path
stringlengths 9
117
| type
stringclasses 2
values | project
stringclasses 10
values | commit_hash
stringlengths 40
40
| commit_message
stringlengths 1
137
| ground_truth
stringlengths 0
2.74k
| main_code
stringlengths 102
3.37k
| context
stringlengths 0
14.7k
|
|---|---|---|---|---|---|---|---|
scripts.prepdocslib.searchmanager/SearchManager.create_index
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
13e4cd8476c42158e708662539ce101b5b6a9655
|
[Backend/Prepdocs] Update to latest version of search SDK (#1010)
|
<16>:<add> vector_search_profile="embedding_config",
<del> vector_search_configuration="default",
|
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def create_index(self):
<0> if self.search_info.verbose:
<1> print(f"Ensuring search index {self.search_info.index_name} exists")
<2>
<3> async with self.search_info.create_search_index_client() as search_index_client:
<4> fields = [
<5> SimpleField(name="id", type="Edm.String", key=True),
<6> SearchableField(name="content", type="Edm.String", analyzer_name=self.search_analyzer_name),
<7> SearchField(
<8> name="embedding",
<9> type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
<10> hidden=False,
<11> searchable=True,
<12> filterable=False,
<13> sortable=False,
<14> facetable=False,
<15> vector_search_dimensions=1536,
<16> vector_search_configuration="default",
<17> ),
<18> SimpleField(name="category", type="Edm.String", filterable=True, facetable=True),
<19> SimpleField(name="sourcepage", type="Edm.String", filterable=True, facetable=True),
<20> SimpleField(name="sourcefile", type="Edm.String", filterable=True, facetable=True),
<21> ]
<22> if self.use_acls:
<23> fields.append(
<24> SimpleField(
<25> name="oids", type=SearchFieldDataType.Collection(SearchFieldDataType.String), filterable=True
<26> )
<27> )
<28> fields.append(
<29> SimpleField(
<30> name="groups", type=SearchFieldDataType.Collection(SearchFieldDataType.String), filterable=True
<31> )
<32> )
<33>
<34> index = SearchIndex(
<35> name=self.search_info.index_name,
<36> fields=fields,
<37> semantic_settings=SemanticSettings(
<38> configurations=[
<39> SemanticConfiguration(
<40> name="default",
<41> prioritized_fields=PrioritizedFields(
<42> title_field</s>
|
===========below chunk 0===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def create_index(self):
# offset: 1
),
)
]
),
vector_search=VectorSearch(
algorithm_configurations=[
VectorSearchAlgorithmConfiguration(
name="default", kind="hnsw", hnsw_parameters=HnswParameters(metric="cosine")
)
]
),
)
if self.search_info.index_name not in [name async for name in search_index_client.list_index_names()]:
if self.search_info.verbose:
print(f"Creating {self.search_info.index_name} search index")
await search_index_client.create_index(index)
else:
if self.search_info.verbose:
print(f"Search index {self.search_info.index_name} already exists")
===========unchanged ref 0===========
at: scripts.prepdocslib.searchmanager.SearchManager.__init__
self.search_info = search_info
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
at: scripts.prepdocslib.strategy.SearchInfo
create_search_index_client() -> SearchIndexClient
at: scripts.prepdocslib.strategy.SearchInfo.__init__
self.index_name = index_name
self.verbose = verbose
|
app.backend.approaches.retrievethenread/RetrieveThenReadApproach.run
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
13e4cd8476c42158e708662539ce101b5b6a9655
|
[Backend/Prepdocs] Update to latest version of search SDK (#1010)
|
<10>:<add> vectors: list[VectorQuery] = []
<14>:<del> else:
<15>:<del> query_vector = None
<16>:<add> vectors.append(RawVectorQuery(vector=query_vector, k=50, fields="embedding"))
|
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
<0> q = messages[-1]["content"]
<1> overrides = context.get("overrides", {})
<2> auth_claims = context.get("auth_claims", {})
<3> has_text = overrides.get("retrieval_mode") in ["text", "hybrid", None]
<4> has_vector = overrides.get("retrieval_mode") in ["vectors", "hybrid", None]
<5> use_semantic_captions = True if overrides.get("semantic_captions") and has_text else False
<6> top = overrides.get("top", 3)
<7> filter = self.build_filter(overrides, auth_claims)
<8>
<9> # If retrieval mode includes vectors, compute an embedding for the query
<10> if has_vector:
<11> embedding_args = {"deployment_id": self.embedding_deployment} if self.openai_host == "azure" else {}
<12> embedding = await openai.Embedding.acreate(**embedding_args, model=self.embedding_model, input=q)
<13> query_vector = embedding["data"][0]["embedding"]
<14> else:
<15> query_vector = None
<16>
<17> # Only keep the text query if the retrieval mode uses text, otherwise drop it
<18> query_text = q if has_text else ""
<19>
<20> # Use semantic ranker if requested and if retrieval mode is text or hybrid (vectors + text)
<21> if overrides.get("semantic_ranker") and has_text:
<22> r = await self.search_client.search(
<23> query_text,
<24> filter=filter,
<25> query_type=QueryType.SEMANTIC,
<26> query_language=self.query_</s>
|
===========below chunk 0===========
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
# offset: 1
query_speller=self.query_speller,
semantic_configuration_name="default",
top=top,
query_caption="extractive|highlight-false" if use_semantic_captions else None,
vector=query_vector,
top_k=50 if query_vector else None,
vector_fields="embedding" if query_vector else None,
)
else:
r = await self.search_client.search(
query_text,
filter=filter,
top=top,
vector=query_vector,
top_k=50 if query_vector else None,
vector_fields="embedding" if query_vector else None,
)
if use_semantic_captions:
results = [
doc[self.sourcepage_field] + ": " + nonewlines(" . ".join([c.text for c in doc["@search.captions"]]))
async for doc in r
]
else:
results = [doc[self.sourcepage_field] + ": " + nonewlines(doc[self.content_field]) async for doc in r]
content = "\n".join(results)
message_builder = MessageBuilder(
overrides.get("prompt_template") or self.system_chat_template, self.chatgpt_model
)
# add user question
user_content = q + "\n" + f"Sources:\n {content}"
message_builder.insert_message("user", user_content)
# Add shots/samples. This helps model to mimic response and make sure they match rules laid out in system message.
</s>
===========below chunk 1===========
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
# offset: 2
<s> # Add shots/samples. This helps model to mimic response and make sure they match rules laid out in system message.
message_builder.insert_message("assistant", self.answer)
message_builder.insert_message("user", self.question)
messages = message_builder.messages
chatgpt_args = {"deployment_id": self.chatgpt_deployment} if self.openai_host == "azure" else {}
chat_completion = await openai.ChatCompletion.acreate(
**chatgpt_args,
model=self.chatgpt_model,
messages=messages,
temperature=overrides.get("temperature") or 0.3,
max_tokens=1024,
n=1,
)
extra_info = {
"data_points": results,
"thoughts": f"Question:<br>{query_text}<br><br>Prompt:<br>"
+ "\n\n".join([str(message) for message in messages]),
}
chat_completion.choices[0]["context"] = extra_info
chat_completion.choices[0]["session_state"] = session_state
return chat_completion
===========unchanged ref 0===========
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach
system_chat_template = (
"You are an intelligent assistant helping Contoso Inc employees with their healthcare plan questions and employee handbook questions. "
+ "Use 'you' to refer to the individual asking the questions even if they ask with 'I'. "
+ "Answer the following question using only the data provided in the sources below. "
+ "For tabular information return it as an html table. Do not return markdown format. "
+ "Each source has a name followed by colon and the actual information, always include the source name for each fact you use in the response. "
+ "If you cannot answer using the sources below, say you don't know. Use below example to answer"
)
question = """
'What is the deductible for the employee plan for a visit to Overlake in Bellevue?'
Sources:
info1.txt: deductibles depend on whether you are in-network or out-of-network. In-network deductibles are $500 for employee and $1000 for family. Out-of-network deductibles are $1000 for employee and $2000 for family.
info2.pdf: Overlake is in-network for the employee plan.
info3.pdf: Overlake is the name of the area that includes a park and ride near Bellevue.
info4.pdf: In-network institutions include Overlake, Swedish and others in the region
"""
answer = "In-network deductibles are $500 for employee and $1000 for family [info1.txt] and Overlake is in-network for the employee plan [info2.pdf][info4.pdf]."
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach.__init__
self.search_client = search_client
self.openai_host = openai_host
self.chatgpt_deployment = chatgpt_deployment
self.chatgpt_model = chatgpt_model
===========unchanged ref 1===========
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
at: approaches.approach.Approach
build_filter(overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]
run(self, messages: list[dict], stream: bool=False, session_state: Any=None, context: dict[str, Any]={}) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]
at: core.messagebuilder
MessageBuilder(system_content: str, chatgpt_model: str)
at: core.messagebuilder.MessageBuilder
insert_message(role: str, content: str, index: int=1)
at: core.messagebuilder.MessageBuilder.__init__
self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
at: openai.api_resources.chat_completion
ChatCompletion(engine: Optional[str]=None, *, id=None, api_key=None, api_version=None, api_type=None, organization=None, response_ms: Optional[int]=None, api_base=None, **params)
at: openai.api_resources.chat_completion.ChatCompletion
engine_required = False
OBJECT_NAME = "chat.completions"
acreate(api_key=None, api_base=None, api_type=None, request_id=None, api_version=None, organization=None, /, *, api_key=None, api_base=None, api_type=None, request_id=None, api_version=None, organization=None, **params)
|
scripts.prepdocslib.embeddings/OpenAIEmbeddings.create_embedding_batch
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<2>:<add> client = await self.create_client()
<4>:<add> retry=retry_if_exception_type(RateLimitError),
<del> retry=retry_if_exception_type(openai.error.RateLimitError),
<10>:<del> emb_args = await self.create_embedding_arguments()
<11>:<add> emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=batch.texts)
<del> emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
<12>:<add> embeddings.extend([data.embedding for data in emb_response.data])
<del> embeddings.extend([data["embedding"] for data in emb_response["data"]])
|
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_batch(self, texts: List[str]) -> List[List[float]]:
<0> batches = self.split_text_into_batches(texts)
<1> embeddings = []
<2> for batch in batches:
<3> async for attempt in AsyncRetrying(
<4> retry=retry_if_exception_type(openai.error.RateLimitError),
<5> wait=wait_random_exponential(min=15, max=60),
<6> stop=stop_after_attempt(15),
<7> before_sleep=self.before_retry_sleep,
<8> ):
<9> with attempt:
<10> emb_args = await self.create_embedding_arguments()
<11> emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
<12> embeddings.extend([data["embedding"] for data in emb_response["data"]])
<13> if self.verbose:
<14> print(f"Batch Completed. Batch size {len(batch.texts)} Token count {batch.token_length}")
<15>
<16> return embeddings
<17>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
|
scripts.prepdocslib.embeddings/OpenAIEmbeddings.create_embedding_single
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> client = await self.create_client()
<1>:<add> retry=retry_if_exception_type(RateLimitError),
<del> retry=retry_if_exception_type(openai.error.RateLimitError),
<7>:<del> emb_args = await self.create_embedding_arguments()
<8>:<add> emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
<del> emb_response = await openai.Embedding.acreate(**emb_args, input=text)
<10>:<add> return emb_response.data[0].embedding
<del> return emb_response["data"][0]["embedding"]
|
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
<0> async for attempt in AsyncRetrying(
<1> retry=retry_if_exception_type(openai.error.RateLimitError),
<2> wait=wait_random_exponential(min=15, max=60),
<3> stop=stop_after_attempt(15),
<4> before_sleep=self.before_retry_sleep,
<5> ):
<6> with attempt:
<7> emb_args = await self.create_embedding_arguments()
<8> emb_response = await openai.Embedding.acreate(**emb_args, input=text)
<9>
<10> return emb_response["data"][0]["embedding"]
<11>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_batch(self, texts: List[str]) -> List[List[float]]:
batches = self.split_text_into_batches(texts)
embeddings = []
+ client = await self.create_client()
for batch in batches:
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=batch.texts)
- emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
+ embeddings.extend([data.embedding for data in emb_response.data])
- embeddings.extend([data["embedding"] for data in emb_response["data"]])
if self.verbose:
print(f"Batch Completed. Batch size {len(batch.texts)} Token count {batch.token_length}")
return embeddings
|
scripts.prepdocslib.embeddings/AzureOpenAIEmbeddingService.wrap_credential
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<9>:<add> raise TypeError("Invalid credential type")
<del> raise Exception("Invalid credential type")
|
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
<0> if isinstance(self.credential, AzureKeyCredential):
<1> return self.credential.key
<2>
<3> if isinstance(self.credential, AsyncTokenCredential):
<4> if not self.cached_token or self.cached_token.expires_on <= time.time():
<5> self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
<6>
<7> return self.cached_token.token
<8>
<9> raise Exception("Invalid credential type")
<10>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
+ client = await self.create_client()
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
- emb_response = await openai.Embedding.acreate(**emb_args, input=text)
+ return emb_response.data[0].embedding
- return emb_response["data"][0]["embedding"]
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_batch(self, texts: List[str]) -> List[List[float]]:
batches = self.split_text_into_batches(texts)
embeddings = []
+ client = await self.create_client()
for batch in batches:
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=batch.texts)
- emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
+ embeddings.extend([data.embedding for data in emb_response.data])
- embeddings.extend([data["embedding"] for data in emb_response["data"]])
if self.verbose:
print(f"Batch Completed. Batch size {len(batch.texts)} Token count {batch.token_length}")
return embeddings
|
app.backend.core.messagebuilder/MessageBuilder.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> self.messages: list[ChatCompletionMessageParam] = [
<add> ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
<del> self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
<1>:<add> ]
|
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
<0> self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
<1> self.model = chatgpt_model
<2>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
+ client = await self.create_client()
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
- emb_response = await openai.Embedding.acreate(**emb_args, input=text)
+ return emb_response.data[0].embedding
- return emb_response["data"][0]["embedding"]
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_batch(self, texts: List[str]) -> List[List[float]]:
batches = self.split_text_into_batches(texts)
embeddings = []
+ client = await self.create_client()
for batch in batches:
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=batch.texts)
- emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
+ embeddings.extend([data.embedding for data in emb_response.data])
- embeddings.extend([data["embedding"] for data in emb_response["data"]])
if self.verbose:
print(f"Batch Completed. Batch size {len(batch.texts)} Token count {batch.token_length}")
return embeddings
|
app.backend.core.messagebuilder/MessageBuilder.insert_message
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<4>:<add> role (str): The role of the message sender (either "user", "system", or "assistant").
<del> role (str): The role of the message sender (either "user" or "system").
<8>:<add> message: ChatCompletionMessageParam
<add> if role == "user":
<add> message = ChatCompletionUserMessageParam(role="user", content=self.normalize_content(content))
<add> elif role == "system":
<add> message = ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(content))
<add> elif role == "assistant":
<add> message = ChatCompletionAssistantMessageParam(role="assistant", content=self.normalize_content(content))
<add> else:
<add> raise ValueError(f"Invalid role: {role}")
<add> self.messages.insert(index, message)
<del> self.messages.insert(index, {"role": role, "content": self.normalize_content(content)})
|
# module: app.backend.core.messagebuilder
class MessageBuilder:
def insert_message(self, role: str, content: str, index: int = 1):
<0> """
<1> Inserts a message into the conversation at the specified index,
<2> or at index 1 (after system message) if no index is specified.
<3> Args:
<4> role (str): The role of the message sender (either "user" or "system").
<5> content (str): The content of the message.
<6> index (int): The index at which to insert the message.
<7> """
<8> self.messages.insert(index, {"role": role, "content": self.normalize_content(content)})
<9>
|
===========unchanged ref 0===========
at: app.backend.core.messagebuilder.MessageBuilder
normalize_content(content: str)
===========changed ref 0===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
+ client = await self.create_client()
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
- emb_response = await openai.Embedding.acreate(**emb_args, input=text)
+ return emb_response.data[0].embedding
- return emb_response["data"][0]["embedding"]
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_batch(self, texts: List[str]) -> List[List[float]]:
batches = self.split_text_into_batches(texts)
embeddings = []
+ client = await self.create_client()
for batch in batches:
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=batch.texts)
- emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
+ embeddings.extend([data.embedding for data in emb_response.data])
- embeddings.extend([data["embedding"] for data in emb_response["data"]])
if self.verbose:
print(f"Batch Completed. Batch size {len(batch.texts)} Token count {batch.token_length}")
return embeddings
|
app.backend.core.modelhelper/num_tokens_from_messages
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<15>:<add> for value in message.values():
<del> for key, value in message.items():
<16>:<add> num_tokens += len(encoding.encode(str(value)))
<del> num_tokens += len(encoding.encode(value))
|
# module: app.backend.core.modelhelper
def num_tokens_from_messages(message: dict[str, str], model: str) -> int:
<0> """
<1> Calculate the number of tokens required to encode a message.
<2> Args:
<3> message (dict): The message to encode, represented as a dictionary.
<4> model (str): The name of the model to use for encoding.
<5> Returns:
<6> int: The total number of tokens required to encode the message.
<7> Example:
<8> message = {'role': 'user', 'content': 'Hello, how are you?'}
<9> model = 'gpt-3.5-turbo'
<10> num_tokens_from_messages(message, model)
<11> output: 11
<12> """
<13> encoding = tiktoken.encoding_for_model(get_oai_chatmodel_tiktok(model))
<14> num_tokens = 2 # For "role" and "content" keys
<15> for key, value in message.items():
<16> num_tokens += len(encoding.encode(value))
<17> return num_tokens
<18>
|
===========unchanged ref 0===========
at: app.backend.core.modelhelper
get_oai_chatmodel_tiktok(aoaimodel: str) -> str
at: tiktoken.core.Encoding
encode(text: str, *, allowed_special: Union[Literal["all"], AbstractSet[str]]=set(), disallowed_special: Union[Literal["all"], Collection[str]]="all") -> list[int]
at: tiktoken.model
encoding_for_model(model_name: str) -> Encoding
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 6===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
+ client = await self.create_client()
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
- emb_response = await openai.Embedding.acreate(**emb_args, input=text)
+ return emb_response.data[0].embedding
- return emb_response["data"][0]["embedding"]
===========changed ref 10===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def insert_message(self, role: str, content: str, index: int = 1):
"""
Inserts a message into the conversation at the specified index,
or at index 1 (after system message) if no index is specified.
Args:
+ role (str): The role of the message sender (either "user", "system", or "assistant").
- role (str): The role of the message sender (either "user" or "system").
content (str): The content of the message.
index (int): The index at which to insert the message.
"""
+ message: ChatCompletionMessageParam
+ if role == "user":
+ message = ChatCompletionUserMessageParam(role="user", content=self.normalize_content(content))
+ elif role == "system":
+ message = ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(content))
+ elif role == "assistant":
+ message = ChatCompletionAssistantMessageParam(role="assistant", content=self.normalize_content(content))
+ else:
+ raise ValueError(f"Invalid role: {role}")
+ self.messages.insert(index, message)
- self.messages.insert(index, {"role": role, "content": self.normalize_content(content)})
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_batch(self, texts: List[str]) -> List[List[float]]:
batches = self.split_text_into_batches(texts)
embeddings = []
+ client = await self.create_client()
for batch in batches:
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=batch.texts)
- emb_response = await openai.Embedding.acreate(**emb_args, input=batch.texts)
+ embeddings.extend([data.embedding for data in emb_response.data])
- embeddings.extend([data["embedding"] for data in emb_response["data"]])
if self.verbose:
print(f"Batch Completed. Batch size {len(batch.texts)} Token count {batch.token_length}")
return embeddings
|
tests.test_prepdocs/test_compute_embedding_success
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> async def mock_create_client(*args, **kwargs):
<del> async def mock_create(*args, **kwargs):
<2>:<add> return MockClient(
<add> embeddings_client=MockEmbeddingsClient(
<add> create_embedding_response=openai.types.CreateEmbeddingResponse(
<del> return {
<3>:<add> object="list",
<del> "object": "list",
<4>:<del> "data": [
<5>:<del> {
<6>:<del> "object": "embedding",
<7>:<add> data=[
<add> openai.types.Embedding(
<add> embedding=[
<del> "embedding": [
<8>:<add> 0.0023064255,
<del> 0.0023064255,
<9>:<add> -0.009327292,
<del> -0.009327292,
<10>:<add> -0.0028842222,
<del> -0.0028842222,
<11>:<add> ],
<add> index=0,
<add> object="embedding",
<add> )
<12>:<add> model="text-embedding-ada-002",
<add> usage=Usage(prompt_tokens=8, total_tokens=8),
<del> "index": 0,
<13>:<add> )
<del> }
<14>:<add> )
<del> ],
<15>:<del> "model": "text-embedding-ada-002",
<16>:<del> "usage": {"prompt_tokens": 8, "total_tokens": 8},
<17>:<add> )
<del> }
<19>:<del> monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
<27>:<add> monkeypatch.setattr(embeddings, "create_client", mock_create_client)
|
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
<0> async def mock_create(*args, **kwargs):
<1> # From https://platform.openai.com/docs/api-reference/embeddings/create
<2> return {
<3> "object": "list",
<4> "data": [
<5> {
<6> "object": "embedding",
<7> "embedding": [
<8> 0.0023064255,
<9> -0.009327292,
<10> -0.0028842222,
<11> ],
<12> "index": 0,
<13> }
<14> ],
<15> "model": "text-embedding-ada-002",
<16> "usage": {"prompt_tokens": 8, "total_tokens": 8},
<17> }
<18>
<19> monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
<20> embeddings = AzureOpenAIEmbeddingService(
<21> open_ai_service="x",
<22> open_ai_deployment="x",
<23> open_ai_model_name="text-ada-003",
<24> credential=MockAzureCredential(),
<25> disable_batch=False,
<26> )
<27> assert await embeddings.create_embeddings(texts=["foo"]) == [
<28> [
<29> 0.0023064255,
<30> -0.009327292,
<31> -0.0028842222,
<32> ]
<33> ]
<34>
<35> embeddings = AzureOpenAIEmbeddingService(
<36> open_ai_service="x",
<37> open_ai_deployment="x",
<38> open_ai_model_name="text-ada-003",
<39> credential=MockAzureCredential(),
<40> disable_batch=True,
<41> )
<42> assert await embeddings.create_embeddings(texts=["foo"]) == [
<43> [
<44> 0.0023064255,
<45> -0.009327292,
<46> -0.002884</s>
|
===========below chunk 0===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
# offset: 1
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=False
)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=True
)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 6===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
+ client = await self.create_client()
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
- emb_response = await openai.Embedding.acreate(**emb_args, input=text)
+ return emb_response.data[0].embedding
- return emb_response["data"][0]["embedding"]
===========changed ref 10===========
# module: app.backend.core.modelhelper
def num_tokens_from_messages(message: dict[str, str], model: str) -> int:
"""
Calculate the number of tokens required to encode a message.
Args:
message (dict): The message to encode, represented as a dictionary.
model (str): The name of the model to use for encoding.
Returns:
int: The total number of tokens required to encode the message.
Example:
message = {'role': 'user', 'content': 'Hello, how are you?'}
model = 'gpt-3.5-turbo'
num_tokens_from_messages(message, model)
output: 11
"""
encoding = tiktoken.encoding_for_model(get_oai_chatmodel_tiktok(model))
num_tokens = 2 # For "role" and "content" keys
+ for value in message.values():
- for key, value in message.items():
+ num_tokens += len(encoding.encode(str(value)))
- num_tokens += len(encoding.encode(value))
return num_tokens
|
tests.test_prepdocs/test_compute_embedding_ratelimiterror_batch
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> async def mock_acreate(*args, **kwargs):
<1>:<del> raise openai.error.RateLimitError
<2>:<del>
<3>:<del> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<14>:<add> monkeypatch.setattr(embeddings, "create_client", create_rate_limit_client)
|
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_ratelimiterror_batch(monkeypatch, capsys):
<0> async def mock_acreate(*args, **kwargs):
<1> raise openai.error.RateLimitError
<2>
<3> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<4> monkeypatch.setattr(tenacity.wait_random_exponential, "__call__", lambda x, y: 0)
<5> with pytest.raises(tenacity.RetryError):
<6> embeddings = AzureOpenAIEmbeddingService(
<7> open_ai_service="x",
<8> open_ai_deployment="x",
<9> open_ai_model_name="text-embedding-ada-002",
<10> credential=MockAzureCredential(),
<11> disable_batch=False,
<12> verbose=True,
<13> )
<14> await embeddings.create_embeddings(texts=["foo"])
<15> captured = capsys.readouterr()
<16> assert captured.out.count("Rate limited on the OpenAI embeddings API") == 14
<17>
|
===========changed ref 0===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 1===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 2===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 3===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
+ async def mock_create_client(*args, **kwargs):
- async def mock_create(*args, **kwargs):
# From https://platform.openai.com/docs/api-reference/embeddings/create
+ return MockClient(
+ embeddings_client=MockEmbeddingsClient(
+ create_embedding_response=openai.types.CreateEmbeddingResponse(
- return {
+ object="list",
- "object": "list",
- "data": [
- {
- "object": "embedding",
+ data=[
+ openai.types.Embedding(
+ embedding=[
- "embedding": [
+ 0.0023064255,
- 0.0023064255,
+ -0.009327292,
- -0.009327292,
+ -0.0028842222,
- -0.0028842222,
+ ],
+ index=0,
+ object="embedding",
+ )
],
+ model="text-embedding-ada-002",
+ usage=Usage(prompt_tokens=8, total_tokens=8),
- "index": 0,
+ )
- }
+ )
- ],
- "model": "text-embedding-ada-002",
- "usage": {"prompt_tokens": 8, "total_tokens": 8},
+ )
- }
- monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-ada-003",
credential=MockAzureCredential(),
disable_batch=False,
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert</s>
===========changed ref 4===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
# offset: 1
<s>
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-ada-003",
credential=MockAzureCredential(),
disable_batch=True,
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=False
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=True
)
+ monkeypatch.setattr(embeddings, "create_client",</s>
===========changed ref 5===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
# offset: 2
<s>create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 12===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
|
tests.test_prepdocs/test_compute_embedding_ratelimiterror_single
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> async def mock_acreate(*args, **kwargs):
<1>:<del> raise openai.error.RateLimitError
<2>:<del>
<3>:<del> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<14>:<add> monkeypatch.setattr(embeddings, "create_client", create_rate_limit_client)
|
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_ratelimiterror_single(monkeypatch, capsys):
<0> async def mock_acreate(*args, **kwargs):
<1> raise openai.error.RateLimitError
<2>
<3> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<4> monkeypatch.setattr(tenacity.wait_random_exponential, "__call__", lambda x, y: 0)
<5> with pytest.raises(tenacity.RetryError):
<6> embeddings = AzureOpenAIEmbeddingService(
<7> open_ai_service="x",
<8> open_ai_deployment="x",
<9> open_ai_model_name="text-embedding-ada-002",
<10> credential=MockAzureCredential(),
<11> disable_batch=True,
<12> verbose=True,
<13> )
<14> await embeddings.create_embeddings(texts=["foo"])
<15> captured = capsys.readouterr()
<16> assert captured.out.count("Rate limited on the OpenAI embeddings API") == 14
<17>
|
===========changed ref 0===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 1===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 2===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 3===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_ratelimiterror_batch(monkeypatch, capsys):
- async def mock_acreate(*args, **kwargs):
- raise openai.error.RateLimitError
-
- monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
monkeypatch.setattr(tenacity.wait_random_exponential, "__call__", lambda x, y: 0)
with pytest.raises(tenacity.RetryError):
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-embedding-ada-002",
credential=MockAzureCredential(),
disable_batch=False,
verbose=True,
)
+ monkeypatch.setattr(embeddings, "create_client", create_rate_limit_client)
await embeddings.create_embeddings(texts=["foo"])
captured = capsys.readouterr()
assert captured.out.count("Rate limited on the OpenAI embeddings API") == 14
===========changed ref 4===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
+ async def mock_create_client(*args, **kwargs):
- async def mock_create(*args, **kwargs):
# From https://platform.openai.com/docs/api-reference/embeddings/create
+ return MockClient(
+ embeddings_client=MockEmbeddingsClient(
+ create_embedding_response=openai.types.CreateEmbeddingResponse(
- return {
+ object="list",
- "object": "list",
- "data": [
- {
- "object": "embedding",
+ data=[
+ openai.types.Embedding(
+ embedding=[
- "embedding": [
+ 0.0023064255,
- 0.0023064255,
+ -0.009327292,
- -0.009327292,
+ -0.0028842222,
- -0.0028842222,
+ ],
+ index=0,
+ object="embedding",
+ )
],
+ model="text-embedding-ada-002",
+ usage=Usage(prompt_tokens=8, total_tokens=8),
- "index": 0,
+ )
- }
+ )
- ],
- "model": "text-embedding-ada-002",
- "usage": {"prompt_tokens": 8, "total_tokens": 8},
+ )
- }
- monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-ada-003",
credential=MockAzureCredential(),
disable_batch=False,
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert</s>
===========changed ref 5===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
# offset: 1
<s>
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-ada-003",
credential=MockAzureCredential(),
disable_batch=True,
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=False
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=True
)
+ monkeypatch.setattr(embeddings, "create_client",</s>
===========changed ref 6===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
# offset: 2
<s>create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 12===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
|
tests.test_prepdocs/test_compute_embedding_autherror
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> async def mock_acreate(*args, **kwargs):
<1>:<del> raise openai.error.AuthenticationError
<2>:<del>
<3>:<del> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<5>:<add> with pytest.raises(openai.AuthenticationError):
<del> with pytest.raises(openai.error.AuthenticationError):
<14>:<add> monkeypatch.setattr(embeddings, "create_client", create_auth_error_limit_client)
<16>:<add> with pytest.raises(openai.AuthenticationError):
<del> with pytest.raises(openai.error.AuthenticationError):
<25>:<add> monkeypatch.setattr(embeddings, "create_client", create_auth_error_limit_client)
|
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_autherror(monkeypatch, capsys):
<0> async def mock_acreate(*args, **kwargs):
<1> raise openai.error.AuthenticationError
<2>
<3> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<4> monkeypatch.setattr(tenacity.wait_random_exponential, "__call__", lambda x, y: 0)
<5> with pytest.raises(openai.error.AuthenticationError):
<6> embeddings = AzureOpenAIEmbeddingService(
<7> open_ai_service="x",
<8> open_ai_deployment="x",
<9> open_ai_model_name="text-embedding-ada-002",
<10> credential=MockAzureCredential(),
<11> disable_batch=False,
<12> verbose=True,
<13> )
<14> await embeddings.create_embeddings(texts=["foo"])
<15>
<16> with pytest.raises(openai.error.AuthenticationError):
<17> embeddings = AzureOpenAIEmbeddingService(
<18> open_ai_service="x",
<19> open_ai_deployment="x",
<20> open_ai_model_name="text-embedding-ada-002",
<21> credential=MockAzureCredential(),
<22> disable_batch=True,
<23> verbose=True,
<24> )
<25> await embeddings.create_embeddings(texts=["foo"])
<26>
|
===========changed ref 0===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 1===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 2===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_ratelimiterror_single(monkeypatch, capsys):
- async def mock_acreate(*args, **kwargs):
- raise openai.error.RateLimitError
-
- monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
monkeypatch.setattr(tenacity.wait_random_exponential, "__call__", lambda x, y: 0)
with pytest.raises(tenacity.RetryError):
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-embedding-ada-002",
credential=MockAzureCredential(),
disable_batch=True,
verbose=True,
)
+ monkeypatch.setattr(embeddings, "create_client", create_rate_limit_client)
await embeddings.create_embeddings(texts=["foo"])
captured = capsys.readouterr()
assert captured.out.count("Rate limited on the OpenAI embeddings API") == 14
===========changed ref 7===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_ratelimiterror_batch(monkeypatch, capsys):
- async def mock_acreate(*args, **kwargs):
- raise openai.error.RateLimitError
-
- monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
monkeypatch.setattr(tenacity.wait_random_exponential, "__call__", lambda x, y: 0)
with pytest.raises(tenacity.RetryError):
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-embedding-ada-002",
credential=MockAzureCredential(),
disable_batch=False,
verbose=True,
)
+ monkeypatch.setattr(embeddings, "create_client", create_rate_limit_client)
await embeddings.create_embeddings(texts=["foo"])
captured = capsys.readouterr()
assert captured.out.count("Rate limited on the OpenAI embeddings API") == 14
===========changed ref 8===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
+ async def mock_create_client(*args, **kwargs):
- async def mock_create(*args, **kwargs):
# From https://platform.openai.com/docs/api-reference/embeddings/create
+ return MockClient(
+ embeddings_client=MockEmbeddingsClient(
+ create_embedding_response=openai.types.CreateEmbeddingResponse(
- return {
+ object="list",
- "object": "list",
- "data": [
- {
- "object": "embedding",
+ data=[
+ openai.types.Embedding(
+ embedding=[
- "embedding": [
+ 0.0023064255,
- 0.0023064255,
+ -0.009327292,
- -0.009327292,
+ -0.0028842222,
- -0.0028842222,
+ ],
+ index=0,
+ object="embedding",
+ )
],
+ model="text-embedding-ada-002",
+ usage=Usage(prompt_tokens=8, total_tokens=8),
- "index": 0,
+ )
- }
+ )
- ],
- "model": "text-embedding-ada-002",
- "usage": {"prompt_tokens": 8, "total_tokens": 8},
+ )
- }
- monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-ada-003",
credential=MockAzureCredential(),
disable_batch=False,
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert</s>
===========changed ref 9===========
# module: tests.test_prepdocs
@pytest.mark.asyncio
async def test_compute_embedding_success(monkeypatch):
# offset: 1
<s>
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = AzureOpenAIEmbeddingService(
open_ai_service="x",
open_ai_deployment="x",
open_ai_model_name="text-ada-003",
credential=MockAzureCredential(),
disable_batch=True,
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=False
)
+ monkeypatch.setattr(embeddings, "create_client", mock_create_client)
assert await embeddings.create_embeddings(texts=["foo"]) == [
[
0.0023064255,
-0.009327292,
-0.0028842222,
]
]
embeddings = OpenAIEmbeddingService(
open_ai_model_name="text-ada-003", credential=MockAzureCredential(), organization="org", disable_batch=True
)
+ monkeypatch.setattr(embeddings, "create_client",</s>
|
tests.test_chatapproach/test_get_search_query
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
<4>:<add> payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"this is the query","role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
<del> payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
|
# module: tests.test_chatapproach
+ def test_get_search_query(chat_approach):
- def test_get_search_query():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
<5> default_query = "hello"
<6> query = chat_approach.get_search_query(json.loads(payload), default_query)
<7>
<8> assert query == "accesstelemedicineservices"
<9>
|
===========unchanged ref 0===========
at: _pytest.fixtures
fixture(fixture_function: FixtureFunction, *, scope: "Union[_ScopeName, Callable[[str, Config], _ScopeName]]"=..., params: Optional[Iterable[object]]=..., autouse: bool=..., ids: Optional[
Union[Sequence[Optional[object]], Callable[[Any], Optional[object]]]
]=..., name: Optional[str]=...) -> FixtureFunction
fixture(fixture_function: None=..., *, scope: "Union[_ScopeName, Callable[[str, Config], _ScopeName]]"=..., params: Optional[Iterable[object]]=..., autouse: bool=..., ids: Optional[
Union[Sequence[Optional[object]], Callable[[Any], Optional[object]]]
]=..., name: Optional[str]=None) -> FixtureFunctionMarker
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 9===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 12===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 14===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
def create_embedding_single(self, text: str) -> List[float]:
+ client = await self.create_client()
async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(RateLimitError),
- retry=retry_if_exception_type(openai.error.RateLimitError),
wait=wait_random_exponential(min=15, max=60),
stop=stop_after_attempt(15),
before_sleep=self.before_retry_sleep,
):
with attempt:
- emb_args = await self.create_embedding_arguments()
+ emb_response = await client.embeddings.create(model=self.open_ai_model_name, input=text)
- emb_response = await openai.Embedding.acreate(**emb_args, input=text)
+ return emb_response.data[0].embedding
- return emb_response["data"][0]["embedding"]
|
tests.test_chatapproach/test_get_search_query_returns_default
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
<4>:<add> payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"","role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
<del> payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
<6>:<add> chatcompletions = ChatCompletion.model_validate(json.loads(payload), strict=False)
<add> query = chat_approach.get_search_query(chat
|
# module: tests.test_chatapproach
+ def test_get_search_query_returns_default(chat_approach):
- def test_get_search_query_returns_default():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
<5> default_query = "hello"
<6> query = chat_approach.get_search_query(json.loads(payload), default_query)
<7>
<8> assert query == default_query
<9>
|
===========unchanged ref 0===========
at: tests.test_chatapproach
chat_approach()
===========changed ref 0===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_get_search_query(chat_approach):
- def test_get_search_query():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
+ payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"this is the query","role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
- payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant","function_call":</s>
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_search_query(chat_approach):
- def test_get_search_query():
# offset: 1
<s>],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
default_query = "hello"
+ chatcompletions = ChatCompletion.model_validate(json.loads(payload), strict=False)
+ query = chat_approach.get_search_query(chatcompletions, default_query)
- query = chat_approach.get_search_query(json.loads(payload), default_query)
assert query == "accesstelemedicineservices"
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 12===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 17===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
|
tests.test_chatapproach/test_get_messages_from_history
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> messages = chat_approach.get_messages_from_history(
<5> system_prompt="You are a bot.",
<6> model_id="gpt-35-turbo",
<7> history=[
<8> {"role": "user", "content": "What happens in a performance review?"},
<9> {
<10> "role": "assistant",
<11> "content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
<12> },
<13> {"role": "user", "content": "What does a Product Manager do?"},
<14> ],
<15> user_content="What does a Product Manager do?",
<16> max_tokens=3000,
<17> )
<18> assert messages == [
<19> {"role": "system", "content": "You are a bot."},
<20> {"role": "user", "content": "What happens in a performance review?"},
<21> {
<22> "role": "assistant",
<23> "content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The</s>
|
===========below chunk 0===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
# offset: 1
},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========unchanged ref 0===========
at: json
loads(s: Union[str, bytes], *, cls: Optional[Type[JSONDecoder]]=..., object_hook: Optional[Callable[[Dict[Any, Any]], Any]]=..., parse_float: Optional[Callable[[str], Any]]=..., parse_int: Optional[Callable[[str], Any]]=..., parse_constant: Optional[Callable[[str], Any]]=..., object_pairs_hook: Optional[Callable[[List[Tuple[Any, Any]]], Any]]=..., **kwds: Any) -> Any
at: tests.test_chatapproach.test_get_search_query
query = chat_approach.get_search_query(chatcompletions, default_query)
===========changed ref 0===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_get_search_query_returns_default(chat_approach):
- def test_get_search_query_returns_default():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
+ payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"","role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
- payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
</s>
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_search_query_returns_default(chat_approach):
- def test_get_search_query_returns_default():
# offset: 1
<s>results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
default_query = "hello"
+ chatcompletions = ChatCompletion.model_validate(json.loads(payload), strict=False)
+ query = chat_approach.get_search_query(chatcompletions, default_query)
- query = chat_approach.get_search_query(json.loads(payload), default_query)
assert query == default_query
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_search_query(chat_approach):
- def test_get_search_query():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
+ payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"this is the query","role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
- payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant","function_call":</s>
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_search_query(chat_approach):
- def test_get_search_query():
# offset: 1
<s>],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant","function_call":{"name":"search_sources","arguments":"{\\n\\"search_query\\":\\"accesstelemedicineservices\\"\\n}"}},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
default_query = "hello"
+ chatcompletions = ChatCompletion.model_validate(json.loads(payload), strict=False)
+ query = chat_approach.get_search_query(chatcompletions, default_query)
- query = chat_approach.get_search_query(json.loads(payload), default_query)
assert query == "accesstelemedicineservices"
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
|
tests.test_chatapproach/test_get_messages_from_history_truncated
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> messages = chat_approach.get_messages_from_history(
<5> system_prompt="You are a bot.",
<6> model_id="gpt-35-turbo",
<7> history=[
<8> {"role": "user", "content": "What happens in a performance review?"},
<9> {
<10> "role": "assistant",
<11> "content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
<12> },
<13> {"role": "user", "content": "What does a Product Manager do?"},
<14> ],
<15> user_content="What does a Product Manager do?",
<16> max_tokens=10,
<17> )
<18> assert messages == [
<19> {"role": "system", "content": "You are a bot."},
<20> {"role": "user", "content": "What does a Product Manager do?"},
<21> ]
<22>
|
===========changed ref 0===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=3000,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a</s>
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
# offset: 1
<s> and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_search_query_returns_default(chat_approach):
- def test_get_search_query_returns_default():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
+ payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"","role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
- payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
</s>
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_search_query_returns_default(chat_approach):
- def test_get_search_query_returns_default():
# offset: 1
<s>results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
default_query = "hello"
+ chatcompletions = ChatCompletion.model_validate(json.loads(payload), strict=False)
+ query = chat_approach.get_search_query(chatcompletions, default_query)
- query = chat_approach.get_search_query(json.loads(payload), default_query)
assert query == default_query
|
tests.test_chatapproach/test_get_messages_from_history_truncated_longer
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> messages = chat_approach.get_messages_from_history(
<5> system_prompt="You are a bot.", # 8 tokens
<6> model_id="gpt-35-turbo",
<7> history=[
<8> {"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
<9> {
<10> "role": "assistant",
<11> "content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
<12> }, # 102 tokens
<13> {"role": "user", "content": "Is there a dress code?"}, # 9 tokens
<14> {
<15> "role": "assistant",
<16> "content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
<17> }, # 26 tokens
<18> {"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
<19> ],
<20> user_content="What does a Product Manager do?",
<21> max_tokens=55,
<22> )
<23> assert messages == [
<24> {"role": "system", "</s>
|
===========below chunk 0===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
# offset: 1
{"role": "user", "content": "Is there a dress code?"},
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 0===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=3000,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a</s>
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
# offset: 1
<s> and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_search_query_returns_default(chat_approach):
- def test_get_search_query_returns_default():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
+ payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"content":"","role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
- payload = '{"id":"chatcmpl-81JkxYqYppUkPtOAia40gki2vJ9QM","object":"chat.completion","created":1695324963,"model":"gpt-35-turbo","prompt_filter_results":[{"prompt_index":0,"content_filter_results":{"hate":{"filtered":false,"severity":"safe"},"self_harm":{"filtered":false,"severity":"safe"},"sexual":{"filtered":false,"severity":"safe"},"violence":{"filtered":false,"severity":"safe"}}}],"choices":[{"index":0,"finish_reason":"function_call","message":{"role":"assistant"},"content_filter_results":{}}],"usage":{"completion_tokens":19,"prompt_tokens":425,"total_tokens":444}}'
</s>
|
tests.test_chatapproach/test_get_messages_from_history_truncated_break_pair
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<1>:<del> chat_approach = ChatReadRetrieveReadApproach(
<2>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<3>:<del> )
<4>:<del>
|
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_break_pair(chat_approach):
- def test_get_messages_from_history_truncated_break_pair():
<0> """Tests that the truncation breaks the pair of messages."""
<1> chat_approach = ChatReadRetrieveReadApproach(
<2> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<3> )
<4>
<5> messages = chat_approach.get_messages_from_history(
<6> system_prompt="You are a bot.", # 8 tokens
<7> model_id="gpt-35-turbo",
<8> history=[
<9> {"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
<10> {
<11> "role": "assistant",
<12> "content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
<13> }, # 102 tokens
<14> {"role": "user", "content": "Is there a dress code?"}, # 9 tokens
<15> {
<16> "role": "assistant",
<17> "content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
<18> }, # 26 tokens
<19> {"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
<20> ],
<21> user_content="What does a Product Manager do?",
<22> max_tokens=147,
<23> )
</s>
|
===========below chunk 0===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_break_pair(chat_approach):
- def test_get_messages_from_history_truncated_break_pair():
# offset: 1
{"role": "system", "content": "You are a bot."},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "Is there a dress code?"},
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 0===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.", # 8 tokens
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
}, # 102 tokens
{"role": "user", "content": "Is there a dress code?"}, # 9 tokens
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
}, # 26 tokens
{"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
],
user_content="What does a Product Manager do?",
max_tokens=55,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "</s>
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
# offset: 1
<s>
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "Is there a dress code?"},
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
|
tests.test_chatapproach/test_extract_followup_questions
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_extract_followup_questions(chat_approach):
- def test_extract_followup_questions():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> content = "Here is answer to your question.<<What is the dress code?>>"
<5> pre_content, followup_questions = chat_approach.extract_followup_questions(content)
<6> assert pre_content == "Here is answer to your question."
<7> assert followup_questions == ["What is the dress code?"]
<8>
|
===========unchanged ref 0===========
at: tests.test_chatapproach.test_extract_followup_questions
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
===========changed ref 0===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.", # 8 tokens
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
}, # 102 tokens
{"role": "user", "content": "Is there a dress code?"}, # 9 tokens
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
}, # 26 tokens
{"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
],
user_content="What does a Product Manager do?",
max_tokens=55,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "</s>
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
# offset: 1
<s>
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "Is there a dress code?"},
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=3000,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a</s>
===========changed ref 5===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history(chat_approach):
- def test_get_messages_from_history():
# offset: 1
<s> and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
|
tests.test_chatapproach/test_extract_followup_questions_three
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_extract_followup_questions_three(chat_approach):
- def test_extract_followup_questions_three():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> content = """Here is answer to your question.
<5>
<6> <<What are some examples of successful product launches they should have experience with?>>
<7> <<Are there any specific technical skills or certifications required for the role?>>
<8> <<Is there a preference for candidates with experience in a specific industry or sector?>>"""
<9> pre_content, followup_questions = chat_approach.extract_followup_questions(content)
<10> assert pre_content == "Here is answer to your question.\n\n"
<11> assert followup_questions == [
<12> "What are some examples of successful product launches they should have experience with?",
<13> "Are there any specific technical skills or certifications required for the role?",
<14> "Is there a preference for candidates with experience in a specific industry or sector?",
<15> ]
<16>
|
===========unchanged ref 0===========
at: tests.test_chatapproach.test_extract_followup_questions_three
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
===========changed ref 0===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions(chat_approach):
- def test_extract_followup_questions():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question.<<What is the dress code?>>"
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == ["What is the dress code?"]
===========changed ref 1===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.", # 8 tokens
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
}, # 102 tokens
{"role": "user", "content": "Is there a dress code?"}, # 9 tokens
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
}, # 26 tokens
{"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
],
user_content="What does a Product Manager do?",
max_tokens=55,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "</s>
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
# offset: 1
<s>
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "Is there a dress code?"},
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
|
tests.test_chatapproach/test_extract_followup_questions_no_followup
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> content = "Here is answer to your question."
<5> pre_content, followup_questions = chat_approach.extract_followup_questions(content)
<6> assert pre_content == "Here is answer to your question."
<7> assert followup_questions == []
<8>
|
===========unchanged ref 0===========
at: tests.test_chatapproach.test_extract_followup_questions_no_pre_content
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
===========changed ref 0===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions(chat_approach):
- def test_extract_followup_questions():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question.<<What is the dress code?>>"
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == ["What is the dress code?"]
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_three(chat_approach):
- def test_extract_followup_questions_three():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = """Here is answer to your question.
<<What are some examples of successful product launches they should have experience with?>>
<<Are there any specific technical skills or certifications required for the role?>>
<<Is there a preference for candidates with experience in a specific industry or sector?>>"""
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question.\n\n"
assert followup_questions == [
"What are some examples of successful product launches they should have experience with?",
"Are there any specific technical skills or certifications required for the role?",
"Is there a preference for candidates with experience in a specific industry or sector?",
]
===========changed ref 2===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.", # 8 tokens
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
}, # 102 tokens
{"role": "user", "content": "Is there a dress code?"}, # 9 tokens
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
}, # 26 tokens
{"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
],
user_content="What does a Product Manager do?",
max_tokens=55,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "</s>
===========changed ref 5===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
# offset: 1
<s>
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "Is there a dress code?"},
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
},
{"role": "user", "content": "What does a Product Manager do?"},
]
|
tests.test_chatapproach/test_extract_followup_questions_no_pre_content
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_pre_content(chat_approach):
- def test_extract_followup_questions_no_pre_content():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> content = "<<What is the dress code?>>"
<5> pre_content, followup_questions = chat_approach.extract_followup_questions(content)
<6> assert pre_content == ""
<7> assert followup_questions == ["What is the dress code?"]
<8>
|
===========unchanged ref 0===========
at: tests.test_chatapproach.test_get_messages_from_history_few_shots
messages = chat_approach.get_messages_from_history(
system_prompt=chat_approach.query_prompt_template,
model_id=chat_approach.chatgpt_model,
user_content=user_query_request,
history=[],
max_tokens=chat_approach.chatgpt_token_limit - len(user_query_request),
few_shots=chat_approach.query_prompt_few_shots,
)
===========changed ref 0===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions(chat_approach):
- def test_extract_followup_questions():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question.<<What is the dress code?>>"
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == ["What is the dress code?"]
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_three(chat_approach):
- def test_extract_followup_questions_three():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = """Here is answer to your question.
<<What are some examples of successful product launches they should have experience with?>>
<<Are there any specific technical skills or certifications required for the role?>>
<<Is there a preference for candidates with experience in a specific industry or sector?>>"""
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question.\n\n"
assert followup_questions == [
"What are some examples of successful product launches they should have experience with?",
"Are there any specific technical skills or certifications required for the role?",
"Is there a preference for candidates with experience in a specific industry or sector?",
]
===========changed ref 3===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 4===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 5===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.", # 8 tokens
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
}, # 102 tokens
{"role": "user", "content": "Is there a dress code?"}, # 9 tokens
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
}, # 26 tokens
{"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
],
user_content="What does a Product Manager do?",
max_tokens=55,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "</s>
|
tests.test_chatapproach/test_get_messages_from_history_few_shots
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> chat_approach = ChatReadRetrieveReadApproach(
<1>:<del> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2>:<del> )
<3>:<del>
|
# module: tests.test_chatapproach
+ def test_get_messages_from_history_few_shots(chat_approach):
- def test_get_messages_from_history_few_shots():
<0> chat_approach = ChatReadRetrieveReadApproach(
<1> None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
<2> )
<3>
<4> user_query_request = "What does a Product manager do?"
<5> messages = chat_approach.get_messages_from_history(
<6> system_prompt=chat_approach.query_prompt_template,
<7> model_id=chat_approach.chatgpt_model,
<8> user_content=user_query_request,
<9> history=[],
<10> max_tokens=chat_approach.chatgpt_token_limit - len(user_query_request),
<11> few_shots=chat_approach.query_prompt_few_shots,
<12> )
<13> # Make sure messages are in the right order
<14> assert messages[0]["role"] == "system"
<15> assert messages[1]["role"] == "user"
<16> assert messages[2]["role"] == "assistant"
<17> assert messages[3]["role"] == "user"
<18> assert messages[4]["role"] == "assistant"
<19> assert messages[5]["role"] == "user"
<20> assert messages[5]["content"] == user_query_request
<21>
|
===========changed ref 0===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_pre_content(chat_approach):
- def test_extract_followup_questions_no_pre_content():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "<<What is the dress code?>>"
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == ""
assert followup_questions == ["What is the dress code?"]
===========changed ref 1===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
===========changed ref 2===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions(chat_approach):
- def test_extract_followup_questions():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question.<<What is the dress code?>>"
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == ["What is the dress code?"]
===========changed ref 3===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_three(chat_approach):
- def test_extract_followup_questions_three():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = """Here is answer to your question.
<<What are some examples of successful product launches they should have experience with?>>
<<Are there any specific technical skills or certifications required for the role?>>
<<Is there a preference for candidates with experience in a specific industry or sector?>>"""
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question.\n\n"
assert followup_questions == [
"What are some examples of successful product launches they should have experience with?",
"Are there any specific technical skills or certifications required for the role?",
"Is there a preference for candidates with experience in a specific industry or sector?",
]
===========changed ref 4===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 5===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated(chat_approach):
- def test_get_messages_from_history_truncated():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.",
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"},
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
},
{"role": "user", "content": "What does a Product Manager do?"},
],
user_content="What does a Product Manager do?",
max_tokens=10,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "What does a Product Manager do?"},
]
===========changed ref 6===========
# module: tests.test_chatapproach
+ def test_get_messages_from_history_truncated_longer(chat_approach):
- def test_get_messages_from_history_truncated_longer():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
messages = chat_approach.get_messages_from_history(
system_prompt="You are a bot.", # 8 tokens
model_id="gpt-35-turbo",
history=[
{"role": "user", "content": "What happens in a performance review?"}, # 10 tokens
{
"role": "assistant",
"content": "During the performance review at Contoso Electronics, the supervisor will discuss the employee's performance over the past year and provide feedback on areas for improvement. They will also provide an opportunity for the employee to discuss their goals and objectives for the upcoming year. The review is a two-way dialogue between managers and employees, and employees will receive a written summary of their performance review which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
}, # 102 tokens
{"role": "user", "content": "Is there a dress code?"}, # 9 tokens
{
"role": "assistant",
"content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]",
}, # 26 tokens
{"role": "user", "content": "What does a Product Manager do?"}, # 10 tokens
],
user_content="What does a Product Manager do?",
max_tokens=55,
)
assert messages == [
{"role": "system", "content": "You are a bot."},
{"role": "user", "content": "</s>
|
tests.conftest/mock_openai_embedding
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<1>:<add> return CreateEmbeddingResponse(
<add> object="list",
<add> data=[
<add> Embedding(
<add> embedding=[
<add> 0.0023064255,
<add> -0.009327292,
<add> -0.0028842222,
<add> ],
<add> index=0,
<add> object="embedding",
<add> )
<add> ],
<add> model="text-embedding-ada-002",
<add> usage=Usage(prompt_tokens=8, total_tokens=8),
<add> )
<del> if openai.api_type == "openai":
<2>:<del> assert kwargs.get("deployment_id") is None
<3>:<del> else:
<4>:<del> assert kwargs.get("deployment_id") is not None
<5>:<del> return {"data": [{"embedding": [0.1, 0.2, 0.3]}]}
<7>:<add> def patch(openai_client):
<add> monkeypatch.setattr(openai_client.embeddings, "create", mock_acreate)
<del> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
|
# module: tests.conftest
@pytest.fixture
def mock_openai_embedding(monkeypatch):
<0> async def mock_acreate(*args, **kwargs):
<1> if openai.api_type == "openai":
<2> assert kwargs.get("deployment_id") is None
<3> else:
<4> assert kwargs.get("deployment_id") is not None
<5> return {"data": [{"embedding": [0.1, 0.2, 0.3]}]}
<6>
<7> monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
<8>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 9===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 12===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 14===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 15===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 17===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
===========changed ref 19===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_pre_content(chat_approach):
- def test_extract_followup_questions_no_pre_content():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "<<What is the dress code?>>"
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == ""
assert followup_questions == ["What is the dress code?"]
|
tests.conftest/mock_openai_chatcompletion
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<2>:<add> chunk_id = "test-id"
<add> model = "gpt-35-turbo"
<3>:<add> {"object": "chat.completion.chunk", "choices": [], "id": chunk_id, "model": model, "created": 1},
<add> {
<add> "object": "chat.completion.chunk",
<del> {"object": "chat.completion.chunk", "choices": []},
<4>:<add> "choices": [{"delta": {"role": "assistant"}, "index": 0, "finish_reason": None}],
<add> "id": chunk_id,
<add> "model": model,
<add> "created": 1,
<add> },
<del> {"object": "chat.completion.chunk", "choices": [{"delta": {"role": "assistant"}}]},
<12>:<add> "choices": [
<add> {
<add> "delta": {"role": "assistant", "content": parts[0] + "<<"},
<del> "choices": [{"delta": {"role": "assistant", "content": parts[0] + "<<"}}],
<13>:<add> "index": 0,
<add> "finish_reason": None,
<add> }
<add> ],
<add> "id": chunk_id,
<add> "model": model,
<add> "created": 1,
<18>:<add> "choices": [
|
# module: tests.conftest
@pytest.fixture
def mock_openai_chatcompletion(monkeypatch):
<0> class AsyncChatCompletionIterator:
<1> def __init__(self, answer: str):
<2> self.responses = [
<3> {"object": "chat.completion.chunk", "choices": []},
<4> {"object": "chat.completion.chunk", "choices": [{"delta": {"role": "assistant"}}]},
<5> ]
<6> # Split at << to simulate chunked responses
<7> if answer.find("<<") > -1:
<8> parts = answer.split("<<")
<9> self.responses.append(
<10> {
<11> "object": "chat.completion.chunk",
<12> "choices": [{"delta": {"role": "assistant", "content": parts[0] + "<<"}}],
<13> }
<14> )
<15> self.responses.append(
<16> {
<17> "object": "chat.completion.chunk",
<18> "choices": [{"delta": {"role": "assistant", "content": parts[1]}}],
<19> }
<20> )
<21> else:
<22> self.responses.append(
<23> {
<24> "object": "chat.completion.chunk",
<25> "choices": [{"delta": {"content": answer}}],
<26> }
<27> )
<28>
<29> def __aiter__(self):
<30> return self
<31>
<32> async def __anext__(self):
<33> if self.responses:
<34> return self.responses.pop(0)
<35> else:
<36> raise StopAsyncIteration
<37>
<38> async def mock_acreate(*args, **kwargs):
<39> if openai.api_type == "openai":
<40> assert kwargs.get("deployment_id") is None
<41> else:
<42> assert kwargs.get("deployment_id") is not None
<43> messages = kwargs["messages"]
<44> if messages[-1]["content"] == "Generate search query for: What is the capital of France?":
<45> answer = "capital of France"
<46> else:
</s>
|
===========below chunk 0===========
# module: tests.conftest
@pytest.fixture
def mock_openai_chatcompletion(monkeypatch):
# offset: 1
if messages[0]["content"].find("Generate 3 very brief follow-up questions") > -1:
answer = "The capital of France is Paris. [Benefit_Options-2.pdf]. <<What is the capital of Spain?>>"
if "stream" in kwargs and kwargs["stream"] is True:
return AsyncChatCompletionIterator(answer)
else:
return openai.util.convert_to_openai_object(
{"object": "chat.completion", "choices": [{"message": {"role": "assistant", "content": answer}}]}
)
monkeypatch.setattr(openai.ChatCompletion, "acreate", mock_acreate)
===========changed ref 0===========
# module: tests.conftest
@pytest.fixture
def mock_openai_embedding(monkeypatch):
async def mock_acreate(*args, **kwargs):
+ return CreateEmbeddingResponse(
+ object="list",
+ data=[
+ Embedding(
+ embedding=[
+ 0.0023064255,
+ -0.009327292,
+ -0.0028842222,
+ ],
+ index=0,
+ object="embedding",
+ )
+ ],
+ model="text-embedding-ada-002",
+ usage=Usage(prompt_tokens=8, total_tokens=8),
+ )
- if openai.api_type == "openai":
- assert kwargs.get("deployment_id") is None
- else:
- assert kwargs.get("deployment_id") is not None
- return {"data": [{"embedding": [0.1, 0.2, 0.3]}]}
+ def patch(openai_client):
+ monkeypatch.setattr(openai_client.embeddings, "create", mock_acreate)
- monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 10===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 14===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 15===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 16===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
|
tests.conftest/client
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<4>:<add> mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
<add> mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
<del>
|
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
<0> quart_app = app.create_app()
<1>
<2> async with quart_app.test_app() as test_app:
<3> quart_app.config.update({"TESTING": True})
<4>
<5> yield test_app.test_client()
<6>
|
===========changed ref 0===========
# module: tests.conftest
@pytest.fixture
def mock_openai_embedding(monkeypatch):
async def mock_acreate(*args, **kwargs):
+ return CreateEmbeddingResponse(
+ object="list",
+ data=[
+ Embedding(
+ embedding=[
+ 0.0023064255,
+ -0.009327292,
+ -0.0028842222,
+ ],
+ index=0,
+ object="embedding",
+ )
+ ],
+ model="text-embedding-ada-002",
+ usage=Usage(prompt_tokens=8, total_tokens=8),
+ )
- if openai.api_type == "openai":
- assert kwargs.get("deployment_id") is None
- else:
- assert kwargs.get("deployment_id") is not None
- return {"data": [{"embedding": [0.1, 0.2, 0.3]}]}
+ def patch(openai_client):
+ monkeypatch.setattr(openai_client.embeddings, "create", mock_acreate)
- monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 10===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 14===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 15===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 16===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 18===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
===========changed ref 19===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
|
tests.conftest/auth_client
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<14>:<add> mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
<add> mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
|
# module: tests.conftest
@pytest_asyncio.fixture(params=auth_envs)
async def auth_client(
monkeypatch,
mock_openai_chatcompletion,
mock_openai_embedding,
mock_confidential_client_success,
mock_list_groups_success,
mock_acs_search_filter,
request,
):
<0> monkeypatch.setenv("AZURE_STORAGE_ACCOUNT", "test-storage-account")
<1> monkeypatch.setenv("AZURE_STORAGE_CONTAINER", "test-storage-container")
<2> monkeypatch.setenv("AZURE_SEARCH_INDEX", "test-search-index")
<3> monkeypatch.setenv("AZURE_SEARCH_SERVICE", "test-search-service")
<4> monkeypatch.setenv("AZURE_OPENAI_CHATGPT_MODEL", "gpt-35-turbo")
<5> for key, value in request.param.items():
<6> monkeypatch.setenv(key, value)
<7>
<8> with mock.patch("app.DefaultAzureCredential") as mock_default_azure_credential:
<9> mock_default_azure_credential.return_value = MockAzureCredential()
<10> quart_app = app.create_app()
<11>
<12> async with quart_app.test_app() as test_app:
<13> quart_app.config.update({"TESTING": True})
<14> client = test_app.test_client()
<15> client.config = quart_app.config
<16>
<17> yield client
<18>
|
===========changed ref 0===========
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
quart_app = app.create_app()
async with quart_app.test_app() as test_app:
quart_app.config.update({"TESTING": True})
+ mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
+ mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
-
yield test_app.test_client()
===========changed ref 1===========
# module: tests.conftest
@pytest.fixture
def mock_openai_embedding(monkeypatch):
async def mock_acreate(*args, **kwargs):
+ return CreateEmbeddingResponse(
+ object="list",
+ data=[
+ Embedding(
+ embedding=[
+ 0.0023064255,
+ -0.009327292,
+ -0.0028842222,
+ ],
+ index=0,
+ object="embedding",
+ )
+ ],
+ model="text-embedding-ada-002",
+ usage=Usage(prompt_tokens=8, total_tokens=8),
+ )
- if openai.api_type == "openai":
- assert kwargs.get("deployment_id") is None
- else:
- assert kwargs.get("deployment_id") is not None
- return {"data": [{"embedding": [0.1, 0.2, 0.3]}]}
+ def patch(openai_client):
+ monkeypatch.setattr(openai_client.embeddings, "create", mock_acreate)
- monkeypatch.setattr(openai.Embedding, "acreate", mock_acreate)
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 11===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 14===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 16===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 17===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 19===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
|
tests.test_app/test_ask_handle_exception_contentsafety
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<2>:<del> mock.Mock(
<3>:<del> side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
<4>:<del> ),
<5>:<add> mock.Mock(side_effect=filtered_response),
|
# module: tests.test_app
@pytest.mark.asyncio
async def test_ask_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
<0> monkeypatch.setattr(
<1> "approaches.retrievethenread.RetrieveThenReadApproach.run",
<2> mock.Mock(
<3> side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
<4> ),
<5> )
<6>
<7> response = await client.post(
<8> "/ask",
<9> json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
<10> )
<11> assert response.status_code == 400
<12> result = await response.get_json()
<13> assert "Exception in /ask: The response was filtered" in caplog.text
<14> snapshot.assert_match(json.dumps(result, indent=4), "result.json")
<15>
|
===========changed ref 0===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 10===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 14===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 15===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 16===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 17===========
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
quart_app = app.create_app()
async with quart_app.test_app() as test_app:
quart_app.config.update({"TESTING": True})
+ mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
+ mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
-
yield test_app.test_client()
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 19===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
===========changed ref 20===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
def wrap_credential(self) -> str:
if isinstance(self.credential, AzureKeyCredential):
return self.credential.key
if isinstance(self.credential, AsyncTokenCredential):
if not self.cached_token or self.cached_token.expires_on <= time.time():
self.cached_token = await self.credential.get_token("https://cognitiveservices.azure.com/.default")
return self.cached_token.token
+ raise TypeError("Invalid credential type")
- raise Exception("Invalid credential type")
|
tests.test_app/test_chat_handle_exception_contentsafety
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<2>:<del> mock.Mock(
<3>:<del> side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
<4>:<del> ),
<5>:<add> mock.Mock(side_effect=filtered_response),
|
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
<0> monkeypatch.setattr(
<1> "approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run",
<2> mock.Mock(
<3> side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
<4> ),
<5> )
<6>
<7> response = await client.post(
<8> "/chat",
<9> json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
<10> )
<11> assert response.status_code == 400
<12> result = await response.get_json()
<13> assert "Exception in /chat: The response was filtered" in caplog.text
<14> snapshot.assert_match(json.dumps(result, indent=4), "result.json")
<15>
|
===========changed ref 0===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 1===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_ask_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
monkeypatch.setattr(
"approaches.retrievethenread.RetrieveThenReadApproach.run",
- mock.Mock(
- side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
- ),
+ mock.Mock(side_effect=filtered_response),
)
response = await client.post(
"/ask",
json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
)
assert response.status_code == 400
result = await response.get_json()
assert "Exception in /ask: The response was filtered" in caplog.text
snapshot.assert_match(json.dumps(result, indent=4), "result.json")
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 11===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 14===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 16===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 17===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 18===========
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
quart_app = app.create_app()
async with quart_app.test_app() as test_app:
quart_app.config.update({"TESTING": True})
+ mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
+ mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
-
yield test_app.test_client()
===========changed ref 19===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 20===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
|
tests.test_app/test_chat_handle_exception_streaming
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> chat_client = client.app.config[app.CONFIG_OPENAI_CLIENT]
<1>:<add> chat_client.chat.completions, "create", mock.Mock(side_effect=ZeroDivisionError("something bad happened"))
<del> "openai.ChatCompletion.acreate", mock.Mock(side_effect=ZeroDivisionError("something bad happened"))
|
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_handle_exception_streaming(client, monkeypatch, snapshot, caplog):
<0> monkeypatch.setattr(
<1> "openai.ChatCompletion.acreate", mock.Mock(side_effect=ZeroDivisionError("something bad happened"))
<2> )
<3>
<4> response = await client.post(
<5> "/chat",
<6> json={"messages": [{"content": "What is the capital of France?", "role": "user"}], "stream": True},
<7> )
<8> assert response.status_code == 200
<9> assert "Exception while generating response stream: something bad happened" in caplog.text
<10> result = await response.get_data()
<11> snapshot.assert_match(result, "result.jsonlines")
<12>
|
===========changed ref 0===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
monkeypatch.setattr(
"approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run",
- mock.Mock(
- side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
- ),
+ mock.Mock(side_effect=filtered_response),
)
response = await client.post(
"/chat",
json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
)
assert response.status_code == 400
result = await response.get_json()
assert "Exception in /chat: The response was filtered" in caplog.text
snapshot.assert_match(json.dumps(result, indent=4), "result.json")
===========changed ref 1===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 2===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_ask_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
monkeypatch.setattr(
"approaches.retrievethenread.RetrieveThenReadApproach.run",
- mock.Mock(
- side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
- ),
+ mock.Mock(side_effect=filtered_response),
)
response = await client.post(
"/ask",
json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
)
assert response.status_code == 400
result = await response.get_json()
assert "Exception in /ask: The response was filtered" in caplog.text
snapshot.assert_match(json.dumps(result, indent=4), "result.json")
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 12===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 17===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 18===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 19===========
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
quart_app = app.create_app()
async with quart_app.test_app() as test_app:
quart_app.config.update({"TESTING": True})
+ mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
+ mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
-
yield test_app.test_client()
|
tests.test_app/test_chat_handle_exception_contentsafety_streaming
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<del> monkeypatch.setattr(
<1>:<del> "openai.ChatCompletion.acreate",
<2>:<del> mock.Mock(
<3>:<del> side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
<4>:<del> ),
<5>:<del> )
<6>:<add> chat_client = client.app.config[app.CONFIG_OPENAI_CLIENT]
<add> monkeypatch.setattr(chat_client.chat.completions, "create", mock.Mock(side_effect=filtered_response))
|
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_handle_exception_contentsafety_streaming(client, monkeypatch, snapshot, caplog):
<0> monkeypatch.setattr(
<1> "openai.ChatCompletion.acreate",
<2> mock.Mock(
<3> side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
<4> ),
<5> )
<6>
<7> response = await client.post(
<8> "/chat",
<9> json={"messages": [{"content": "How do I do something bad?", "role": "user"}], "stream": True},
<10> )
<11> assert response.status_code == 200
<12> assert "Exception while generating response stream: The response was filtered" in caplog.text
<13> result = await response.get_data()
<14> snapshot.assert_match(result, "result.jsonlines")
<15>
|
===========changed ref 0===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_handle_exception_streaming(client, monkeypatch, snapshot, caplog):
+ chat_client = client.app.config[app.CONFIG_OPENAI_CLIENT]
monkeypatch.setattr(
+ chat_client.chat.completions, "create", mock.Mock(side_effect=ZeroDivisionError("something bad happened"))
- "openai.ChatCompletion.acreate", mock.Mock(side_effect=ZeroDivisionError("something bad happened"))
)
response = await client.post(
"/chat",
json={"messages": [{"content": "What is the capital of France?", "role": "user"}], "stream": True},
)
assert response.status_code == 200
assert "Exception while generating response stream: something bad happened" in caplog.text
result = await response.get_data()
snapshot.assert_match(result, "result.jsonlines")
===========changed ref 1===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
monkeypatch.setattr(
"approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run",
- mock.Mock(
- side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
- ),
+ mock.Mock(side_effect=filtered_response),
)
response = await client.post(
"/chat",
json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
)
assert response.status_code == 400
result = await response.get_json()
assert "Exception in /chat: The response was filtered" in caplog.text
snapshot.assert_match(json.dumps(result, indent=4), "result.json")
===========changed ref 2===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 3===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_ask_handle_exception_contentsafety(client, monkeypatch, snapshot, caplog):
monkeypatch.setattr(
"approaches.retrievethenread.RetrieveThenReadApproach.run",
- mock.Mock(
- side_effect=openai.error.InvalidRequestError("The response was filtered", "prompt", code="content_filter")
- ),
+ mock.Mock(side_effect=filtered_response),
)
response = await client.post(
"/ask",
json={"messages": [{"content": "How do I do something bad?", "role": "user"}]},
)
assert response.status_code == 400
result = await response.get_json()
assert "Exception in /ask: The response was filtered" in caplog.text
snapshot.assert_match(json.dumps(result, indent=4), "result.json")
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 12===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 13===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 15===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 18===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 19===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
|
tests.test_searchmanager/test_update_content_with_embeddings
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> async def mock_create_client(*args, **kwargs):
<del> async def mock_create(*args, **kwargs):
<2>:<add> return MockClient(
<add> embeddings_client=MockEmbeddingsClient(
<add> create_embedding_response=openai.types.CreateEmbeddingResponse(
<del> return {
<3>:<add> object="list",
<del> "object": "list",
<4>:<del> "data": [
<5>:<del> {
<6>:<del> "object": "embedding",
<7>:<add> data=[
<add> openai.types.Embedding(
<add> embedding=[
<del> "embedding": [
<8>:<add> 0.0023064255,
<del> 0.0023064255,
<9>:<add> -0.009327292,
<del> -0.009327292,
<10>:<add> -0.0028842222,
<del> -0.0028842222,
<11>:<add> ],
<add> index=0,
<add> object="embedding",
<add> )
<12>:<add> model="text-embedding-ada-002",
<add> usage=Usage(prompt_tokens=8, total_tokens=8),
<del> "index": 0,
<13>:<add> )
<del> }
<14>:<add> )
<del> ],
<15>:<del> "model": "text-embedding-ada-002",
<16>:<del> "usage": {"prompt_tokens": 8, "total_tokens": 8},
<17>:<add> )
<del> }
<18>:<del>
<19>:<del> monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
<27>:<add> embeddings = AzureOpenAIEmbeddingService(
<add> open_ai_service="x",
<add>
|
# module: tests.test_searchmanager
@pytest.mark.asyncio
async def test_update_content_with_embeddings(monkeypatch, search_info):
<0> async def mock_create(*args, **kwargs):
<1> # From https://platform.openai.com/docs/api-reference/embeddings/create
<2> return {
<3> "object": "list",
<4> "data": [
<5> {
<6> "object": "embedding",
<7> "embedding": [
<8> 0.0023064255,
<9> -0.009327292,
<10> -0.0028842222,
<11> ],
<12> "index": 0,
<13> }
<14> ],
<15> "model": "text-embedding-ada-002",
<16> "usage": {"prompt_tokens": 8, "total_tokens": 8},
<17> }
<18>
<19> monkeypatch.setattr(openai.Embedding, "acreate", mock_create)
<20>
<21> documents_uploaded = []
<22>
<23> async def mock_upload_documents(self, documents):
<24> documents_uploaded.extend(documents)
<25>
<26> monkeypatch.setattr(SearchClient, "upload_documents", mock_upload_documents)
<27>
<28> manager = SearchManager(
<29> search_info,
<30> embeddings=AzureOpenAIEmbeddingService(
<31> open_ai_service="x",
<32> open_ai_deployment="x",
<33> open_ai_model_name="text-ada-003",
<34> credential=AzureKeyCredential("test"),
<35> disable_batch=True,
<36> ),
<37> )
<38>
<39> test_io = io.BytesIO(b"test content")
<40> test_io.name = "test/foo.pdf"
<41> file = File(test_io)
<42>
<43> await manager.update_content(
<44> [
<45> Section(
<46> split_page=SplitPage(
<47> page_num=0,
<48> text="test content",
<49> ),
<50> content=file</s>
|
===========below chunk 0===========
# module: tests.test_searchmanager
@pytest.mark.asyncio
async def test_update_content_with_embeddings(monkeypatch, search_info):
# offset: 1
category="test",
)
]
)
assert len(documents_uploaded) == 1, "It should have uploaded one document"
assert documents_uploaded[0]["embedding"] == [
0.0023064255,
-0.009327292,
-0.0028842222,
]
===========unchanged ref 0===========
at: _pytest.mark.structures
MARK_GEN = MarkGenerator(_ispytest=True)
at: _pytest.monkeypatch
monkeypatch() -> Generator["MonkeyPatch", None, None]
at: io
BytesIO(initial_bytes: bytes=...)
at: io.BytesIO
name: Any
at: scripts.prepdocslib.embeddings
AzureOpenAIEmbeddingService(open_ai_service: str, open_ai_deployment: str, open_ai_model_name: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], disable_batch: bool=False, verbose: bool=False)
at: scripts.prepdocslib.listfilestrategy
File(content: IO, acls: Optional[dict[str, list]]=None)
at: scripts.prepdocslib.searchmanager
SearchManager(search_info: SearchInfo, search_analyzer_name: Optional[str]=None, use_acls: bool=False, embeddings: Optional[OpenAIEmbeddings]=None)
at: scripts.prepdocslib.searchmanager.SearchManager
update_content(sections: List[Section])
at: scripts.prepdocslib.textsplitter
SplitPage(page_num: int, text: str)
at: tests.test_searchmanager
MockEmbeddingsClient(create_embedding_response: openai.types.CreateEmbeddingResponse)
MockClient(embeddings_client)
at: tests.test_searchmanager.test_update_content_many
ids = []
manager = SearchManager(
search_info,
)
sections = []
file = File(test_io)
===========changed ref 0===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 1===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 2===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 12===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 15===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 18===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 19===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<1>:<add> self.openai_client = openai_client
<del> self.openai_host = openai_host
<2>:<add> self.chatgpt_model = chatgpt_model
<3>:<del> self.chatgpt_model = chatgpt_model
|
<s>pt_model: str,
- openai_host: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
- chatgpt_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
<0> self.search_client = search_client
<1> self.openai_host = openai_host
<2> self.chatgpt_deployment = chatgpt_deployment
<3> self.chatgpt_model = chatgpt_model
<4> self.embedding_deployment = embedding_deployment
<5> self.embedding_model = embedding_model
<6> self.sourcepage_field = sourcepage_field
<7> self.content_field = content_field
<8> self.query_language = query_language
<9> self.query_speller = query_speller
<10> self.chatgpt_token_limit = get_token_limit(chatgpt_model)
<11>
|
===========unchanged ref 0===========
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach
SYSTEM = "system"
USER = "user"
ASSISTANT = "assistant"
NO_RESPONSE = "0"
system_message_chat_conversation = """Assistant helps the company employees with their healthcare plan questions, and questions about the employee handbook. Be brief in your answers.
Answer ONLY with the facts listed in the list of sources below. If there isn't enough information below, say you don't know. Do not generate answers that don't use the sources below. If asking a clarifying question to the user would help, ask the question.
For tabular information return it as an html table. Do not return markdown format. If the question is not in English, answer in the language used in the question.
Each source has a name followed by colon and the actual information, always include the source name for each fact you use in the response. Use square brackets to reference the source, for example [info1.txt]. Don't combine sources, list each source separately, for example [info1.txt][info2.pdf].
{follow_up_questions_prompt}
{injected_prompt}
"""
follow_up_questions_prompt_content = """Generate 3 very brief follow-up questions that the user would likely ask next.
Enclose the follow-up questions in double angle brackets. Example:
<<Are there exclusions for prescriptions?>>
<<Which pharmacies can be ordered from?>>
<<What is the limit for over-the-counter medication?>>
Do no repeat questions that have already been asked.
Make sure the last question ends with ">>"."""
===========unchanged ref 1===========
query_prompt_template = """Below is a history of the conversation so far, and a new question asked by the user that needs to be answered by searching in a knowledge base about employee healthcare plans and the employee handbook.
You have access to an Azure AI Search index with 100's of documents.
Generate a search query based on the conversation and the new question.
Do not include cited source filenames and document names e.g info.txt or doc.pdf in the search query terms.
Do not include any text inside [] or <<>> in the search query terms.
Do not include any special characters like '+'.
If the question is not in English, translate the question to English before generating the search query.
If you cannot generate a search query, return just the number 0.
"""
query_prompt_few_shots = [
{"role": USER, "content": "What are my health plans?"},
{"role": ASSISTANT, "content": "Show available health plans"},
{"role": USER, "content": "does my plan cover cardio?"},
{"role": ASSISTANT, "content": "Health plan cardio coverage"},
]
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 4===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 6===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 12===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 15===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 18===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.run_without_streaming
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<3>:<add> chat_completion_response: ChatCompletion = await chat_coroutine
<add> chat_resp = chat_completion_response.model_dump() # Convert to dict to make it JSON serializable
<del> chat_resp = dict(await chat_coroutine)
|
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_without_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> dict[str, Any]:
<0> extra_info, chat_coroutine = await self.run_until_final_call(
<1> history, overrides, auth_claims, should_stream=False
<2> )
<3> chat_resp = dict(await chat_coroutine)
<4> chat_resp["choices"][0]["context"] = extra_info
<5> if overrides.get("suggest_followup_questions"):
<6> content, followup_questions = self.extract_followup_questions(chat_resp["choices"][0]["message"]["content"])
<7> chat_resp["choices"][0]["message"]["content"] = content
<8> chat_resp["choices"][0]["context"]["followup_questions"] = followup_questions
<9> chat_resp["choices"][0]["session_state"] = session_state
<10> return chat_resp
<11>
|
===========unchanged ref 0===========
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach
run_until_final_call(history: list[dict[str, str]], overrides: dict[str, Any], auth_claims: dict[str, Any], should_stream: bool=False) -> tuple
extract_followup_questions(content: str)
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.__init__
self.openai_client = openai_client
self.chatgpt_model = chatgpt_model
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run_until_final_call
original_user_query = history[-1]["content"]
query_text = self.get_search_query(chat_completion, original_user_query)
query_text = None
results = [doc[self.sourcepage_field] + ": " + nonewlines(doc[self.content_field]) async for doc in r]
results = [
doc[self.sourcepage_field] + ": " + nonewlines(" . ".join([c.text for c in doc["@search.captions"]]))
async for doc in r
]
content = "\n".join(results)
messages_token_limit = self.chatgpt_token_limit - response_token_limit
messages = self.get_messages_from_history(
system_prompt=system_message,
model_id=self.chatgpt_model,
history=history,
# Model does not handle lengthy system messages well. Moving sources to latest user conversation to solve follow up questions prompt.
user_content=original_user_query + "\n\nSources:\n" + content,
max_tokens=messages_token_limit,
)
===========unchanged ref 1===========
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
===========changed ref 0===========
<s>: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_until_final_call(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
should_stream: bool = False,
+ ) -> tuple[dict[str, Any], Coroutine[Any, Any, Union[ChatCompletion, AsyncStream[ChatCompletionChunk]]]]:
- ) -> tuple:
has_text = overrides.get("retrieval_mode") in ["text", "hybrid", None]
has_vector = overrides.get("retrieval_mode") in ["vectors", "hybrid", None]
use_semantic_captions = True if overrides.get("semantic_captions") and has_text else False
top = overrides.get("top", 3)
filter = self.build_filter(overrides, auth_claims)
-
original_user_query = history[-1]["content"]
user_query_request = "Generate search query for: " + original_user_query
functions = [
{
"name": "search_sources",
"description": "Retrieve sources from the Azure AI Search index",
"parameters": {
"type": "object",
"properties": {
"search_query": {
"type": "string",
"description": "Query string to retrieve documents from azure search eg: 'Health care plan'",
}
},
"required": ["search_query"],
},
}
]
# STEP 1: Generate an optimized keyword search query based on the chat history and the last question
messages = self.get_messages_from_history(
system_prompt=self.query_prompt_template,
model_id=self.chatgpt_model,
history=history,
user_content=user_query_request,
max_tokens=self.chatgpt_token_limit - len(user_query_request),
few_shots=self.query_prompt_few</s>
===========changed ref 1===========
<s>approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_until_final_call(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
should_stream: bool = False,
+ ) -> tuple[dict[str, Any], Coroutine[Any, Any, Union[ChatCompletion, AsyncStream[ChatCompletionChunk]]]]:
- ) -> tuple:
# offset: 1
<s>gpt_token_limit - len(user_query_request),
few_shots=self.query_prompt_few_shots,
)
-
- chatgpt_args = {"deployment_id": self.chatgpt_deployment} if self.openai_host == "azure" else {}
- chat_completion = await openai.ChatCompletion.acreate(
- **chatgpt_args,
- model=self.chatgpt_model,
+ chat_completion: ChatCompletion = await self.openai_client.chat.completions.create(
+ messages=messages, # type: ignore
- messages=messages,
+ # Azure Open AI takes the deployment name as the model name
+ model=self.chatgpt_deployment if self.chatgpt_deployment else self.chatgpt_model,
temperature=0.0,
max_tokens=100, # Setting too low risks malformed JSON, setting too high may affect performance
n=1,
functions=functions,
function_call="auto",
)
query_text = self.get_search_query(chat_completion, original_user_query)
# STEP 2: Retrieve relevant documents from the search index with the GPT optimized query
# If retrieval mode includes vectors, compute an embedding for the query
vectors: list[VectorQuery] = []
if has_vector:
+ embedding = await self.openai_client</s>
===========changed ref 2===========
<s>approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_until_final_call(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
should_stream: bool = False,
+ ) -> tuple[dict[str, Any], Coroutine[Any, Any, Union[ChatCompletion, AsyncStream[ChatCompletionChunk]]]]:
- ) -> tuple:
# offset: 2
<s>dings.create(
+ # Azure Open AI takes the deployment name as the model name
+ model=self.embedding_deployment if self.embedding_deployment else self.embedding_model,
+ input=query_text,
+ )
- embedding_args = {"deployment_id": self.embedding_deployment} if self.openai_host == "azure" else {}
- embedding = await openai.Embedding.acreate(**embedding_args, model=self.embedding_model, input=query_text)
+ query_vector = embedding.data[0].embedding
- query_vector = embedding["data"][0]["embedding"]
vectors.append(RawVectorQuery(vector=query_vector, k=50, fields="embedding"))
# Only keep the text query if the retrieval mode uses text, otherwise drop it
if not has_text:
query_text = None
# Use semantic L2 reranker if requested and if retrieval mode is text or hybrid (vectors + text)
if overrides.get("semantic_ranker") and has_text:
r = await self.search_client.search(
query_text,
filter=filter,
query_type=QueryType.SEMANTIC,
query_language=self.query_language,
query_speller=self.query_speller,
semantic_configuration_name="default",
top=top,
query_caption="extractive|highlight-</s>
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.run_with_streaming
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<18>:<add> async for event_chunk in await chat_coroutine:
<del> async for event in await chat_coroutine:
<20>:<add> event = event_chunk.model_dump() # Convert pydantic model to dict
<22>:<add> content = event["choices"][0]["delta"].get("content")
<del> content = event["choices"][0]["delta"].get("content", "")
<23>:<add> content = content or "" # content may either not exist in delta, or explicitly be None
|
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
<0> extra_info, chat_coroutine = await self.run_until_final_call(
<1> history, overrides, auth_claims, should_stream=True
<2> )
<3> yield {
<4> "choices": [
<5> {
<6> "delta": {"role": self.ASSISTANT},
<7> "context": extra_info,
<8> "session_state": session_state,
<9> "finish_reason": None,
<10> "index": 0,
<11> }
<12> ],
<13> "object": "chat.completion.chunk",
<14> }
<15>
<16> followup_questions_started = False
<17> followup_content = ""
<18> async for event in await chat_coroutine:
<19> # "2023-07-01-preview" API version has a bug where first response has empty choices
<20> if event["choices"]:
<21> # if event contains << and not >>, it is start of follow-up question, truncate
<22> content = event["choices"][0]["delta"].get("content", "")
<23> if overrides.get("suggest_followup_questions") and "<<" in content:
<24> followup_questions_started = True
<25> earlier_content = content[: content.index("<<")]
<26> if earlier_content:
<27> event["choices"][0]["delta"]["content"] = earlier_content
<28> yield event
<29> followup_content += content[content.index("<<") :]
<30> elif followup_questions_started:
<31> followup_content += content
<32> else:
<33> yield event
<34> if followup_content:
<35> _, followup_questions = self.extract_followup_questions(follow</s>
|
===========below chunk 0===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
# offset: 1
yield {
"choices": [
{
"delta": {"role": self.ASSISTANT},
"context": {"followup_questions": followup_questions},
"finish_reason": None,
"index": 0,
}
],
"object": "chat.completion.chunk",
}
===========unchanged ref 0===========
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach
ASSISTANT = "assistant"
run_until_final_call(history: list[dict[str, str]], overrides: dict[str, Any], auth_claims: dict[str, Any], should_stream: bool=False) -> tuple
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run_without_streaming
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=False
)
at: typing
AsyncGenerator = _alias(collections.abc.AsyncGenerator, 2)
===========changed ref 0===========
<s>: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_until_final_call(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
should_stream: bool = False,
+ ) -> tuple[dict[str, Any], Coroutine[Any, Any, Union[ChatCompletion, AsyncStream[ChatCompletionChunk]]]]:
- ) -> tuple:
has_text = overrides.get("retrieval_mode") in ["text", "hybrid", None]
has_vector = overrides.get("retrieval_mode") in ["vectors", "hybrid", None]
use_semantic_captions = True if overrides.get("semantic_captions") and has_text else False
top = overrides.get("top", 3)
filter = self.build_filter(overrides, auth_claims)
-
original_user_query = history[-1]["content"]
user_query_request = "Generate search query for: " + original_user_query
functions = [
{
"name": "search_sources",
"description": "Retrieve sources from the Azure AI Search index",
"parameters": {
"type": "object",
"properties": {
"search_query": {
"type": "string",
"description": "Query string to retrieve documents from azure search eg: 'Health care plan'",
}
},
"required": ["search_query"],
},
}
]
# STEP 1: Generate an optimized keyword search query based on the chat history and the last question
messages = self.get_messages_from_history(
system_prompt=self.query_prompt_template,
model_id=self.chatgpt_model,
history=history,
user_content=user_query_request,
max_tokens=self.chatgpt_token_limit - len(user_query_request),
few_shots=self.query_prompt_few</s>
===========changed ref 1===========
<s>approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_until_final_call(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
should_stream: bool = False,
+ ) -> tuple[dict[str, Any], Coroutine[Any, Any, Union[ChatCompletion, AsyncStream[ChatCompletionChunk]]]]:
- ) -> tuple:
# offset: 1
<s>gpt_token_limit - len(user_query_request),
few_shots=self.query_prompt_few_shots,
)
-
- chatgpt_args = {"deployment_id": self.chatgpt_deployment} if self.openai_host == "azure" else {}
- chat_completion = await openai.ChatCompletion.acreate(
- **chatgpt_args,
- model=self.chatgpt_model,
+ chat_completion: ChatCompletion = await self.openai_client.chat.completions.create(
+ messages=messages, # type: ignore
- messages=messages,
+ # Azure Open AI takes the deployment name as the model name
+ model=self.chatgpt_deployment if self.chatgpt_deployment else self.chatgpt_model,
temperature=0.0,
max_tokens=100, # Setting too low risks malformed JSON, setting too high may affect performance
n=1,
functions=functions,
function_call="auto",
)
query_text = self.get_search_query(chat_completion, original_user_query)
# STEP 2: Retrieve relevant documents from the search index with the GPT optimized query
# If retrieval mode includes vectors, compute an embedding for the query
vectors: list[VectorQuery] = []
if has_vector:
+ embedding = await self.openai_client</s>
===========changed ref 2===========
<s>approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_until_final_call(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
should_stream: bool = False,
+ ) -> tuple[dict[str, Any], Coroutine[Any, Any, Union[ChatCompletion, AsyncStream[ChatCompletionChunk]]]]:
- ) -> tuple:
# offset: 2
<s>dings.create(
+ # Azure Open AI takes the deployment name as the model name
+ model=self.embedding_deployment if self.embedding_deployment else self.embedding_model,
+ input=query_text,
+ )
- embedding_args = {"deployment_id": self.embedding_deployment} if self.openai_host == "azure" else {}
- embedding = await openai.Embedding.acreate(**embedding_args, model=self.embedding_model, input=query_text)
+ query_vector = embedding.data[0].embedding
- query_vector = embedding["data"][0]["embedding"]
vectors.append(RawVectorQuery(vector=query_vector, k=50, fields="embedding"))
# Only keep the text query if the retrieval mode uses text, otherwise drop it
if not has_text:
query_text = None
# Use semantic L2 reranker if requested and if retrieval mode is text or hybrid (vectors + text)
if overrides.get("semantic_ranker") and has_text:
r = await self.search_client.search(
query_text,
filter=filter,
query_type=QueryType.SEMANTIC,
query_language=self.query_language,
query_speller=self.query_speller,
semantic_configuration_name="default",
top=top,
query_caption="extractive|highlight-</s>
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.run
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<3>:<del> # Workaround for: https://github.com/openai/openai-python/issues/371
<4>:<del> async with aiohttp.ClientSession() as s:
<5>:<del> openai.aiosession.set(s)
<6>:<add> return await self.run_without_streaming(messages, overrides, auth_claims, session_state)
<del> response = await self.run_without_streaming(messages, overrides, auth_claims, session_state)
<7>:<del> return response
|
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run(
self, messages: list[dict], stream: bool = False, session_state: Any = None, context: dict[str, Any] = {}
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
<0> overrides = context.get("overrides", {})
<1> auth_claims = context.get("auth_claims", {})
<2> if stream is False:
<3> # Workaround for: https://github.com/openai/openai-python/issues/371
<4> async with aiohttp.ClientSession() as s:
<5> openai.aiosession.set(s)
<6> response = await self.run_without_streaming(messages, overrides, auth_claims, session_state)
<7> return response
<8> else:
<9> return self.run_with_streaming(messages, overrides, auth_claims, session_state)
<10>
|
===========unchanged ref 0===========
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run_with_streaming
event = event_chunk.model_dump() # Convert pydantic model to dict
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
===========changed ref 0===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_without_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> dict[str, Any]:
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=False
)
+ chat_completion_response: ChatCompletion = await chat_coroutine
+ chat_resp = chat_completion_response.model_dump() # Convert to dict to make it JSON serializable
- chat_resp = dict(await chat_coroutine)
chat_resp["choices"][0]["context"] = extra_info
if overrides.get("suggest_followup_questions"):
content, followup_questions = self.extract_followup_questions(chat_resp["choices"][0]["message"]["content"])
chat_resp["choices"][0]["message"]["content"] = content
chat_resp["choices"][0]["context"]["followup_questions"] = followup_questions
chat_resp["choices"][0]["session_state"] = session_state
return chat_resp
===========changed ref 1===========
<s>pt_model: str,
- openai_host: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
- chatgpt_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
self.search_client = search_client
+ self.openai_client = openai_client
- self.openai_host = openai_host
+ self.chatgpt_model = chatgpt_model
self.chatgpt_deployment = chatgpt_deployment
- self.chatgpt_model = chatgpt_model
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
self.chatgpt_token_limit = get_token_limit(chatgpt_model)
===========changed ref 2===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=True
)
yield {
"choices": [
{
"delta": {"role": self.ASSISTANT},
"context": extra_info,
"session_state": session_state,
"finish_reason": None,
"index": 0,
}
],
"object": "chat.completion.chunk",
}
followup_questions_started = False
followup_content = ""
+ async for event_chunk in await chat_coroutine:
- async for event in await chat_coroutine:
# "2023-07-01-preview" API version has a bug where first response has empty choices
+ event = event_chunk.model_dump() # Convert pydantic model to dict
if event["choices"]:
# if event contains << and not >>, it is start of follow-up question, truncate
+ content = event["choices"][0]["delta"].get("content")
- content = event["choices"][0]["delta"].get("content", "")
+ content = content or "" # content may either not exist in delta, or explicitly be None
if overrides.get("suggest_followup_questions") and "<<" in content:
followup_questions_started = True
earlier_content = content[: content.index("<<")]
if earlier_content:
event["choices"][0]["delta"]["content"] = earlier_content
yield event
followup_content += content[content.index("<<") :]
elif followup_questions_</s>
===========changed ref 3===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
# offset: 1
<s>content
yield event
followup_content += content[content.index("<<") :]
elif followup_questions_started:
followup_content += content
else:
yield event
if followup_content:
_, followup_questions = self.extract_followup_questions(followup_content)
yield {
"choices": [
{
"delta": {"role": self.ASSISTANT},
"context": {"followup_questions": followup_questions},
"finish_reason": None,
"index": 0,
}
],
"object": "chat.completion.chunk",
}
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 6===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 8===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 10===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.get_messages_from_history
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<9>:<add> total_token_count = message_builder.count_tokens_for_message(dict(message_builder.messages[-1])) # type: ignore
<del> total_token_count = message_builder.count_tokens_for_message(message_builder.messages[-1])
|
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def get_messages_from_history(
self,
system_prompt: str,
model_id: str,
history: list[dict[str, str]],
user_content: str,
max_tokens: int,
few_shots=[],
+ ) -> list[ChatCompletionMessageParam]:
- ) -> list:
<0> message_builder = MessageBuilder(system_prompt, model_id)
<1>
<2> # Add examples to show the chat what responses we want. It will try to mimic any responses and make sure they match the rules laid out in the system message.
<3> for shot in reversed(few_shots):
<4> message_builder.insert_message(shot.get("role"), shot.get("content"))
<5>
<6> append_index = len(few_shots) + 1
<7>
<8> message_builder.insert_message(self.USER, user_content, index=append_index)
<9> total_token_count = message_builder.count_tokens_for_message(message_builder.messages[-1])
<10>
<11> newest_to_oldest = list(reversed(history[:-1]))
<12> for message in newest_to_oldest:
<13> potential_message_count = message_builder.count_tokens_for_message(message)
<14> if (total_token_count + potential_message_count) > max_tokens:
<15> logging.debug("Reached max tokens of %d, history will be truncated", max_tokens)
<16> break
<17> message_builder.insert_message(message["role"], message["content"], index=append_index)
<18> total_token_count += potential_message_count
<19> return message_builder.messages
<20>
|
===========unchanged ref 0===========
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach
ASSISTANT = "assistant"
extract_followup_questions(content: str)
at: app.backend.approaches.chatreadretrieveread.ChatReadRetrieveReadApproach.run_with_streaming
followup_content += content[content.index("<<") :]
followup_content = ""
followup_content += content
at: approaches.approach.Approach
run(self, messages: list[dict], stream: bool=False, session_state: Any=None, context: dict[str, Any]={}) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]
at: typing
AsyncGenerator = _alias(collections.abc.AsyncGenerator, 2)
===========changed ref 0===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run(
self, messages: list[dict], stream: bool = False, session_state: Any = None, context: dict[str, Any] = {}
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
overrides = context.get("overrides", {})
auth_claims = context.get("auth_claims", {})
if stream is False:
- # Workaround for: https://github.com/openai/openai-python/issues/371
- async with aiohttp.ClientSession() as s:
- openai.aiosession.set(s)
+ return await self.run_without_streaming(messages, overrides, auth_claims, session_state)
- response = await self.run_without_streaming(messages, overrides, auth_claims, session_state)
- return response
else:
return self.run_with_streaming(messages, overrides, auth_claims, session_state)
===========changed ref 1===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_without_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> dict[str, Any]:
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=False
)
+ chat_completion_response: ChatCompletion = await chat_coroutine
+ chat_resp = chat_completion_response.model_dump() # Convert to dict to make it JSON serializable
- chat_resp = dict(await chat_coroutine)
chat_resp["choices"][0]["context"] = extra_info
if overrides.get("suggest_followup_questions"):
content, followup_questions = self.extract_followup_questions(chat_resp["choices"][0]["message"]["content"])
chat_resp["choices"][0]["message"]["content"] = content
chat_resp["choices"][0]["context"]["followup_questions"] = followup_questions
chat_resp["choices"][0]["session_state"] = session_state
return chat_resp
===========changed ref 2===========
<s>pt_model: str,
- openai_host: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
- chatgpt_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
self.search_client = search_client
+ self.openai_client = openai_client
- self.openai_host = openai_host
+ self.chatgpt_model = chatgpt_model
self.chatgpt_deployment = chatgpt_deployment
- self.chatgpt_model = chatgpt_model
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
self.chatgpt_token_limit = get_token_limit(chatgpt_model)
===========changed ref 3===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=True
)
yield {
"choices": [
{
"delta": {"role": self.ASSISTANT},
"context": extra_info,
"session_state": session_state,
"finish_reason": None,
"index": 0,
}
],
"object": "chat.completion.chunk",
}
followup_questions_started = False
followup_content = ""
+ async for event_chunk in await chat_coroutine:
- async for event in await chat_coroutine:
# "2023-07-01-preview" API version has a bug where first response has empty choices
+ event = event_chunk.model_dump() # Convert pydantic model to dict
if event["choices"]:
# if event contains << and not >>, it is start of follow-up question, truncate
+ content = event["choices"][0]["delta"].get("content")
- content = event["choices"][0]["delta"].get("content", "")
+ content = content or "" # content may either not exist in delta, or explicitly be None
if overrides.get("suggest_followup_questions") and "<<" in content:
followup_questions_started = True
earlier_content = content[: content.index("<<")]
if earlier_content:
event["choices"][0]["delta"]["content"] = earlier_content
yield event
followup_content += content[content.index("<<") :]
elif followup_questions_</s>
===========changed ref 4===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
# offset: 1
<s>content
yield event
followup_content += content[content.index("<<") :]
elif followup_questions_started:
followup_content += content
else:
yield event
if followup_content:
_, followup_questions = self.extract_followup_questions(followup_content)
yield {
"choices": [
{
"delta": {"role": self.ASSISTANT},
"context": {"followup_questions": followup_questions},
"finish_reason": None,
"index": 0,
}
],
"object": "chat.completion.chunk",
}
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.get_search_query
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> response_message = chat_completion.choices[0].message
<del> response_message = chat_completion["choices"][0]["message"]
<1>:<add> if function_call := response_message.function_call:
<del> if function_call := response_message.get("function_call"):
<2>:<add> if function_call.name == "search_sources":
<del> if function_call["name"] == "search_sources":
<3>:<add> arg = json.loads(function_call.arguments)
<del> arg = json.loads(function_call["arguments"])
<7>:<add> elif query_text := response_message.content:
<del> elif query_text := response_message.get("content"):
|
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
+ def get_search_query(self, chat_completion: ChatCompletion, user_query: str):
- def get_search_query(self, chat_completion: dict[str, Any], user_query: str):
<0> response_message = chat_completion["choices"][0]["message"]
<1> if function_call := response_message.get("function_call"):
<2> if function_call["name"] == "search_sources":
<3> arg = json.loads(function_call["arguments"])
<4> search_query = arg.get("search_query", self.NO_RESPONSE)
<5> if search_query != self.NO_RESPONSE:
<6> return search_query
<7> elif query_text := response_message.get("content"):
<8> if query_text.strip() != self.NO_RESPONSE:
<9> return query_text
<10> return user_query
<11>
|
===========unchanged ref 0===========
at: core.messagebuilder
MessageBuilder(system_content: str, chatgpt_model: str)
===========changed ref 0===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run(
self, messages: list[dict], stream: bool = False, session_state: Any = None, context: dict[str, Any] = {}
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
overrides = context.get("overrides", {})
auth_claims = context.get("auth_claims", {})
if stream is False:
- # Workaround for: https://github.com/openai/openai-python/issues/371
- async with aiohttp.ClientSession() as s:
- openai.aiosession.set(s)
+ return await self.run_without_streaming(messages, overrides, auth_claims, session_state)
- response = await self.run_without_streaming(messages, overrides, auth_claims, session_state)
- return response
else:
return self.run_with_streaming(messages, overrides, auth_claims, session_state)
===========changed ref 1===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_without_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> dict[str, Any]:
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=False
)
+ chat_completion_response: ChatCompletion = await chat_coroutine
+ chat_resp = chat_completion_response.model_dump() # Convert to dict to make it JSON serializable
- chat_resp = dict(await chat_coroutine)
chat_resp["choices"][0]["context"] = extra_info
if overrides.get("suggest_followup_questions"):
content, followup_questions = self.extract_followup_questions(chat_resp["choices"][0]["message"]["content"])
chat_resp["choices"][0]["message"]["content"] = content
chat_resp["choices"][0]["context"]["followup_questions"] = followup_questions
chat_resp["choices"][0]["session_state"] = session_state
return chat_resp
===========changed ref 2===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def get_messages_from_history(
self,
system_prompt: str,
model_id: str,
history: list[dict[str, str]],
user_content: str,
max_tokens: int,
few_shots=[],
+ ) -> list[ChatCompletionMessageParam]:
- ) -> list:
message_builder = MessageBuilder(system_prompt, model_id)
# Add examples to show the chat what responses we want. It will try to mimic any responses and make sure they match the rules laid out in the system message.
for shot in reversed(few_shots):
message_builder.insert_message(shot.get("role"), shot.get("content"))
append_index = len(few_shots) + 1
message_builder.insert_message(self.USER, user_content, index=append_index)
+ total_token_count = message_builder.count_tokens_for_message(dict(message_builder.messages[-1])) # type: ignore
- total_token_count = message_builder.count_tokens_for_message(message_builder.messages[-1])
newest_to_oldest = list(reversed(history[:-1]))
for message in newest_to_oldest:
potential_message_count = message_builder.count_tokens_for_message(message)
if (total_token_count + potential_message_count) > max_tokens:
logging.debug("Reached max tokens of %d, history will be truncated", max_tokens)
break
message_builder.insert_message(message["role"], message["content"], index=append_index)
total_token_count += potential_message_count
return message_builder.messages
===========changed ref 3===========
<s>pt_model: str,
- openai_host: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
- chatgpt_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
self.search_client = search_client
+ self.openai_client = openai_client
- self.openai_host = openai_host
+ self.chatgpt_model = chatgpt_model
self.chatgpt_deployment = chatgpt_deployment
- self.chatgpt_model = chatgpt_model
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
self.chatgpt_token_limit = get_token_limit(chatgpt_model)
===========changed ref 4===========
# module: app.backend.approaches.chatreadretrieveread
class ChatReadRetrieveReadApproach(Approach):
def run_with_streaming(
self,
history: list[dict[str, str]],
overrides: dict[str, Any],
auth_claims: dict[str, Any],
session_state: Any = None,
) -> AsyncGenerator[dict, None]:
extra_info, chat_coroutine = await self.run_until_final_call(
history, overrides, auth_claims, should_stream=True
)
yield {
"choices": [
{
"delta": {"role": self.ASSISTANT},
"context": extra_info,
"session_state": session_state,
"finish_reason": None,
"index": 0,
}
],
"object": "chat.completion.chunk",
}
followup_questions_started = False
followup_content = ""
+ async for event_chunk in await chat_coroutine:
- async for event in await chat_coroutine:
# "2023-07-01-preview" API version has a bug where first response has empty choices
+ event = event_chunk.model_dump() # Convert pydantic model to dict
if event["choices"]:
# if event contains << and not >>, it is start of follow-up question, truncate
+ content = event["choices"][0]["delta"].get("content")
- content = event["choices"][0]["delta"].get("content", "")
+ content = content or "" # content may either not exist in delta, or explicitly be None
if overrides.get("suggest_followup_questions") and "<<" in content:
followup_questions_started = True
earlier_content = content[: content.index("<<")]
if earlier_content:
event["choices"][0]["delta"]["content"] = earlier_content
yield event
followup_content += content[content.index("<<") :]
elif followup_questions_</s>
|
app.backend.approaches.retrievethenread/RetrieveThenReadApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<1>:<add> self.openai_client = openai_client
<del> self.openai_host = openai_host
<2>:<del> self.chatgpt_deployment = chatgpt_deployment
<5>:<add> self.chatgpt_deployment = chatgpt_deployment
|
<s>ai_host: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
+ embedding_model: str,
- chatgpt_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
- embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
<0> self.search_client = search_client
<1> self.openai_host = openai_host
<2> self.chatgpt_deployment = chatgpt_deployment
<3> self.chatgpt_model = chatgpt_model
<4> self.embedding_model = embedding_model
<5> self.embedding_deployment = embedding_deployment
<6> self.sourcepage_field = sourcepage_field
<7> self.content_field = content_field
<8> self.query_language = query_language
<9> self.query_speller = query_speller
<10>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 3===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 4===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 6===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 10===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 12===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 15===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 18===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 19===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 20===========
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
quart_app = app.create_app()
async with quart_app.test_app() as test_app:
quart_app.config.update({"TESTING": True})
+ mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
+ mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
-
yield test_app.test_client()
===========changed ref 21===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "deployment_id": self.open_ai_deployment,
- "api_type": self.get_api_type(),
- "api_key": await self.wrap_credential(),
- "api_version": "2023-05-15",
- "api_base": f"https://{self.open_ai_service}.openai.azure.com",
- }
-
===========changed ref 22===========
# module: tests.test_chatapproach
+ def test_extract_followup_questions_no_followup(chat_approach):
- def test_extract_followup_questions_no_followup():
- chat_approach = ChatReadRetrieveReadApproach(
- None, "", "gpt-35-turbo", "gpt-35-turbo", "", "", "", "", "en-us", "lexicon"
- )
-
content = "Here is answer to your question."
pre_content, followup_questions = chat_approach.extract_followup_questions(content)
assert pre_content == "Here is answer to your question."
assert followup_questions == []
|
app.backend.approaches.retrievethenread/RetrieveThenReadApproach.run
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<8>:<del>
<12>:<add> embedding = await self.openai_client.embeddings.create(
<add> # Azure Open AI takes the deployment name as the model name
<add> model=self.embedding_deployment if self.embedding_deployment else self.embedding_model,
<add> input=q,
<add> )
<del> embedding_args = {"deployment_id": self.embedding_deployment} if self.openai_host == "azure" else {}
<13>:<del> embedding = await openai.Embedding.acreate(**embedding_args, model=self.embedding_model, input=q)
<14>:<add> query_vector = embedding.data[0].embedding
<del> query_vector = embedding["data"][0]["embedding"]
|
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
<0> q = messages[-1]["content"]
<1> overrides = context.get("overrides", {})
<2> auth_claims = context.get("auth_claims", {})
<3> has_text = overrides.get("retrieval_mode") in ["text", "hybrid", None]
<4> has_vector = overrides.get("retrieval_mode") in ["vectors", "hybrid", None]
<5> use_semantic_captions = True if overrides.get("semantic_captions") and has_text else False
<6> top = overrides.get("top", 3)
<7> filter = self.build_filter(overrides, auth_claims)
<8>
<9> # If retrieval mode includes vectors, compute an embedding for the query
<10> vectors: list[VectorQuery] = []
<11> if has_vector:
<12> embedding_args = {"deployment_id": self.embedding_deployment} if self.openai_host == "azure" else {}
<13> embedding = await openai.Embedding.acreate(**embedding_args, model=self.embedding_model, input=q)
<14> query_vector = embedding["data"][0]["embedding"]
<15> vectors.append(RawVectorQuery(vector=query_vector, k=50, fields="embedding"))
<16>
<17> # Only keep the text query if the retrieval mode uses text, otherwise drop it
<18> query_text = q if has_text else ""
<19>
<20> # Use semantic ranker if requested and if retrieval mode is text or hybrid (vectors + text)
<21> if overrides.get("semantic_ranker") and has_text:
<22> r = await self.search_client.search(
<23> query_text,
<24> filter=filter,
</s>
|
===========below chunk 0===========
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
# offset: 1
query_language=self.query_language,
query_speller=self.query_speller,
semantic_configuration_name="default",
top=top,
query_caption="extractive|highlight-false" if use_semantic_captions else None,
vector_queries=vectors,
)
else:
r = await self.search_client.search(
query_text,
filter=filter,
top=top,
vector_queries=vectors,
)
if use_semantic_captions:
results = [
doc[self.sourcepage_field] + ": " + nonewlines(" . ".join([c.text for c in doc["@search.captions"]]))
async for doc in r
]
else:
results = [doc[self.sourcepage_field] + ": " + nonewlines(doc[self.content_field]) async for doc in r]
content = "\n".join(results)
message_builder = MessageBuilder(
overrides.get("prompt_template") or self.system_chat_template, self.chatgpt_model
)
# add user question
user_content = q + "\n" + f"Sources:\n {content}"
message_builder.insert_message("user", user_content)
# Add shots/samples. This helps model to mimic response and make sure they match rules laid out in system message.
message_builder.insert_message("assistant", self.answer)
message_builder.insert_message("user", self.question)
messages = message_builder.messages
chatgpt</s>
===========below chunk 1===========
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
# offset: 2
<s> message_builder.insert_message("user", self.question)
messages = message_builder.messages
chatgpt_args = {"deployment_id": self.chatgpt_deployment} if self.openai_host == "azure" else {}
chat_completion = await openai.ChatCompletion.acreate(
**chatgpt_args,
model=self.chatgpt_model,
messages=messages,
temperature=overrides.get("temperature") or 0.3,
max_tokens=1024,
n=1,
)
extra_info = {
"data_points": results,
"thoughts": f"Question:<br>{query_text}<br><br>Prompt:<br>"
+ "\n\n".join([str(message) for message in messages]),
}
chat_completion.choices[0]["context"] = extra_info
chat_completion.choices[0]["session_state"] = session_state
return chat_completion
===========unchanged ref 0===========
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach
system_chat_template = (
"You are an intelligent assistant helping Contoso Inc employees with their healthcare plan questions and employee handbook questions. "
+ "Use 'you' to refer to the individual asking the questions even if they ask with 'I'. "
+ "Answer the following question using only the data provided in the sources below. "
+ "For tabular information return it as an html table. Do not return markdown format. "
+ "Each source has a name followed by colon and the actual information, always include the source name for each fact you use in the response. "
+ "If you cannot answer using the sources below, say you don't know. Use below example to answer"
)
question = """
'What is the deductible for the employee plan for a visit to Overlake in Bellevue?'
Sources:
info1.txt: deductibles depend on whether you are in-network or out-of-network. In-network deductibles are $500 for employee and $1000 for family. Out-of-network deductibles are $1000 for employee and $2000 for family.
info2.pdf: Overlake is in-network for the employee plan.
info3.pdf: Overlake is the name of the area that includes a park and ride near Bellevue.
info4.pdf: In-network institutions include Overlake, Swedish and others in the region
"""
answer = "In-network deductibles are $500 for employee and $1000 for family [info1.txt] and Overlake is in-network for the employee plan [info2.pdf][info4.pdf]."
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach.__init__
self.search_client = search_client
self.chatgpt_deployment = chatgpt_deployment
self.openai_client = openai_client
self.chatgpt_model = chatgpt_model
===========unchanged ref 1===========
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
at: approaches.approach.Approach
build_filter(overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]
run(self, messages: list[dict], stream: bool=False, session_state: Any=None, context: dict[str, Any]={}) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]
at: core.messagebuilder
MessageBuilder(system_content: str, chatgpt_model: str)
at: core.messagebuilder.MessageBuilder
insert_message(role: str, content: str, index: int=1)
at: core.messagebuilder.MessageBuilder.__init__
self.messages: list[ChatCompletionMessageParam] = [
ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
]
at: text
nonewlines(s: str) -> str
at: typing
AsyncGenerator = _alias(collections.abc.AsyncGenerator, 2)
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
|
app.backend.app/error_dict
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<0>:<add> if isinstance(error, APIError) and error.code == "content_filter":
<del> if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
|
# module: app.backend.app
def error_dict(error: Exception) -> dict:
<0> if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
<1> return {"error": ERROR_MESSAGE_FILTER}
<2> return {"error": ERROR_MESSAGE.format(error_type=type(error))}
<3>
|
===========unchanged ref 0===========
at: app.backend.app
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
at: openai.error
InvalidRequestError(message, param, code=None, http_body=None, http_status=None, json_body=None, headers=None)
===========changed ref 0===========
# module: app.backend.app
- CONFIG_OPENAI_TOKEN = "openai_token"
- CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
CONFIG_SEARCH_CLIENT = "search_client"
+ CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 4===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 5===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 7===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 11===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 12===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 13===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 15===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 16===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 19===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 20===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
===========changed ref 21===========
# module: tests.conftest
@pytest_asyncio.fixture()
async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
quart_app = app.create_app()
async with quart_app.test_app() as test_app:
quart_app.config.update({"TESTING": True})
+ mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
+ mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
-
yield test_app.test_client()
|
app.backend.app/error_response
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<1>:<add> if isinstance(error, APIError) and error.code == "content_filter":
<del> if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
|
# module: app.backend.app
def error_response(error: Exception, route: str, status_code: int = 500):
<0> logging.exception("Exception in %s: %s", route, error)
<1> if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
<2> status_code = 400
<3> return jsonify(error_dict(error)), status_code
<4>
|
===========unchanged ref 0===========
at: app.backend.app
error_dict(error: Exception) -> dict
at: logging
exception(msg: Any, *args: Any, exc_info: _ExcInfoType=..., stack_info: bool=..., extra: Optional[Dict[str, Any]]=..., **kwargs: Any) -> None
at: openai.error
InvalidRequestError(message, param, code=None, http_body=None, http_status=None, json_body=None, headers=None)
===========changed ref 0===========
# module: app.backend.app
def error_dict(error: Exception) -> dict:
+ if isinstance(error, APIError) and error.code == "content_filter":
- if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
return {"error": ERROR_MESSAGE_FILTER}
return {"error": ERROR_MESSAGE.format(error_type=type(error))}
===========changed ref 1===========
# module: app.backend.app
- CONFIG_OPENAI_TOKEN = "openai_token"
- CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
CONFIG_SEARCH_CLIENT = "search_client"
+ CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 5===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 6===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 7===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 8===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 9===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 12===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 14===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 15===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 16===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 17===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
===========changed ref 19===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncAzureOpenAI(
+ azure_endpoint=f"https://{self.open_ai_service}.openai.azure.com",
+ azure_deployment=self.open_ai_deployment,
+ api_key=await self.wrap_credential(),
+ api_version="2023-05-15",
+ )
+
===========changed ref 20===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
+ self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
- self.messages = [{"role": "system", "content": self.normalize_content(system_content)}]
+ ]
self.model = chatgpt_model
===========changed ref 21===========
# module: tests.test_chatapproach
+ @pytest.fixture
+ def chat_approach():
+ return ChatReadRetrieveReadApproach(
+ search_client=None,
+ openai_client=None,
+ chatgpt_model="gpt-35-turbo",
+ chatgpt_deployment="chat",
+ embedding_deployment="embeddings",
+ embedding_model="text-",
+ sourcepage_field="",
+ content_field="",
+ query_language="en-us",
+ query_speller="lexicon",
+ )
+
|
app.backend.app/ask
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<8>:<del> # Workaround for: https://github.com/openai/openai-python/issues/371
<9>:<del> async with aiohttp.ClientSession() as s:
<10>:<del> openai.aiosession.set(s)
<11>:<add> r = await approach.run(
<del> r = await approach.run(
<12>:<add> request_json["messages"], context=context, session_state=request_json.get("session_state")
<del> request_json["messages"], context=context, session_state=request_json.get("session_state")
<13>:<add> )
<del> )
|
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
<0> if not request.is_json:
<1> return jsonify({"error": "request must be json"}), 415
<2> request_json = await request.get_json()
<3> context = request_json.get("context", {})
<4> auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
<5> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<6> try:
<7> approach = current_app.config[CONFIG_ASK_APPROACH]
<8> # Workaround for: https://github.com/openai/openai-python/issues/371
<9> async with aiohttp.ClientSession() as s:
<10> openai.aiosession.set(s)
<11> r = await approach.run(
<12> request_json["messages"], context=context, session_state=request_json.get("session_state")
<13> )
<14> return jsonify(r)
<15> except Exception as error:
<16> return error_response(error, "/ask")
<17>
|
===========unchanged ref 0===========
at: _contextvars.ContextVar
set(value, /)
at: aiohttp.client
ClientSession(base_url: Optional[StrOrURL]=None, *, connector: Optional[BaseConnector]=None, loop: Optional[asyncio.AbstractEventLoop]=None, cookies: Optional[LooseCookies]=None, headers: Optional[LooseHeaders]=None, skip_auto_headers: Optional[Iterable[str]]=None, auth: Optional[BasicAuth]=None, json_serialize: JSONEncoder=json.dumps, request_class: Type[ClientRequest]=ClientRequest, response_class: Type[ClientResponse]=ClientResponse, ws_response_class: Type[ClientWebSocketResponse]=ClientWebSocketResponse, version: HttpVersion=http.HttpVersion11, cookie_jar: Optional[AbstractCookieJar]=None, connector_owner: bool=True, raise_for_status: bool=False, read_timeout: Union[float, object]=sentinel, conn_timeout: Optional[float]=None, timeout: Union[object, ClientTimeout]=sentinel, auto_decompress: bool=True, trust_env: bool=False, requote_redirect_url: bool=True, trace_configs: Optional[List[TraceConfig]]=None, read_bufsize: int=2**16, fallback_charset_resolver: _CharsetResolver=(
_default_fallback_charset_resolver
))
at: app.backend.app
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_AUTH_CLIENT = "auth_client"
bp = Blueprint("routes", __name__, static_folder="static")
error_response(error: Exception, route: str, status_code: int=500)
at: openai
aiosession: ContextVar[Optional["ClientSession"]] = ContextVar(
"aiohttp-session", default=None
) # Acts as a global aiohttp ClientSession that reuses connections.
===========changed ref 0===========
# module: app.backend.app
def error_response(error: Exception, route: str, status_code: int = 500):
logging.exception("Exception in %s: %s", route, error)
+ if isinstance(error, APIError) and error.code == "content_filter":
- if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
status_code = 400
return jsonify(error_dict(error)), status_code
===========changed ref 1===========
# module: app.backend.app
def error_dict(error: Exception) -> dict:
+ if isinstance(error, APIError) and error.code == "content_filter":
- if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
return {"error": ERROR_MESSAGE_FILTER}
return {"error": ERROR_MESSAGE.format(error_type=type(error))}
===========changed ref 2===========
# module: app.backend.app
- CONFIG_OPENAI_TOKEN = "openai_token"
- CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
CONFIG_SEARCH_CLIENT = "search_client"
+ CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 5===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 6===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 7===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 9===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 11===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 13===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 14===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 15===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 16===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 17===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
===========changed ref 18===========
# module: tests.test_prepdocs
+ class RateLimitMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.RateLimitError(
+ message="Rate limited on the OpenAI embeddings API", response=fake_response(409), body=None
+ )
+
===========changed ref 19===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
|
app.backend.app/setup_clients
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<11>:<add> AZURE_OPENAI_CHATGPT_DEPLOYMENT = os.getenv("AZURE_OPENAI_CHATGPT_DEPLOYMENT") if OPENAI_HOST == "azure" else None
<del> AZURE_OPENAI_CHATGPT_DEPLOYMENT = os.getenv("AZURE_OPENAI_CHATGPT_DEPLOYMENT")
<12>:<add> AZURE_OPENAI_EMB_DEPLOYMENT = os.getenv("AZURE_OPENAI_EMB_DEPLOYMENT") if OPENAI_HOST == "azure" else None
<del> AZURE_OPENAI_EMB_DEPLOYMENT = os.getenv("AZURE_OPENAI_EMB_DEPLOYMENT")
|
# module: app.backend.app
@bp.before_app_serving
async def setup_clients():
<0> # Replace these with your own values, either in environment variables or directly here
<1> AZURE_STORAGE_ACCOUNT = os.environ["AZURE_STORAGE_ACCOUNT"]
<2> AZURE_STORAGE_CONTAINER = os.environ["AZURE_STORAGE_CONTAINER"]
<3> AZURE_SEARCH_SERVICE = os.environ["AZURE_SEARCH_SERVICE"]
<4> AZURE_SEARCH_INDEX = os.environ["AZURE_SEARCH_INDEX"]
<5> # Shared by all OpenAI deployments
<6> OPENAI_HOST = os.getenv("OPENAI_HOST", "azure")
<7> OPENAI_CHATGPT_MODEL = os.environ["AZURE_OPENAI_CHATGPT_MODEL"]
<8> OPENAI_EMB_MODEL = os.getenv("AZURE_OPENAI_EMB_MODEL_NAME", "text-embedding-ada-002")
<9> # Used with Azure OpenAI deployments
<10> AZURE_OPENAI_SERVICE = os.getenv("AZURE_OPENAI_SERVICE")
<11> AZURE_OPENAI_CHATGPT_DEPLOYMENT = os.getenv("AZURE_OPENAI_CHATGPT_DEPLOYMENT")
<12> AZURE_OPENAI_EMB_DEPLOYMENT = os.getenv("AZURE_OPENAI_EMB_DEPLOYMENT")
<13> # Used only with non-Azure OpenAI deployments
<14> OPENAI_API_KEY = os.getenv("OPENAI_API_KEY")
<15> OPENAI_ORGANIZATION = os.getenv("OPENAI_ORGANIZATION")
<16> AZURE_USE_AUTHENTICATION = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
<17> AZURE_SERVER_APP_ID = os.getenv("AZURE_SERVER_APP_ID")
<18> AZURE_SERVER_APP_SECRET = os.getenv("AZURE_SERVER_APP_SECRET")
<19> AZURE_CLIENT_APP_ID = os.getenv("AZURE_CLIENT_APP_ID")</s>
|
===========below chunk 0===========
# module: app.backend.app
@bp.before_app_serving
async def setup_clients():
# offset: 1
TOKEN_CACHE_PATH = os.getenv("TOKEN_CACHE_PATH")
KB_FIELDS_CONTENT = os.getenv("KB_FIELDS_CONTENT", "content")
KB_FIELDS_SOURCEPAGE = os.getenv("KB_FIELDS_SOURCEPAGE", "sourcepage")
AZURE_SEARCH_QUERY_LANGUAGE = os.getenv("AZURE_SEARCH_QUERY_LANGUAGE", "en-us")
AZURE_SEARCH_QUERY_SPELLER = os.getenv("AZURE_SEARCH_QUERY_SPELLER", "lexicon")
# Use the current user identity to authenticate with Azure OpenAI, AI Search and Blob Storage (no secrets needed,
# just use 'az login' locally, and managed identity when deployed on Azure). If you need to use keys, use separate AzureKeyCredential instances with the
# keys for each service
# If you encounter a blocking error during a DefaultAzureCredential resolution, you can exclude the problematic credential by using a parameter (ex. exclude_shared_token_cache_credential=True)
azure_credential = DefaultAzureCredential(exclude_shared_token_cache_credential=True)
# Set up authentication helper
auth_helper = AuthenticationHelper(
use_authentication=AZURE_USE_AUTHENTICATION,
server_app_id=AZURE_SERVER_APP_ID,
server_app_secret=AZURE_SERVER_APP_SECRET,
client_app_id=AZURE_CLIENT_APP_ID,
tenant_id=AZURE_TENANT_ID,
token_cache_path=TOKEN_CACHE_PATH,
)
# Set up clients for AI Search and Storage
search_client = SearchClient(
endpoint=f"https://{AZURE_SEARCH_SERVICE}.search.windows.net",
index_name=AZURE_SEARCH_INDEX,
credential=azure_credential,
)
blob_client = BlobServiceClient(
account_url=f"https://{AZURE_STORAGE_ACCOUNT}.blob.core.</s>
===========below chunk 1===========
# module: app.backend.app
@bp.before_app_serving
async def setup_clients():
# offset: 2
<s> blob_client = BlobServiceClient(
account_url=f"https://{AZURE_STORAGE_ACCOUNT}.blob.core.windows.net", credential=azure_credential
)
blob_container_client = blob_client.get_container_client(AZURE_STORAGE_CONTAINER)
# Used by the OpenAI SDK
if OPENAI_HOST == "azure":
openai.api_type = "azure_ad"
openai.api_base = f"https://{AZURE_OPENAI_SERVICE}.openai.azure.com"
openai.api_version = "2023-07-01-preview"
openai_token = await azure_credential.get_token("https://cognitiveservices.azure.com/.default")
openai.api_key = openai_token.token
# Store on app.config for later use inside requests
current_app.config[CONFIG_OPENAI_TOKEN] = openai_token
else:
openai.api_type = "openai"
openai.api_key = OPENAI_API_KEY
openai.organization = OPENAI_ORGANIZATION
current_app.config[CONFIG_CREDENTIAL] = azure_credential
current_app.config[CONFIG_SEARCH_CLIENT] = search_client
current_app.config[CONFIG_BLOB_CONTAINER_CLIENT] = blob_container_client
current_app.config[CONFIG_AUTH_CLIENT] = auth_helper
# Various approaches to integrate GPT and external knowledge, most applications will use a single one of these patterns
# or some derivative, here we include several for exploration purposes
current_app.config[CONFIG_ASK_APPROACH] = RetrieveThenReadApproach(
search_client,
OPENAI_HOST,
AZURE_OPENAI_CHATGPT_DEPLOYMENT,
</s>
===========below chunk 2===========
# module: app.backend.app
@bp.before_app_serving
async def setup_clients():
# offset: 3
<s>_CHATGPT_MODEL,
AZURE_OPENAI_EMB_DEPLOYMENT,
OPENAI_EMB_MODEL,
KB_FIELDS_SOURCEPAGE,
KB_FIELDS_CONTENT,
AZURE_SEARCH_QUERY_LANGUAGE,
AZURE_SEARCH_QUERY_SPELLER,
)
current_app.config[CONFIG_CHAT_APPROACH] = ChatReadRetrieveReadApproach(
search_client,
OPENAI_HOST,
AZURE_OPENAI_CHATGPT_DEPLOYMENT,
OPENAI_CHATGPT_MODEL,
AZURE_OPENAI_EMB_DEPLOYMENT,
OPENAI_EMB_MODEL,
KB_FIELDS_SOURCEPAGE,
KB_FIELDS_CONTENT,
AZURE_SEARCH_QUERY_LANGUAGE,
AZURE_SEARCH_QUERY_SPELLER,
)
===========unchanged ref 0===========
at: app.backend.app
CONFIG_OPENAI_TOKEN = "openai_token"
CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
CONFIG_SEARCH_CLIENT = "search_client"
bp = Blueprint("routes", __name__, static_folder="static")
at: approaches.chatreadretrieveread
ChatReadRetrieveReadApproach(search_client: SearchClient, openai_host: str, chatgpt_deployment: Optional[str], chatgpt_model: str, embedding_deployment: Optional[str], embedding_model: str, sourcepage_field: str, content_field: str, query_language: str, query_speller: str)
at: approaches.retrievethenread
RetrieveThenReadApproach(search_client: SearchClient, openai_host: str, chatgpt_deployment: Optional[str], chatgpt_model: str, embedding_deployment: Optional[str], embedding_model: str, sourcepage_field: str, content_field: str, query_language: str, query_speller: str)
at: core.authentication
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
at: openai
api_key = os.environ.get("OPENAI_API_KEY")
organization = os.environ.get("OPENAI_ORGANIZATION")
api_base = os.environ.get("OPENAI_API_BASE", "https://api.openai.com/v1")
api_type = os.environ.get("OPENAI_API_TYPE", "open_ai")
===========unchanged ref 1===========
api_version = os.environ.get(
"OPENAI_API_VERSION",
("2023-05-15" if api_type in ("azure", "azure_ad", "azuread") else None),
)
at: os
environ = _createenviron()
getenv(key: str, default: _T) -> Union[str, _T]
getenv(key: str) -> Optional[str]
|
app.backend.app/create_app
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
aa02563ff18ce4f5f0cca15eaa59eb6155672f8e
|
Upgrade OpenAI SDK to v1 (#1017)
|
<7>:<add> # This tracks HTTP requests made by httpx/openai:
<add> HTTPXClientInstrumentor().instrument()
|
# module: app.backend.app
def create_app():
<0> app = Quart(__name__)
<1> app.register_blueprint(bp)
<2>
<3> if os.getenv("APPLICATIONINSIGHTS_CONNECTION_STRING"):
<4> configure_azure_monitor()
<5> # This tracks HTTP requests made by aiohttp:
<6> AioHttpClientInstrumentor().instrument()
<7> # This middleware tracks app route requests:
<8> app.asgi_app = OpenTelemetryMiddleware(app.asgi_app) # type: ignore[method-assign]
<9>
<10> # Level should be one of https://docs.python.org/3/library/logging.html#logging-levels
<11> default_level = "INFO" # In development, log more verbosely
<12> if os.getenv("WEBSITE_HOSTNAME"): # In production, don't log as heavily
<13> default_level = "WARNING"
<14> logging.basicConfig(level=os.getenv("APP_LOG_LEVEL", default_level))
<15>
<16> if allowed_origin := os.getenv("ALLOWED_ORIGIN"):
<17> app.logger.info("CORS enabled for %s", allowed_origin)
<18> cors(app, allow_origin=allowed_origin, allow_methods=["GET", "POST"])
<19> return app
<20>
|
===========unchanged ref 0===========
at: app.backend.app
bp = Blueprint("routes", __name__, static_folder="static")
at: logging
basicConfig(*, filename: Optional[StrPath]=..., filemode: str=..., format: str=..., datefmt: Optional[str]=..., style: str=..., level: Optional[_Level]=..., stream: Optional[IO[str]]=..., handlers: Optional[Iterable[Handler]]=...) -> None
at: os
getenv(key: str, default: _T) -> Union[str, _T]
getenv(key: str) -> Optional[str]
===========changed ref 0===========
# module: app.backend.app
- @bp.before_request
- async def ensure_openai_token():
- if openai.api_type != "azure_ad":
- return
- openai_token = current_app.config[CONFIG_OPENAI_TOKEN]
- if openai_token.expires_on < time.time() + 60:
- openai_token = await current_app.config[CONFIG_CREDENTIAL].get_token(
- "https://cognitiveservices.azure.com/.default"
- )
- current_app.config[CONFIG_OPENAI_TOKEN] = openai_token
- openai.api_key = openai_token.token
-
===========changed ref 1===========
# module: app.backend.app
def error_response(error: Exception, route: str, status_code: int = 500):
logging.exception("Exception in %s: %s", route, error)
+ if isinstance(error, APIError) and error.code == "content_filter":
- if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
status_code = 400
return jsonify(error_dict(error)), status_code
===========changed ref 2===========
# module: app.backend.app
def error_dict(error: Exception) -> dict:
+ if isinstance(error, APIError) and error.code == "content_filter":
- if isinstance(error, openai.error.InvalidRequestError) and error.code == "content_filter":
return {"error": ERROR_MESSAGE_FILTER}
return {"error": ERROR_MESSAGE.format(error_type=type(error))}
===========changed ref 3===========
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
if not request.is_json:
return jsonify({"error": "request must be json"}), 415
request_json = await request.get_json()
context = request_json.get("context", {})
auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
try:
approach = current_app.config[CONFIG_ASK_APPROACH]
- # Workaround for: https://github.com/openai/openai-python/issues/371
- async with aiohttp.ClientSession() as s:
- openai.aiosession.set(s)
+ r = await approach.run(
- r = await approach.run(
+ request_json["messages"], context=context, session_state=request_json.get("session_state")
- request_json["messages"], context=context, session_state=request_json.get("session_state")
+ )
- )
return jsonify(r)
except Exception as error:
return error_response(error, "/ask")
===========changed ref 4===========
# module: app.backend.app
- CONFIG_OPENAI_TOKEN = "openai_token"
- CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
CONFIG_SEARCH_CLIENT = "search_client"
+ CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddings(ABC):
- def create_embedding_arguments(self) -> dict[str, Any]:
- raise NotImplementedError
-
===========changed ref 7===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 8===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ return self.create_embedding_response
+
===========changed ref 9===========
# module: tests.test_searchmanager
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 10===========
# module: tests.test_prepdocs
+ class MockClient:
+ def __init__(self, embeddings_client):
+ self.embeddings = embeddings_client
+
===========changed ref 11===========
# module: tests.test_searchmanager
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 12===========
# module: tests.test_prepdocs
+ class MockEmbeddingsClient:
+ def __init__(self, create_embedding_response: openai.types.CreateEmbeddingResponse):
+ self.create_embedding_response = create_embedding_response
+
===========changed ref 13===========
# module: tests.test_prepdocs
+ def create_rate_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=RateLimitMockEmbeddingsClient())
+
===========changed ref 14===========
# module: tests.test_prepdocs
+ def create_auth_error_limit_client(*args, **kwargs):
+ return MockClient(embeddings_client=AuthenticationErrorMockEmbeddingsClient())
+
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class AzureOpenAIEmbeddingService(OpenAIEmbeddings):
- def get_api_type(self) -> str:
- return "azure_ad" if isinstance(self.credential, AsyncTokenCredential) else "azure"
-
===========changed ref 17===========
# module: tests.test_app
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 18===========
# module: tests.test_prepdocs
+ def fake_response(http_code):
+ return Response(http_code, request=Request(method="get", url="https://foo.bar/"))
+
===========changed ref 19===========
# module: tests.test_prepdocs
+ class AuthenticationErrorMockEmbeddingsClient:
+ def create(self, *args, **kwargs) -> openai.types.CreateEmbeddingResponse:
+ raise openai.AuthenticationError(message="Bad things happened.", response=fake_response(403), body=None)
+
|
scripts.prepdocslib.textsplitter/TextSplitter.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<6>:<add> self.has_image_embeddings = has_image_embeddings
|
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
+ def __init__(self, has_image_embeddings, verbose: bool = False):
- def __init__(self, verbose: bool = False):
<0> self.sentence_endings = [".", "!", "?"]
<1> self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
<2> self.max_section_length = 1000
<3> self.sentence_search_limit = 100
<4> self.section_overlap = 100
<5> self.verbose = verbose
<6>
| |
scripts.prepdocslib.textsplitter/TextSplitter.split_pages
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<0>:<add> # Chunking is disabled when using GPT4V. To be updated in the future.
<add> if self.has_image_embeddings:
<add> for i, page in enumerate(pages):
<add> yield SplitPage(page_num=i, text=page.text)
<add>
|
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
def split_pages(self, pages: List[Page]) -> Generator[SplitPage, None, None]:
<0> def find_page(offset):
<1> num_pages = len(pages)
<2> for i in range(num_pages - 1):
<3> if offset >= pages[i].offset and offset < pages[i + 1].offset:
<4> return pages[i].page_num
<5> return pages[num_pages - 1].page_num
<6>
<7> all_text = "".join(page.text for page in pages)
<8> length = len(all_text)
<9> start = 0
<10> end = length
<11> while start + self.section_overlap < length:
<12> last_word = -1
<13> end = start + self.max_section_length
<14>
<15> if end > length:
<16> end = length
<17> else:
<18> # Try to find the end of the sentence
<19> while (
<20> end < length
<21> and (end - start - self.max_section_length) < self.sentence_search_limit
<22> and all_text[end] not in self.sentence_endings
<23> ):
<24> if all_text[end] in self.word_breaks:
<25> last_word = end
<26> end += 1
<27> if end < length and all_text[end] not in self.sentence_endings and last_word > 0:
<28> end = last_word # Fall back to at least keeping a whole word
<29> if end < length:
<30> end += 1
<31>
<32> # Try to find the start of the sentence or at least a whole word boundary
<33> last_word = -1
<34> while (
<35> start > 0
<36> and start > end - self.max_section_length - 2 * self.sentence_search_limit
<37> and all_text[start] not in self.sentence_endings
<38> ):
<39> if all_text[start] in self.word_breaks:
<40> last_word = start
<41> start -= 1</s>
|
===========below chunk 0===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
def split_pages(self, pages: List[Page]) -> Generator[SplitPage, None, None]:
# offset: 1
start = last_word
if start > 0:
start += 1
section_text = all_text[start:end]
yield SplitPage(page_num=find_page(start), text=section_text)
last_table_start = section_text.rfind("<table")
if last_table_start > 2 * self.sentence_search_limit and last_table_start > section_text.rfind("</table"):
# If the section ends with an unclosed table, we need to start the next section with the table.
# If table starts inside sentence_search_limit, we ignore it, as that will cause an infinite loop for tables longer than MAX_SECTION_LENGTH
# If last table starts inside section_overlap, keep overlapping
if self.verbose:
print(
f"Section ends with unclosed table, starting next section with the table at page {find_page(start)} offset {start} table start {last_table_start}"
)
start = min(end - self.section_overlap, start + last_table_start)
else:
start = end - self.section_overlap
if start + self.section_overlap < end:
yield SplitPage(page_num=find_page(start), text=all_text[start:end])
===========unchanged ref 0===========
at: scripts.prepdocslib.pdfparser
Page(page_num: int, offset: int, text: str)
at: scripts.prepdocslib.pdfparser.Page.__init__
self.page_num = page_num
self.offset = offset
self.text = text
at: scripts.prepdocslib.textsplitter
SplitPage(page_num: int, text: str)
at: scripts.prepdocslib.textsplitter.TextSplitter.__init__
self.sentence_endings = [".", "!", "?"]
self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
self.max_section_length = 1000
self.sentence_search_limit = 100
self.section_overlap = 100
self.has_image_embeddings = has_image_embeddings
at: typing
List = _alias(list, 1, inst=False, name='List')
Generator = _alias(collections.abc.Generator, 3)
===========changed ref 0===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
+ def __init__(self, has_image_embeddings, verbose: bool = False):
- def __init__(self, verbose: bool = False):
self.sentence_endings = [".", "!", "?"]
self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
self.max_section_length = 1000
self.sentence_search_limit = 100
self.section_overlap = 100
self.verbose = verbose
+ self.has_image_embeddings = has_image_embeddings
|
scripts.prepdocslib.blobmanager/BlobManager.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<3>:<add> self.store_page_images = store_page_images
<4>:<add> self.user_delegation_key: Optional[UserDelegationKey] = None
|
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def __init__(
self,
endpoint: str,
container: str,
credential: Union[AsyncTokenCredential, str],
+ store_page_images: bool = False,
verbose: bool = False,
):
<0> self.endpoint = endpoint
<1> self.credential = credential
<2> self.container = container
<3> self.verbose = verbose
<4>
|
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ """
+ Class for using image embeddings from Azure AI Vision
+ To learn more, please visit https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api
+ """
+
===========changed ref 5===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
+ def __init__(self, has_image_embeddings, verbose: bool = False):
- def __init__(self, verbose: bool = False):
self.sentence_endings = [".", "!", "?"]
self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
self.max_section_length = 1000
self.sentence_search_limit = 100
self.section_overlap = 100
self.verbose = verbose
+ self.has_image_embeddings = has_image_embeddings
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def create_embeddings(self, blob_urls: List[str]) -> List[List[float]]:
+ headers = {"Ocp-Apim-Subscription-Key": self.credential}
+ params = {"api-version": "2023-02-01-preview", "modelVersion": "latest"}
+ endpoint = urljoin(self.endpoint, "computervision/retrieval:vectorizeImage")
+ embeddings: List[List[float]] = []
+ async with aiohttp.ClientSession(headers=headers) as session:
+ for blob_url in blob_urls:
+ async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(Exception),
+ wait=wait_random_exponential(min=15, max=60),
+ stop=stop_after_attempt(15),
+ before_sleep=self.before_retry_sleep,
+ ):
+ with attempt:
+ body = {"url": blob_url}
+ async with session.post(url=endpoint, params=params, json=body) as resp:
+ resp_json = await resp.json()
+ embeddings.append(resp_json["vector"])
+
+ return embeddings
+
===========changed ref 7===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
def split_pages(self, pages: List[Page]) -> Generator[SplitPage, None, None]:
+ # Chunking is disabled when using GPT4V. To be updated in the future.
+ if self.has_image_embeddings:
+ for i, page in enumerate(pages):
+ yield SplitPage(page_num=i, text=page.text)
+
def find_page(offset):
num_pages = len(pages)
for i in range(num_pages - 1):
if offset >= pages[i].offset and offset < pages[i + 1].offset:
return pages[i].page_num
return pages[num_pages - 1].page_num
all_text = "".join(page.text for page in pages)
length = len(all_text)
start = 0
end = length
while start + self.section_overlap < length:
last_word = -1
end = start + self.max_section_length
if end > length:
end = length
else:
# Try to find the end of the sentence
while (
end < length
and (end - start - self.max_section_length) < self.sentence_search_limit
and all_text[end] not in self.sentence_endings
):
if all_text[end] in self.word_breaks:
last_word = end
end += 1
if end < length and all_text[end] not in self.sentence_endings and last_word > 0:
end = last_word # Fall back to at least keeping a whole word
if end < length:
end += 1
# Try to find the start of the sentence or at least a whole word boundary
last_word = -1
while (
start > 0
and start > end - self.max_section_length - 2 * self.sentence_search_limit
and all_text[start] not in self.sentence_endings
):
</s>
===========changed ref 8===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
def split_pages(self, pages: List[Page]) -> Generator[SplitPage, None, None]:
# offset: 1
<s> self.sentence_search_limit
and all_text[start] not in self.sentence_endings
):
if all_text[start] in self.word_breaks:
last_word = start
start -= 1
if all_text[start] not in self.sentence_endings and last_word > 0:
start = last_word
if start > 0:
start += 1
section_text = all_text[start:end]
yield SplitPage(page_num=find_page(start), text=section_text)
last_table_start = section_text.rfind("<table")
if last_table_start > 2 * self.sentence_search_limit and last_table_start > section_text.rfind("</table"):
# If the section ends with an unclosed table, we need to start the next section with the table.
# If table starts inside sentence_search_limit, we ignore it, as that will cause an infinite loop for tables longer than MAX_SECTION_LENGTH
# If last table starts inside section_overlap, keep overlapping
if self.verbose:
print(
f"Section ends with unclosed table, starting next section with the table at page {find_page(start)} offset {start} table start {last_table_start}"
)
start = min(end - self.section_overlap, start + last_table_start)
else:
start = end - self.section_overlap
if start + self.section_overlap < end:
yield SplitPage(page_num=find_page(start), text=all_text[start:end])
|
scripts.prepdocslib.blobmanager/BlobManager.remove_blob
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<13>:<add> if (
<add> prefix is not None
<add> and (
<add> not re.match(rf"{prefix}-\d+\.pdf", blob_path) or not re.match(rf"{prefix}-\d+\.png", blob_path)
<add> )
<del> if (prefix is not None and not re.match(rf"{prefix}-\d+\.pdf", blob_path)) or (
<14>:<add> ) or (path is not None and blob_path == os.path.basename(path)):
<del> path is not None and blob_path == os.path.basename(path)
<15>:<del> ):
|
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def remove_blob(self, path: Optional[str] = None):
<0> async with BlobServiceClient(
<1> account_url=self.endpoint, credential=self.credential
<2> ) as service_client, service_client.get_container_client(self.container) as container_client:
<3> if not await container_client.exists():
<4> return
<5> if path is None:
<6> prefix = None
<7> blobs = container_client.list_blob_names()
<8> else:
<9> prefix = os.path.splitext(os.path.basename(path))[0]
<10> blobs = container_client.list_blob_names(name_starts_with=os.path.splitext(os.path.basename(prefix))[0])
<11> async for blob_path in blobs:
<12> # This still supports PDFs split into individual pages, but we could remove in future to simplify code
<13> if (prefix is not None and not re.match(rf"{prefix}-\d+\.pdf", blob_path)) or (
<14> path is not None and blob_path == os.path.basename(path)
<15> ):
<16> continue
<17> if self.verbose:
<18> print(f"\tRemoving blob {blob_path}")
<19> await container_client.delete_blob(blob_path)
<20>
|
===========unchanged ref 0===========
at: os.path
splitext(p: AnyStr) -> Tuple[AnyStr, AnyStr]
splitext(p: _PathLike[AnyStr]) -> Tuple[AnyStr, AnyStr]
at: scripts.prepdocslib.blobmanager
BlobManager(endpoint: str, container: str, credential: Union[AsyncTokenCredential, str], store_page_images: bool=False, verbose: bool=False)
at: scripts.prepdocslib.blobmanager.BlobManager
blob_name_from_file_name(filename) -> str
at: scripts.prepdocslib.blobmanager.BlobManager.__init__
self.store_page_images = store_page_images
at: scripts.prepdocslib.listfilestrategy
File(content: IO, acls: Optional[dict[str, list]]=None)
at: scripts.prepdocslib.listfilestrategy.File.__init__
self.content = content
at: typing
List = _alias(list, 1, inst=False, name='List')
at: typing.IO
__slots__ = ()
===========changed ref 0===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def __init__(
self,
endpoint: str,
container: str,
credential: Union[AsyncTokenCredential, str],
+ store_page_images: bool = False,
verbose: bool = False,
):
self.endpoint = endpoint
self.credential = credential
self.container = container
+ self.store_page_images = store_page_images
self.verbose = verbose
+ self.user_delegation_key: Optional[UserDelegationKey] = None
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ """
+ Class for using image embeddings from Azure AI Vision
+ To learn more, please visit https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api
+ """
+
===========changed ref 6===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
+ def __init__(self, has_image_embeddings, verbose: bool = False):
- def __init__(self, verbose: bool = False):
self.sentence_endings = [".", "!", "?"]
self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
self.max_section_length = 1000
self.sentence_search_limit = 100
self.section_overlap = 100
self.verbose = verbose
+ self.has_image_embeddings = has_image_embeddings
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def create_embeddings(self, blob_urls: List[str]) -> List[List[float]]:
+ headers = {"Ocp-Apim-Subscription-Key": self.credential}
+ params = {"api-version": "2023-02-01-preview", "modelVersion": "latest"}
+ endpoint = urljoin(self.endpoint, "computervision/retrieval:vectorizeImage")
+ embeddings: List[List[float]] = []
+ async with aiohttp.ClientSession(headers=headers) as session:
+ for blob_url in blob_urls:
+ async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(Exception),
+ wait=wait_random_exponential(min=15, max=60),
+ stop=stop_after_attempt(15),
+ before_sleep=self.before_retry_sleep,
+ ):
+ with attempt:
+ body = {"url": blob_url}
+ async with session.post(url=endpoint, params=params, json=body) as resp:
+ resp_json = await resp.json()
+ embeddings.append(resp_json["vector"])
+
+ return embeddings
+
===========changed ref 8===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
def split_pages(self, pages: List[Page]) -> Generator[SplitPage, None, None]:
+ # Chunking is disabled when using GPT4V. To be updated in the future.
+ if self.has_image_embeddings:
+ for i, page in enumerate(pages):
+ yield SplitPage(page_num=i, text=page.text)
+
def find_page(offset):
num_pages = len(pages)
for i in range(num_pages - 1):
if offset >= pages[i].offset and offset < pages[i + 1].offset:
return pages[i].page_num
return pages[num_pages - 1].page_num
all_text = "".join(page.text for page in pages)
length = len(all_text)
start = 0
end = length
while start + self.section_overlap < length:
last_word = -1
end = start + self.max_section_length
if end > length:
end = length
else:
# Try to find the end of the sentence
while (
end < length
and (end - start - self.max_section_length) < self.sentence_search_limit
and all_text[end] not in self.sentence_endings
):
if all_text[end] in self.word_breaks:
last_word = end
end += 1
if end < length and all_text[end] not in self.sentence_endings and last_word > 0:
end = last_word # Fall back to at least keeping a whole word
if end < length:
end += 1
# Try to find the start of the sentence or at least a whole word boundary
last_word = -1
while (
start > 0
and start > end - self.max_section_length - 2 * self.sentence_search_limit
and all_text[start] not in self.sentence_endings
):
</s>
|
scripts.prepdocslib.searchmanager/SearchManager.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<4>:<add> self.search_images = search_images
|
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def __init__(
self,
search_info: SearchInfo,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
embeddings: Optional[OpenAIEmbeddings] = None,
+ search_images: bool = False,
):
<0> self.search_info = search_info
<1> self.search_analyzer_name = search_analyzer_name
<2> self.use_acls = use_acls
<3> self.embeddings = embeddings
<4>
|
===========unchanged ref 0===========
at: scripts.prepdocslib.embeddings
OpenAIEmbeddings(open_ai_model_name: str, disable_batch: bool=False, verbose: bool=False)
at: scripts.prepdocslib.strategy
SearchInfo(endpoint: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], index_name: str, verbose: bool=False)
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 3===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 4===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 5===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def __init__(
self,
endpoint: str,
container: str,
credential: Union[AsyncTokenCredential, str],
+ store_page_images: bool = False,
verbose: bool = False,
):
self.endpoint = endpoint
self.credential = credential
self.container = container
+ self.store_page_images = store_page_images
self.verbose = verbose
+ self.user_delegation_key: Optional[UserDelegationKey] = None
===========changed ref 6===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ """
+ Class for using image embeddings from Azure AI Vision
+ To learn more, please visit https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api
+ """
+
===========changed ref 7===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
+ def __init__(self, has_image_embeddings, verbose: bool = False):
- def __init__(self, verbose: bool = False):
self.sentence_endings = [".", "!", "?"]
self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
self.max_section_length = 1000
self.sentence_search_limit = 100
self.section_overlap = 100
self.verbose = verbose
+ self.has_image_embeddings = has_image_embeddings
===========changed ref 8===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def create_embeddings(self, blob_urls: List[str]) -> List[List[float]]:
+ headers = {"Ocp-Apim-Subscription-Key": self.credential}
+ params = {"api-version": "2023-02-01-preview", "modelVersion": "latest"}
+ endpoint = urljoin(self.endpoint, "computervision/retrieval:vectorizeImage")
+ embeddings: List[List[float]] = []
+ async with aiohttp.ClientSession(headers=headers) as session:
+ for blob_url in blob_urls:
+ async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(Exception),
+ wait=wait_random_exponential(min=15, max=60),
+ stop=stop_after_attempt(15),
+ before_sleep=self.before_retry_sleep,
+ ):
+ with attempt:
+ body = {"url": blob_url}
+ async with session.post(url=endpoint, params=params, json=body) as resp:
+ resp_json = await resp.json()
+ embeddings.append(resp_json["vector"])
+
+ return embeddings
+
===========changed ref 9===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def remove_blob(self, path: Optional[str] = None):
async with BlobServiceClient(
account_url=self.endpoint, credential=self.credential
) as service_client, service_client.get_container_client(self.container) as container_client:
if not await container_client.exists():
return
if path is None:
prefix = None
blobs = container_client.list_blob_names()
else:
prefix = os.path.splitext(os.path.basename(path))[0]
blobs = container_client.list_blob_names(name_starts_with=os.path.splitext(os.path.basename(prefix))[0])
async for blob_path in blobs:
# This still supports PDFs split into individual pages, but we could remove in future to simplify code
+ if (
+ prefix is not None
+ and (
+ not re.match(rf"{prefix}-\d+\.pdf", blob_path) or not re.match(rf"{prefix}-\d+\.png", blob_path)
+ )
- if (prefix is not None and not re.match(rf"{prefix}-\d+\.pdf", blob_path)) or (
+ ) or (path is not None and blob_path == os.path.basename(path)):
- path is not None and blob_path == os.path.basename(path)
- ):
continue
if self.verbose:
print(f"\tRemoving blob {blob_path}")
await container_client.delete_blob(blob_path)
|
scripts.prepdocslib.searchmanager/SearchManager.create_index
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<32>:<add> )
<add> if self.search_images:
<add> fields.append(
<add> SearchField(
<add> name="imageEmbedding",
<add> type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
<add> hidden=False,
<add> searchable=True,
<add> filterable=False,
<add> sortable=False,
<add> facetable=False,
<add> vector_search_dimensions=1024,
<add> vector_search_profile="embedding_config",
<add> ),
|
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def create_index(self):
<0> if self.search_info.verbose:
<1> print(f"Ensuring search index {self.search_info.index_name} exists")
<2>
<3> async with self.search_info.create_search_index_client() as search_index_client:
<4> fields = [
<5> SimpleField(name="id", type="Edm.String", key=True),
<6> SearchableField(name="content", type="Edm.String", analyzer_name=self.search_analyzer_name),
<7> SearchField(
<8> name="embedding",
<9> type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
<10> hidden=False,
<11> searchable=True,
<12> filterable=False,
<13> sortable=False,
<14> facetable=False,
<15> vector_search_dimensions=1536,
<16> vector_search_profile="embedding_config",
<17> ),
<18> SimpleField(name="category", type="Edm.String", filterable=True, facetable=True),
<19> SimpleField(name="sourcepage", type="Edm.String", filterable=True, facetable=True),
<20> SimpleField(name="sourcefile", type="Edm.String", filterable=True, facetable=True),
<21> ]
<22> if self.use_acls:
<23> fields.append(
<24> SimpleField(
<25> name="oids", type=SearchFieldDataType.Collection(SearchFieldDataType.String), filterable=True
<26> )
<27> )
<28> fields.append(
<29> SimpleField(
<30> name="groups", type=SearchFieldDataType.Collection(SearchFieldDataType.String), filterable=True
<31> )
<32> )
<33>
<34> index = SearchIndex(
<35> name=self.search_info.index_name,
<36> fields=fields,
<37> semantic_settings=SemanticSettings(
<38> configurations=[
<39> SemanticConfiguration(
<40> name="default",
<41> prioritized_fields=PrioritizedFields(
<42> title</s>
|
===========below chunk 0===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def create_index(self):
# offset: 1
),
)
]
),
vector_search=VectorSearch(
algorithms=[
HnswVectorSearchAlgorithmConfiguration(
name="hnsw_config",
kind=VectorSearchAlgorithmKind.HNSW,
parameters=HnswParameters(metric="cosine"),
)
],
profiles=[
VectorSearchProfile(
name="embedding_config",
algorithm="hnsw_config",
),
],
),
)
if self.search_info.index_name not in [name async for name in search_index_client.list_index_names()]:
if self.search_info.verbose:
print(f"Creating {self.search_info.index_name} search index")
await search_index_client.create_index(index)
else:
if self.search_info.verbose:
print(f"Search index {self.search_info.index_name} already exists")
===========unchanged ref 0===========
at: scripts.prepdocslib.searchmanager.SearchManager.__init__
self.search_info = search_info
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
at: scripts.prepdocslib.strategy.SearchInfo
create_search_index_client() -> SearchIndexClient
at: scripts.prepdocslib.strategy.SearchInfo.__init__
self.index_name = index_name
self.verbose = verbose
===========changed ref 0===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def __init__(
self,
search_info: SearchInfo,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
embeddings: Optional[OpenAIEmbeddings] = None,
+ search_images: bool = False,
):
self.search_info = search_info
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.embeddings = embeddings
+ self.search_images = search_images
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 2===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 3===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 4===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 5===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 6===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def __init__(
self,
endpoint: str,
container: str,
credential: Union[AsyncTokenCredential, str],
+ store_page_images: bool = False,
verbose: bool = False,
):
self.endpoint = endpoint
self.credential = credential
self.container = container
+ self.store_page_images = store_page_images
self.verbose = verbose
+ self.user_delegation_key: Optional[UserDelegationKey] = None
===========changed ref 7===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ """
+ Class for using image embeddings from Azure AI Vision
+ To learn more, please visit https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api
+ """
+
===========changed ref 8===========
# module: scripts.prepdocslib.textsplitter
class TextSplitter:
+ def __init__(self, has_image_embeddings, verbose: bool = False):
- def __init__(self, verbose: bool = False):
self.sentence_endings = [".", "!", "?"]
self.word_breaks = [",", ";", ":", " ", "(", ")", "[", "]", "{", "}", "\t", "\n"]
self.max_section_length = 1000
self.sentence_search_limit = 100
self.section_overlap = 100
self.verbose = verbose
+ self.has_image_embeddings = has_image_embeddings
===========changed ref 9===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def create_embeddings(self, blob_urls: List[str]) -> List[List[float]]:
+ headers = {"Ocp-Apim-Subscription-Key": self.credential}
+ params = {"api-version": "2023-02-01-preview", "modelVersion": "latest"}
+ endpoint = urljoin(self.endpoint, "computervision/retrieval:vectorizeImage")
+ embeddings: List[List[float]] = []
+ async with aiohttp.ClientSession(headers=headers) as session:
+ for blob_url in blob_urls:
+ async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(Exception),
+ wait=wait_random_exponential(min=15, max=60),
+ stop=stop_after_attempt(15),
+ before_sleep=self.before_retry_sleep,
+ ):
+ with attempt:
+ body = {"url": blob_url}
+ async with session.post(url=endpoint, params=params, json=body) as resp:
+ resp_json = await resp.json()
+ embeddings.append(resp_json["vector"])
+
+ return embeddings
+
|
scripts.prepdocslib.searchmanager/SearchManager.update_content
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<10>:<add> "sourcepage": BlobManager.blob_image_name_from_file_page(
<add> filename=section.content.filename(), page=section.split_page.page_num
<add> )
<add> if image_embeddings
<add> else BlobManager.sourcepage_from_file_page(
<del> "sourcepage": BlobManager.sourcepage_from_file_page(
<24>:<add> if image_embeddings:
<add> for i, (document, section) in enumerate(zip(documents, batch)):
<add> document["imageEmbedding"] = image_embeddings[section.split_page.page_num]
|
# module: scripts.prepdocslib.searchmanager
class SearchManager:
+ def update_content(self, sections: List[Section], image_embeddings: Optional[List[List[float]]] = None):
- def update_content(self, sections: List[Section]):
<0> MAX_BATCH_SIZE = 1000
<1> section_batches = [sections[i : i + MAX_BATCH_SIZE] for i in range(0, len(sections), MAX_BATCH_SIZE)]
<2>
<3> async with self.search_info.create_search_client() as search_client:
<4> for batch_index, batch in enumerate(section_batches):
<5> documents = [
<6> {
<7> "id": f"{section.content.filename_to_id()}-page-{section_index + batch_index * MAX_BATCH_SIZE}",
<8> "content": section.split_page.text,
<9> "category": section.category,
<10> "sourcepage": BlobManager.sourcepage_from_file_page(
<11> filename=section.content.filename(), page=section.split_page.page_num
<12> ),
<13> "sourcefile": section.content.filename(),
<14> **section.content.acls,
<15> }
<16> for section_index, section in enumerate(batch)
<17> ]
<18> if self.embeddings:
<19> embeddings = await self.embeddings.create_embeddings(
<20> texts=[section.split_page.text for section in batch]
<21> )
<22> for i, document in enumerate(documents):
<23> document["embedding"] = embeddings[i]
<24>
<25> await search_client.upload_documents(documents)
<26>
|
===========unchanged ref 0===========
at: scripts.prepdocslib.listfilestrategy.File
filename_to_id()
at: scripts.prepdocslib.searchmanager
Section(split_page: SplitPage, content: File, category: Optional[str]=None)
documents = [
{
"id": f"{section.content.filename_to_id()}-page-{section_index + batch_index * MAX_BATCH_SIZE}",
"content": section.split_page.text,
"category": section.category,
"sourcepage": BlobManager.blob_image_name_from_file_page(
filename=section.content.filename(), page=section.split_page.page_num
)
if image_embeddings
else BlobManager.sourcepage_from_file_page(
filename=section.content.filename(), page=section.split_page.page_num
),
"sourcefile": section.content.filename(),
**section.content.acls,
}
for section_index, section in enumerate(batch)
]
documents = [
{
"id": f"{section.content.filename_to_id()}-page-{section_index + batch_index * MAX_BATCH_SIZE}",
"content": section.split_page.text,
"category": section.category,
"sourcepage": BlobManager.blob_image_name_from_file_page(
filename=section.content.filename(), page=section.split_page.page_num
)
if image_embeddings
else BlobManager.sourcepage_from_file_page(
filename=section.content.filename(), page=section.split_page.page_num
),
"sourcefile": section.content.filename(),
**section.content.acls,
}
for section_index, section in enumerate(batch)
]
at: scripts.prepdocslib.searchmanager.SearchManager.__init__
self.search_info = search_info
===========unchanged ref 1===========
at: scripts.prepdocslib.searchmanager.SearchManager.create_index
index = SearchIndex(
name=self.search_info.index_name,
fields=fields,
semantic_settings=SemanticSettings(
configurations=[
SemanticConfiguration(
name="default",
prioritized_fields=PrioritizedFields(
title_field=None, prioritized_content_fields=[SemanticField(field_name="content")]
),
)
]
),
vector_search=VectorSearch(
algorithms=[
HnswVectorSearchAlgorithmConfiguration(
name="hnsw_config",
kind=VectorSearchAlgorithmKind.HNSW,
parameters=HnswParameters(metric="cosine"),
)
],
profiles=[
VectorSearchProfile(
name="embedding_config",
algorithm="hnsw_config",
),
],
),
)
at: scripts.prepdocslib.searchmanager.Section.__init__
self.split_page = split_page
self.content = content
self.category = category
at: scripts.prepdocslib.strategy.SearchInfo
create_search_client() -> SearchClient
at: scripts.prepdocslib.strategy.SearchInfo.__init__
self.index_name = index_name
self.verbose = verbose
at: scripts.prepdocslib.textsplitter.SplitPage.__init__
self.text = text
at: typing
List = _alias(list, 1, inst=False, name='List')
===========changed ref 0===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def __init__(
self,
search_info: SearchInfo,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
embeddings: Optional[OpenAIEmbeddings] = None,
+ search_images: bool = False,
):
self.search_info = search_info
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.embeddings = embeddings
+ self.search_images = search_images
===========changed ref 1===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def create_index(self):
if self.search_info.verbose:
print(f"Ensuring search index {self.search_info.index_name} exists")
async with self.search_info.create_search_index_client() as search_index_client:
fields = [
SimpleField(name="id", type="Edm.String", key=True),
SearchableField(name="content", type="Edm.String", analyzer_name=self.search_analyzer_name),
SearchField(
name="embedding",
type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
hidden=False,
searchable=True,
filterable=False,
sortable=False,
facetable=False,
vector_search_dimensions=1536,
vector_search_profile="embedding_config",
),
SimpleField(name="category", type="Edm.String", filterable=True, facetable=True),
SimpleField(name="sourcepage", type="Edm.String", filterable=True, facetable=True),
SimpleField(name="sourcefile", type="Edm.String", filterable=True, facetable=True),
]
if self.use_acls:
fields.append(
SimpleField(
name="oids", type=SearchFieldDataType.Collection(SearchFieldDataType.String), filterable=True
)
)
fields.append(
SimpleField(
name="groups", type=SearchFieldDataType.Collection(SearchFieldDataType.String), filterable=True
)
+ )
+ if self.search_images:
+ fields.append(
+ SearchField(
+ name="imageEmbedding",
+ type=SearchFieldDataType.Collection(SearchFieldDataType.Single),
+ hidden=False,
+ searchable=True,
+ filterable=False,
+ sortable=False,
+ facetable=False,
+ vector_search_dimensions=1024,
+ vector_search_profile="embedding_config</s>
===========changed ref 2===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def create_index(self):
# offset: 1
<s> facetable=False,
+ vector_search_dimensions=1024,
+ vector_search_profile="embedding_config",
+ ),
)
index = SearchIndex(
name=self.search_info.index_name,
fields=fields,
semantic_settings=SemanticSettings(
configurations=[
SemanticConfiguration(
name="default",
prioritized_fields=PrioritizedFields(
title_field=None, prioritized_content_fields=[SemanticField(field_name="content")]
),
)
]
),
vector_search=VectorSearch(
algorithms=[
HnswVectorSearchAlgorithmConfiguration(
name="hnsw_config",
kind=VectorSearchAlgorithmKind.HNSW,
parameters=HnswParameters(metric="cosine"),
)
],
profiles=[
VectorSearchProfile(
name="embedding_config",
algorithm="hnsw_config",
),
],
),
)
if self.search_info.index_name not in [name async for name in search_index_client.list_index_names()]:
if self.search_info.verbose:
print(f"Creating {self.search_info.index_name} search index")
await search_index_client.create_index(index)
else:
if self.search_info.verbose:
print(f"Search index {self.search_info.index_name} already exists")
|
tests.test_blob_manager/test_upload_and_remove
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<2>:<add> filename = os.path.basename(f.content.name)
<del> filename = f.content.name.split("/tmp/")[1]
|
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove(monkeypatch, mock_env, blob_manager):
<0> with NamedTemporaryFile(suffix=".pdf") as temp_file:
<1> f = File(temp_file.file)
<2> filename = f.content.name.split("/tmp/")[1]
<3>
<4> # Set up mocks used by upload_blob
<5> async def mock_exists(*args, **kwargs):
<6> return True
<7>
<8> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.exists", mock_exists)
<9>
<10> async def mock_upload_blob(self, name, *args, **kwargs):
<11> assert name == filename
<12> return True
<13>
<14> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.upload_blob", mock_upload_blob)
<15>
<16> await blob_manager.upload_blob(f)
<17>
<18> # Set up mocks used by remove_blob
<19> def mock_list_blob_names(*args, **kwargs):
<20> assert kwargs.get("name_starts_with") == filename.split(".pdf")[0]
<21>
<22> class AsyncBlobItemsIterator:
<23> def __init__(self, file):
<24> self.files = [file, "dontdelete.pdf"]
<25>
<26> def __aiter__(self):
<27> return self
<28>
<29> async def __anext__(self):
<30> if self.files:
<31> return self.files.pop()
<32> raise StopAsyncIteration
<33>
<34> return AsyncBlobItemsIterator(filename)
<35>
<36> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.list_blob_names", mock_list_blob_names)
<37>
<38> async def mock_delete_blob(self, name, *args, **kwargs):
<39> assert name == filename
<40> return True
<41>
<42> </s>
|
===========below chunk 0===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove(monkeypatch, mock_env, blob_manager):
# offset: 1
await blob_manager.remove_blob(f.content.name)
===========unchanged ref 0===========
at: _pytest.mark.structures
MARK_GEN = MarkGenerator(_ispytest=True)
at: _pytest.mark.structures.MarkGenerator
skip: _SkipMarkDecorator
skipif: _SkipifMarkDecorator
xfail: _XfailMarkDecorator
parametrize: _ParametrizeMarkDecorator
usefixtures: _UsefixturesMarkDecorator
filterwarnings: _FilterwarningsMarkDecorator
at: _pytest.monkeypatch
monkeypatch() -> Generator["MonkeyPatch", None, None]
at: os.path
basename(p: _PathLike[AnyStr]) -> AnyStr
basename(p: AnyStr) -> AnyStr
at: scripts.prepdocslib.listfilestrategy
File(content: IO, acls: Optional[dict[str, list]]=None)
at: scripts.prepdocslib.listfilestrategy.File.__init__
self.content = content
at: sys
version_info: _version_info
at: sys._version_info
major: int
minor: int
micro: int
releaselevel: str
serial: int
===========unchanged ref 1===========
at: tempfile
NamedTemporaryFile(mode: str=..., buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[Any]
NamedTemporaryFile(mode: Literal["r", "w", "a", "x", "r+", "w+", "a+", "x+", "rt", "wt", "at", "xt", "r+t", "w+t", "a+t", "x+t"], buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[str]
NamedTemporaryFile(mode: Literal["rb", "wb", "ab", "xb", "r+b", "w+b", "a+b", "x+b"]=..., buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[bytes]
at: typing.IO
__slots__ = ()
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
===========changed ref 0===========
+ # module: tests.mocks
+
+
===========changed ref 1===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 2===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 3===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 4===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 5===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 6===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 7===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 8===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 9===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 10===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 14===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 15===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 20===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockBlob:
+ def __init__(self):
+ self.properties = BlobProperties(
+ name="Financial Market Analysis Report 2023-7.png", content_settings={"content_type": "image/png"}
+ )
+
|
tests.test_blob_manager/test_upload_and_remove_all
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<3>:<add> filename = os.path.basename(f.content.name)
<del> filename = f.content.name.split("/tmp/")[1]
|
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove_all(monkeypatch, mock_env, blob_manager):
<0> with NamedTemporaryFile(suffix=".pdf") as temp_file:
<1> f = File(temp_file.file)
<2> print(f.content.name)
<3> filename = f.content.name.split("/tmp/")[1]
<4>
<5> # Set up mocks used by upload_blob
<6> async def mock_exists(*args, **kwargs):
<7> return True
<8>
<9> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.exists", mock_exists)
<10>
<11> async def mock_upload_blob(self, name, *args, **kwargs):
<12> assert name == filename
<13> return True
<14>
<15> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.upload_blob", mock_upload_blob)
<16>
<17> await blob_manager.upload_blob(f)
<18>
<19> # Set up mocks used by remove_blob
<20> def mock_list_blob_names(*args, **kwargs):
<21> assert kwargs.get("name_starts_with") is None
<22>
<23> class AsyncBlobItemsIterator:
<24> def __init__(self, file):
<25> self.files = [file]
<26>
<27> def __aiter__(self):
<28> return self
<29>
<30> async def __anext__(self):
<31> if self.files:
<32> return self.files.pop()
<33> raise StopAsyncIteration
<34>
<35> return AsyncBlobItemsIterator(filename)
<36>
<37> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.list_blob_names", mock_list_blob_names)
<38>
<39> async def mock_delete_blob(self, name, *args, **kwargs):
<40> assert name == filename
<41> return True
<42>
<43> monkey</s>
|
===========below chunk 0===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove_all(monkeypatch, mock_env, blob_manager):
# offset: 1
await blob_manager.remove_blob()
===========unchanged ref 0===========
at: _pytest.mark.structures
MARK_GEN = MarkGenerator(_ispytest=True)
at: _pytest.mark.structures.MarkGenerator
skipif: _SkipifMarkDecorator
at: os.path
basename(p: _PathLike[AnyStr]) -> AnyStr
basename(p: AnyStr) -> AnyStr
at: scripts.prepdocslib.listfilestrategy
File(content: IO, acls: Optional[dict[str, list]]=None)
at: scripts.prepdocslib.listfilestrategy.File.__init__
self.content = content
at: sys
version_info: _version_info
at: sys._version_info
minor: int
===========unchanged ref 1===========
at: tempfile
NamedTemporaryFile(mode: str=..., buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[Any]
NamedTemporaryFile(mode: Literal["r", "w", "a", "x", "r+", "w+", "a+", "x+", "rt", "wt", "at", "xt", "r+t", "w+t", "a+t", "x+t"], buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[str]
NamedTemporaryFile(mode: Literal["rb", "wb", "ab", "xb", "r+b", "w+b", "a+b", "x+b"]=..., buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[bytes]
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
===========changed ref 0===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove(monkeypatch, mock_env, blob_manager):
with NamedTemporaryFile(suffix=".pdf") as temp_file:
f = File(temp_file.file)
+ filename = os.path.basename(f.content.name)
- filename = f.content.name.split("/tmp/")[1]
# Set up mocks used by upload_blob
async def mock_exists(*args, **kwargs):
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.exists", mock_exists)
async def mock_upload_blob(self, name, *args, **kwargs):
assert name == filename
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.upload_blob", mock_upload_blob)
await blob_manager.upload_blob(f)
# Set up mocks used by remove_blob
def mock_list_blob_names(*args, **kwargs):
assert kwargs.get("name_starts_with") == filename.split(".pdf")[0]
class AsyncBlobItemsIterator:
def __init__(self, file):
self.files = [file, "dontdelete.pdf"]
def __aiter__(self):
return self
async def __anext__(self):
if self.files:
return self.files.pop()
raise StopAsyncIteration
return AsyncBlobItemsIterator(filename)
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.list_blob_names", mock_list_blob_names)
async def mock_delete_blob(self, name, *args, **kwargs):
assert name == filename
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.delete_blob", mock</s>
===========changed ref 1===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove(monkeypatch, mock_env, blob_manager):
# offset: 1
<s> return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.delete_blob", mock_delete_blob)
await blob_manager.remove_blob(f.content.name)
===========changed ref 2===========
+ # module: tests.mocks
+
+
===========changed ref 3===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 4===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 5===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 6===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 7===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 8===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 9===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 10===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 16===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
|
tests.test_blob_manager/test_create_container_upon_upload
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<2>:<add> filename = os.path.basename(f.content.name)
<del> filename = f.content.name.split("/tmp/")[1]
|
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_create_container_upon_upload(monkeypatch, mock_env, blob_manager):
<0> with NamedTemporaryFile(suffix=".pdf") as temp_file:
<1> f = File(temp_file.file)
<2> filename = f.content.name.split("/tmp/")[1]
<3>
<4> # Set up mocks used by upload_blob
<5> async def mock_exists(*args, **kwargs):
<6> return False
<7>
<8> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.exists", mock_exists)
<9>
<10> async def mock_create_container(*args, **kwargs):
<11> return
<12>
<13> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.create_container", mock_create_container)
<14>
<15> async def mock_upload_blob(self, name, *args, **kwargs):
<16> assert name == filename
<17> return True
<18>
<19> monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.upload_blob", mock_upload_blob)
<20>
<21> await blob_manager.upload_blob(f)
<22>
|
===========unchanged ref 0===========
at: _pytest.mark.structures
MARK_GEN = MarkGenerator(_ispytest=True)
at: _pytest.mark.structures.MarkGenerator
skipif: _SkipifMarkDecorator
at: os.path
basename(p: _PathLike[AnyStr]) -> AnyStr
basename(p: AnyStr) -> AnyStr
at: scripts.prepdocslib.listfilestrategy
File(content: IO, acls: Optional[dict[str, list]]=None)
at: scripts.prepdocslib.listfilestrategy.File.__init__
self.content = content
at: sys
version_info: _version_info
at: sys._version_info
minor: int
===========unchanged ref 1===========
at: tempfile
NamedTemporaryFile(mode: str=..., buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[Any]
NamedTemporaryFile(mode: Literal["r", "w", "a", "x", "r+", "w+", "a+", "x+", "rt", "wt", "at", "xt", "r+t", "w+t", "a+t", "x+t"], buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[str]
NamedTemporaryFile(mode: Literal["rb", "wb", "ab", "xb", "r+b", "w+b", "a+b", "x+b"]=..., buffering: int=..., encoding: Optional[str]=..., newline: Optional[str]=..., suffix: Optional[AnyStr]=..., prefix: Optional[AnyStr]=..., dir: Optional[_DirT[AnyStr]]=..., delete: bool=..., *, errors: Optional[str]=...) -> IO[bytes]
===========changed ref 0===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove_all(monkeypatch, mock_env, blob_manager):
with NamedTemporaryFile(suffix=".pdf") as temp_file:
f = File(temp_file.file)
print(f.content.name)
+ filename = os.path.basename(f.content.name)
- filename = f.content.name.split("/tmp/")[1]
# Set up mocks used by upload_blob
async def mock_exists(*args, **kwargs):
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.exists", mock_exists)
async def mock_upload_blob(self, name, *args, **kwargs):
assert name == filename
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.upload_blob", mock_upload_blob)
await blob_manager.upload_blob(f)
# Set up mocks used by remove_blob
def mock_list_blob_names(*args, **kwargs):
assert kwargs.get("name_starts_with") is None
class AsyncBlobItemsIterator:
def __init__(self, file):
self.files = [file]
def __aiter__(self):
return self
async def __anext__(self):
if self.files:
return self.files.pop()
raise StopAsyncIteration
return AsyncBlobItemsIterator(filename)
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.list_blob_names", mock_list_blob_names)
async def mock_delete_blob(self, name, *args, **kwargs):
assert name == filename
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.delete_blob", mock_delete_blob</s>
===========changed ref 1===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove_all(monkeypatch, mock_env, blob_manager):
# offset: 1
<s>
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.delete_blob", mock_delete_blob)
await blob_manager.remove_blob()
===========changed ref 2===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove(monkeypatch, mock_env, blob_manager):
with NamedTemporaryFile(suffix=".pdf") as temp_file:
f = File(temp_file.file)
+ filename = os.path.basename(f.content.name)
- filename = f.content.name.split("/tmp/")[1]
# Set up mocks used by upload_blob
async def mock_exists(*args, **kwargs):
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.exists", mock_exists)
async def mock_upload_blob(self, name, *args, **kwargs):
assert name == filename
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.upload_blob", mock_upload_blob)
await blob_manager.upload_blob(f)
# Set up mocks used by remove_blob
def mock_list_blob_names(*args, **kwargs):
assert kwargs.get("name_starts_with") == filename.split(".pdf")[0]
class AsyncBlobItemsIterator:
def __init__(self, file):
self.files = [file, "dontdelete.pdf"]
def __aiter__(self):
return self
async def __anext__(self):
if self.files:
return self.files.pop()
raise StopAsyncIteration
return AsyncBlobItemsIterator(filename)
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.list_blob_names", mock_list_blob_names)
async def mock_delete_blob(self, name, *args, **kwargs):
assert name == filename
return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.delete_blob", mock</s>
===========changed ref 3===========
# module: tests.test_blob_manager
@pytest.mark.asyncio
@pytest.mark.skipif(sys.version_info.minor < 10, reason="requires Python 3.10 or higher")
async def test_upload_and_remove(monkeypatch, mock_env, blob_manager):
# offset: 1
<s> return True
monkeypatch.setattr("azure.storage.blob.aio.ContainerClient.delete_blob", mock_delete_blob)
await blob_manager.remove_blob(f.content.name)
===========changed ref 4===========
+ # module: tests.mocks
+
+
===========changed ref 5===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 6===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 7===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 8===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
|
app.backend.core.modelhelper/get_token_limit
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<1>:<add> raise ValueError(f"Expected model gpt-35-turbo and above. Received: {model_id}")
<del> raise ValueError("Expected model gpt-35-turbo and above")
|
# module: app.backend.core.modelhelper
def get_token_limit(model_id: str) -> int:
<0> if model_id not in MODELS_2_TOKEN_LIMITS:
<1> raise ValueError("Expected model gpt-35-turbo and above")
<2> return MODELS_2_TOKEN_LIMITS[model_id]
<3>
|
===========unchanged ref 0===========
at: app.backend.core.modelhelper
MODELS_2_TOKEN_LIMITS = {
"gpt-35-turbo": 4000,
"gpt-3.5-turbo": 4000,
"gpt-35-turbo-16k": 16000,
"gpt-3.5-turbo-16k": 16000,
"gpt-4": 8100,
"gpt-4-32k": 32000,
"gpt-4v": 128000,
}
===========changed ref 0===========
# module: app.backend.core.modelhelper
MODELS_2_TOKEN_LIMITS = {
"gpt-35-turbo": 4000,
"gpt-3.5-turbo": 4000,
"gpt-35-turbo-16k": 16000,
"gpt-3.5-turbo-16k": 16000,
"gpt-4": 8100,
"gpt-4-32k": 32000,
+ "gpt-4v": 128000,
}
- AOAI_2_OAI = {"gpt-35-turbo": "gpt-3.5-turbo", "gpt-35-turbo-16k": "gpt-3.5-turbo-16k"}
+ AOAI_2_OAI = {"gpt-35-turbo": "gpt-3.5-turbo", "gpt-35-turbo-16k": "gpt-3.5-turbo-16k", "gpt-4v": "gpt-4-turbo-vision"}
+
===========changed ref 1===========
+ # module: tests.mocks
+
+
===========changed ref 2===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 3===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 4===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 5===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 6===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 7===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 8===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 9===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 10===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 15===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 16===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 21===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockBlob:
+ def __init__(self):
+ self.properties = BlobProperties(
+ name="Financial Market Analysis Report 2023-7.png", content_settings={"content_type": "image/png"}
+ )
+
===========changed ref 24===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 25===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def __init__(
self,
search_info: SearchInfo,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
embeddings: Optional[OpenAIEmbeddings] = None,
+ search_images: bool = False,
):
self.search_info = search_info
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.embeddings = embeddings
+ self.search_images = search_images
===========changed ref 26===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def __init__(
self,
endpoint: str,
container: str,
credential: Union[AsyncTokenCredential, str],
+ store_page_images: bool = False,
verbose: bool = False,
):
self.endpoint = endpoint
self.credential = credential
self.container = container
+ self.store_page_images = store_page_images
self.verbose = verbose
+ self.user_delegation_key: Optional[UserDelegationKey] = None
===========changed ref 27===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ """
+ Class for using image embeddings from Azure AI Vision
+ To learn more, please visit https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api
+ """
+
|
app.backend.core.modelhelper/num_tokens_from_messages
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<13>:<add>
<15>:<add> for key, value in message.items():
<del> for value in message.values():
<16>:<add> if isinstance(value, list):
<add> for v in value:
<add> # TODO: Update token count for images https://github.com/openai/openai-cookbook/pull/881/files
<add> if isinstance(v, str):
<add> num_tokens += len(encoding.encode(v))
<add> else:
<add> num_tokens += len(encoding.encode(value))
<del> num_tokens += len(encoding.encode(str(value)))
|
# module: app.backend.core.modelhelper
def num_tokens_from_messages(message: dict[str, str], model: str) -> int:
<0> """
<1> Calculate the number of tokens required to encode a message.
<2> Args:
<3> message (dict): The message to encode, represented as a dictionary.
<4> model (str): The name of the model to use for encoding.
<5> Returns:
<6> int: The total number of tokens required to encode the message.
<7> Example:
<8> message = {'role': 'user', 'content': 'Hello, how are you?'}
<9> model = 'gpt-3.5-turbo'
<10> num_tokens_from_messages(message, model)
<11> output: 11
<12> """
<13> encoding = tiktoken.encoding_for_model(get_oai_chatmodel_tiktok(model))
<14> num_tokens = 2 # For "role" and "content" keys
<15> for value in message.values():
<16> num_tokens += len(encoding.encode(str(value)))
<17> return num_tokens
<18>
|
===========unchanged ref 0===========
at: app.backend.core.modelhelper
get_oai_chatmodel_tiktok(aoaimodel: str) -> str
at: tiktoken.model
encoding_for_model(model_name: str) -> Encoding
===========changed ref 0===========
# module: app.backend.core.modelhelper
def get_token_limit(model_id: str) -> int:
if model_id not in MODELS_2_TOKEN_LIMITS:
+ raise ValueError(f"Expected model gpt-35-turbo and above. Received: {model_id}")
- raise ValueError("Expected model gpt-35-turbo and above")
return MODELS_2_TOKEN_LIMITS[model_id]
===========changed ref 1===========
# module: app.backend.core.modelhelper
MODELS_2_TOKEN_LIMITS = {
"gpt-35-turbo": 4000,
"gpt-3.5-turbo": 4000,
"gpt-35-turbo-16k": 16000,
"gpt-3.5-turbo-16k": 16000,
"gpt-4": 8100,
"gpt-4-32k": 32000,
+ "gpt-4v": 128000,
}
- AOAI_2_OAI = {"gpt-35-turbo": "gpt-3.5-turbo", "gpt-35-turbo-16k": "gpt-3.5-turbo-16k"}
+ AOAI_2_OAI = {"gpt-35-turbo": "gpt-3.5-turbo", "gpt-35-turbo-16k": "gpt-3.5-turbo-16k", "gpt-4v": "gpt-4-turbo-vision"}
+
===========changed ref 2===========
+ # module: tests.mocks
+
+
===========changed ref 3===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 4===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 5===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 6===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 7===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 8===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 9===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 10===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 16===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 17===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 18===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 19===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 22===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockBlob:
+ def __init__(self):
+ self.properties = BlobProperties(
+ name="Financial Market Analysis Report 2023-7.png", content_settings={"content_type": "image/png"}
+ )
+
===========changed ref 25===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 26===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
def __init__(
self,
search_info: SearchInfo,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
embeddings: Optional[OpenAIEmbeddings] = None,
+ search_images: bool = False,
):
self.search_info = search_info
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.embeddings = embeddings
+ self.search_images = search_images
===========changed ref 27===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def __init__(
self,
endpoint: str,
container: str,
credential: Union[AsyncTokenCredential, str],
+ store_page_images: bool = False,
verbose: bool = False,
):
self.endpoint = endpoint
self.credential = credential
self.container = container
+ self.store_page_images = store_page_images
self.verbose = verbose
+ self.user_delegation_key: Optional[UserDelegationKey] = None
|
tests.e2e/test_chat
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
# module: tests.e2e
def test_chat(page: Page, live_server_url: str):
<0> # Set up a mock route to the /chat endpoint with streaming results
<1> def handle(route: Route):
<2> # Assert that session_state is specified in the request (None for now)
<3> session_state = route.request.post_data_json["session_state"]
<4> assert session_state is None
<5> # Read the JSONL from our snapshot results and return as the response
<6> f = open("tests/snapshots/test_app/test_chat_stream_text/client0/result.jsonlines")
<7> jsonl = f.read()
<8> f.close()
<9> route.fulfill(body=jsonl, status=200, headers={"Transfer-encoding": "Chunked"})
<10>
<11> page.route("*/**/chat", handle)
<12>
<13> # Check initial page state
<14> page.goto(live_server_url)
<15> expect(page).to_have_title("GPT + Enterprise data | Sample")
<16> expect(page.get_by_role("heading", name="Chat with your data")).to_be_visible()
<17> expect(page.get_by_role("button", name="Clear chat")).to_be_disabled()
<18> expect(page.get_by_role("button", name="Developer settings")).to_be_enabled()
<19>
<20> # Ask a question and wait for the message to appear
<21> page.get_by_placeholder("Type a new question (e.g. does my plan cover annual eye exams?)").click()
<22> page.get_by_placeholder("Type a new question (e.g. does my plan cover annual eye exams?)").fill(
<23> "Whats the dental plan?"
<24> )
<25> page.get_by_role("button", name="Ask question button").click()
<26>
<27> expect(page.get_by_text("Whats the dental plan?")).to_be_visible()
<28> expect(page.get_by_text("The capital of France is</s>
|
===========below chunk 0===========
# module: tests.e2e
def test_chat(page: Page, live_server_url: str):
# offset: 1
expect(page.get_by_role("button", name="Clear chat")).to_be_enabled()
# Show the citation document
page.get_by_text("1. Benefit_Options-2.pdf").click()
expect(page.get_by_role("tab", name="Citation")).to_be_visible()
expect(page.get_by_title("Citation")).to_be_visible()
# Show the thought process
page.get_by_label("Show thought process").click()
expect(page.get_by_title("Thought process")).to_be_visible()
expect(page.get_by_text("Searched for:")).to_be_visible()
# Show the supporting content
page.get_by_label("Show supporting content").click()
expect(page.get_by_title("Supporting content")).to_be_visible()
expect(page.get_by_role("heading", name="Benefit_Options-2.pdf")).to_be_visible()
# Clear the chat
page.get_by_role("button", name="Clear chat").click()
expect(page.get_by_text("Whats the dental plan?")).not_to_be_visible()
expect(page.get_by_text("The capital of France is Paris.")).not_to_be_visible()
expect(page.get_by_role("button", name="Clear chat")).to_be_disabled()
===========unchanged ref 0===========
at: io.BufferedRandom
close(self) -> None
at: io.BufferedWriter
read(self, size: Optional[int]=..., /) -> bytes
at: typing.IO
__slots__ = ()
close() -> None
read(n: int=...) -> AnyStr
===========changed ref 0===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 1===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 2===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 4===========
+ # module: tests
+
+
===========changed ref 5===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 6===========
+ # module: tests.mocks
+
+
===========changed ref 7===========
# module: tests.test_content_file
-
-
===========changed ref 8===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 10===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 20===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 25===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 26===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 27===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 28===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 29===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 32===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 33===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 35===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
|
|
app.backend.core.messagebuilder/MessageBuilder.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<1>:<add> ChatCompletionSystemMessageParam(role="system", content=unicodedata.normalize("NFC", system_content))
<del> ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
|
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
<0> self.messages: list[ChatCompletionMessageParam] = [
<1> ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
<2> ]
<3> self.model = chatgpt_model
<4>
|
===========unchanged ref 0===========
at: unicodedata
normalize(form: Text, unistr: Text, /) -> Text
===========changed ref 0===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 1===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 2===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 4===========
+ # module: tests
+
+
===========changed ref 5===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 6===========
+ # module: tests.mocks
+
+
===========changed ref 7===========
# module: tests.test_content_file
-
-
===========changed ref 8===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 10===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 20===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 25===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 26===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 27===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 28===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 29===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 32===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 33===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 35===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 36===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
===========changed ref 37===========
+ # module: tests.test_chatvisionapproach
+ def test_build_filter(chat_approach):
+ result = chat_approach.build_filter({"exclude_category": "test_category"}, {})
+ assert result == "category ne 'test_category'"
+
===========changed ref 38===========
+ # module: tests.mocks
+ class MockBlob:
+ def __init__(self):
+ self.properties = BlobProperties(
+ name="Financial Market Analysis Report 2023-7.png", content_settings={"content_type": "image/png"}
+ )
+
===========changed ref 39===========
+ # module: app.backend.core.imageshelper
+ class ImageURL(TypedDict, total=False):
+ url: Required[str]
+ """Either a URL of the image or the base64 encoded image data."""
+
+ detail: Literal["auto", "low", "high"]
+ """Specifies the detail level of the image."""
+
===========changed ref 40===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_embedding_arguments(self) -> dict[str, Any]:
+ return {
+ "model": self.open_ai_model_name,
+ "api_key": self.credential,
+ "api_type": "openai",
+ "organization": self.organization,
+ }
+
===========changed ref 41===========
+ # module: app.backend.approaches.retrievethenreadvision
+ # Replace these with your own values, either in environment variables or directly here
+ AZURE_STORAGE_ACCOUNT = os.getenv("AZURE_STORAGE_ACCOUNT")
+ AZURE_STORAGE_CONTAINER = os.getenv("AZURE_STORAGE_CONTAINER")
+
|
app.backend.core.messagebuilder/MessageBuilder.insert_message
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<5>:<add> content (str | List[ChatCompletionContentPartParam]): The content of the message.
<del> content (str): The content of the message.
<11>:<add> elif role == "system" and isinstance(content, str):
<del> elif role == "system":
<12>:<add> message = ChatCompletionSystemMessageParam(role="system", content=unicodedata.normalize("NFC", content))
<del> message = ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(content))
<13>:<add> elif role == "assistant" and isinstance(content, str):
<add> message = ChatCompletionAssistantMessageParam(
<add> role="assistant", content=unicodedata.normalize("NFC", content)
<add> )
<del> elif role == "assistant":
<14>:<del> message = ChatCompletionAssistantMessageParam(role="assistant", content=self.normalize_content(content))
|
# module: app.backend.core.messagebuilder
class MessageBuilder:
+ def insert_message(self, role: str, content: Union[str, List[ChatCompletionContentPartParam]], index: int = 1):
- def insert_message(self, role: str, content: str, index: int = 1):
<0> """
<1> Inserts a message into the conversation at the specified index,
<2> or at index 1 (after system message) if no index is specified.
<3> Args:
<4> role (str): The role of the message sender (either "user", "system", or "assistant").
<5> content (str): The content of the message.
<6> index (int): The index at which to insert the message.
<7> """
<8> message: ChatCompletionMessageParam
<9> if role == "user":
<10> message = ChatCompletionUserMessageParam(role="user", content=self.normalize_content(content))
<11> elif role == "system":
<12> message = ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(content))
<13> elif role == "assistant":
<14> message = ChatCompletionAssistantMessageParam(role="assistant", content=self.normalize_content(content))
<15> else:
<16> raise ValueError(f"Invalid role: {role}")
<17> self.messages.insert(index, message)
<18>
|
===========unchanged ref 0===========
at: app.backend.core.messagebuilder.MessageBuilder
normalize_content(content: Union[str, List[ChatCompletionContentPartParam]])
at: typing
List = _alias(list, 1, inst=False, name='List')
at: unicodedata
normalize(form: Text, unistr: Text, /) -> Text
===========changed ref 0===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=unicodedata.normalize("NFC", system_content))
- ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
]
self.model = chatgpt_model
===========changed ref 1===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 2===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 3===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 4===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 5===========
+ # module: tests
+
+
===========changed ref 6===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 7===========
+ # module: tests.mocks
+
+
===========changed ref 8===========
# module: tests.test_content_file
-
-
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 11===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 26===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 28===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 29===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 33===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 35===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 36===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 37===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
===========changed ref 38===========
+ # module: tests.test_chatvisionapproach
+ def test_build_filter(chat_approach):
+ result = chat_approach.build_filter({"exclude_category": "test_category"}, {})
+ assert result == "category ne 'test_category'"
+
===========changed ref 39===========
+ # module: tests.mocks
+ class MockBlob:
+ def __init__(self):
+ self.properties = BlobProperties(
+ name="Financial Market Analysis Report 2023-7.png", content_settings={"content_type": "image/png"}
+ )
+
===========changed ref 40===========
+ # module: app.backend.core.imageshelper
+ class ImageURL(TypedDict, total=False):
+ url: Required[str]
+ """Either a URL of the image or the base64 encoded image data."""
+
+ detail: Literal["auto", "low", "high"]
+ """Specifies the detail level of the image."""
+
|
app.backend.core.messagebuilder/MessageBuilder.normalize_content
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<0>:<add> if isinstance(content, str):
<add> return unicodedata.normalize("NFC", content)
<del> return unicodedata.normalize("NFC", content)
<1>:<add> elif isinstance(content, list):
<add> for part in content:
<add> if "image_url" not in part:
<add> part["text"] = unicodedata.normalize("NFC", part["text"])
<add> return content
|
# module: app.backend.core.messagebuilder
class MessageBuilder:
+ def normalize_content(self, content: Union[str, List[ChatCompletionContentPartParam]]):
- def normalize_content(self, content: str):
<0> return unicodedata.normalize("NFC", content)
<1>
|
===========changed ref 0===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
def __init__(self, system_content: str, chatgpt_model: str):
self.messages: list[ChatCompletionMessageParam] = [
+ ChatCompletionSystemMessageParam(role="system", content=unicodedata.normalize("NFC", system_content))
- ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(system_content))
]
self.model = chatgpt_model
===========changed ref 1===========
# module: app.backend.core.messagebuilder
class MessageBuilder:
+ def insert_message(self, role: str, content: Union[str, List[ChatCompletionContentPartParam]], index: int = 1):
- def insert_message(self, role: str, content: str, index: int = 1):
"""
Inserts a message into the conversation at the specified index,
or at index 1 (after system message) if no index is specified.
Args:
role (str): The role of the message sender (either "user", "system", or "assistant").
+ content (str | List[ChatCompletionContentPartParam]): The content of the message.
- content (str): The content of the message.
index (int): The index at which to insert the message.
"""
message: ChatCompletionMessageParam
if role == "user":
message = ChatCompletionUserMessageParam(role="user", content=self.normalize_content(content))
+ elif role == "system" and isinstance(content, str):
- elif role == "system":
+ message = ChatCompletionSystemMessageParam(role="system", content=unicodedata.normalize("NFC", content))
- message = ChatCompletionSystemMessageParam(role="system", content=self.normalize_content(content))
+ elif role == "assistant" and isinstance(content, str):
+ message = ChatCompletionAssistantMessageParam(
+ role="assistant", content=unicodedata.normalize("NFC", content)
+ )
- elif role == "assistant":
- message = ChatCompletionAssistantMessageParam(role="assistant", content=self.normalize_content(content))
else:
raise ValueError(f"Invalid role: {role}")
self.messages.insert(index, message)
===========changed ref 2===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 4===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 5===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 6===========
+ # module: tests
+
+
===========changed ref 7===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 8===========
+ # module: tests.mocks
+
+
===========changed ref 9===========
# module: tests.test_content_file
-
-
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 11===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 12===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 23===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 26===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 27===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 28===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 29===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 30===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 33===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 34===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 35===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 36===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
|
tests.test_app/test_chat_with_history
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<18>:<add> assert thoughts_contains_text(result["choices"][0]["context"]["thoughts"], "performance review")
<del> assert result["choices"][0]["context"]["thoughts"].find("performance review") != -1
|
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_with_history(client, snapshot):
<0> response = await client.post(
<1> "/chat",
<2> json={
<3> "messages": [
<4> {"content": "What happens in a performance review?", "role": "user"},
<5> {
<6> "content": "During a performance review, employees will receive feedback on their performance over the past year, including both successes and areas for improvement. The feedback will be provided by the employee's supervisor and is intended to help the employee develop and grow in their role [employee_handbook-3.pdf]. The review is a two-way dialogue between the employee and their manager, so employees are encouraged to be honest and open during the process [employee_handbook-3.pdf]. The employee will also have the opportunity to discuss their goals and objectives for the upcoming year [employee_handbook-3.pdf]. A written summary of the performance review will be provided to the employee, which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
<7> "role": "assistant",
<8> },
<9> {"content": "Is dental covered?", "role": "user"},
<10> ],
<11> "context": {
<12> "overrides": {"retrieval_mode": "text"},
<13> },
<14> },
<15> )
<16> assert response.status_code == 200
<17> result = await response.get_json()
<18> assert result["choices"][0]["context"]["thoughts"].find("performance review") != -1
<19> snapshot.assert_match(json.dumps(result, indent=4), "result.json")
<20>
|
===========unchanged ref 0===========
at: _pytest.mark.structures
MARK_GEN = MarkGenerator(_ispytest=True)
at: tests.test_app.test_chat_stream_text_filter
response = await auth_client.post(
"/chat",
headers={"Authorization": "Bearer MockToken"},
json={
"stream": True,
"messages": [{"content": "What is the capital of France?", "role": "user"}],
"context": {
"overrides": {
"retrieval_mode": "text",
"use_oid_security_filter": True,
"use_groups_security_filter": True,
"exclude_category": "excluded",
}
},
},
)
===========changed ref 0===========
# module: tests.test_app
+ def thoughts_contains_text(thoughts, text):
+ found = False
+ for thought in thoughts:
+ description = thought["description"]
+ if isinstance(description, str) and text in description:
+ found = True
+ break
+ elif isinstance(description, list) and any(text in item for item in description):
+ found = True
+ break
+ return found
+
===========changed ref 1===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 2===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 3===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 4===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 5===========
+ # module: tests
+
+
===========changed ref 6===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 7===========
+ # module: tests.mocks
+
+
===========changed ref 8===========
# module: tests.test_content_file
-
-
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 11===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 26===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 28===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 29===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 33===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 35===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 36===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 37===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
===========changed ref 38===========
+ # module: tests.test_chatvisionapproach
+ def test_build_filter(chat_approach):
+ result = chat_approach.build_filter({"exclude_category": "test_category"}, {})
+ assert result == "category ne 'test_category'"
+
===========changed ref 39===========
+ # module: tests.mocks
+ class MockBlob:
+ def __init__(self):
+ self.properties = BlobProperties(
+ name="Financial Market Analysis Report 2023-7.png", content_settings={"content_type": "image/png"}
+ )
+
|
tests.test_app/test_chat_with_long_history
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<22>:<add> assert not thoughts_contains_text(result["choices"][0]["context"]["thoughts"], "Is there a dress code?")
<del> assert result["choices"][0]["context"]["thoughts"].find("Is there a dress code?") == -1
|
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_with_long_history(client, snapshot, caplog):
<0> """This test makes sure that the history is truncated to max tokens minus 1024."""
<1> caplog.set_level(logging.DEBUG)
<2> response = await client.post(
<3> "/chat",
<4> json={
<5> "messages": [
<6> {"role": "user", "content": "Is there a dress code?"}, # 9 tokens
<7> {
<8> "role": "assistant",
<9> "content": "Yes, there is a dress code at Contoso Electronics. Look sharp! [employee_handbook-1.pdf]"
<10> * 150,
<11> }, # 3900 tokens
<12> {"role": "user", "content": "What does a product manager do?"}, # 10 tokens
<13> ],
<14> "context": {
<15> "overrides": {"retrieval_mode": "text"},
<16> },
<17> },
<18> )
<19> assert response.status_code == 200
<20> result = await response.get_json()
<21> # Assert that it doesn't find the first message, since it wouldn't fit in the max tokens.
<22> assert result["choices"][0]["context"]["thoughts"].find("Is there a dress code?") == -1
<23> assert "Reached max tokens" in caplog.text
<24> snapshot.assert_match(json.dumps(result, indent=4), "result.json")
<25>
|
===========unchanged ref 0===========
at: _pytest.mark.structures
MARK_GEN = MarkGenerator(_ispytest=True)
at: json
dumps(obj: Any, *, skipkeys: bool=..., ensure_ascii: bool=..., check_circular: bool=..., allow_nan: bool=..., cls: Optional[Type[JSONEncoder]]=..., indent: Union[None, int, str]=..., separators: Optional[Tuple[str, str]]=..., default: Optional[Callable[[Any], Any]]=..., sort_keys: bool=..., **kwds: Any) -> str
at: logging
DEBUG = 10
at: tests.test_app
thoughts_contains_text(thoughts, text)
at: tests.test_app.test_chat_with_history
response = await client.post(
"/chat",
json={
"messages": [
{"content": "What happens in a performance review?", "role": "user"},
{
"content": "During a performance review, employees will receive feedback on their performance over the past year, including both successes and areas for improvement. The feedback will be provided by the employee's supervisor and is intended to help the employee develop and grow in their role [employee_handbook-3.pdf]. The review is a two-way dialogue between the employee and their manager, so employees are encouraged to be honest and open during the process [employee_handbook-3.pdf]. The employee will also have the opportunity to discuss their goals and objectives for the upcoming year [employee_handbook-3.pdf]. A written summary of the performance review will be provided to the employee, which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
"role": "assistant",
},
{"content": "Is dental covered?", "role": "user"},
],
"context": {
"overrides": {"retrieval_mode": "text"},
},
},
)
===========changed ref 0===========
# module: tests.test_app
+ def thoughts_contains_text(thoughts, text):
+ found = False
+ for thought in thoughts:
+ description = thought["description"]
+ if isinstance(description, str) and text in description:
+ found = True
+ break
+ elif isinstance(description, list) and any(text in item for item in description):
+ found = True
+ break
+ return found
+
===========changed ref 1===========
# module: tests.test_app
@pytest.mark.asyncio
async def test_chat_with_history(client, snapshot):
response = await client.post(
"/chat",
json={
"messages": [
{"content": "What happens in a performance review?", "role": "user"},
{
"content": "During a performance review, employees will receive feedback on their performance over the past year, including both successes and areas for improvement. The feedback will be provided by the employee's supervisor and is intended to help the employee develop and grow in their role [employee_handbook-3.pdf]. The review is a two-way dialogue between the employee and their manager, so employees are encouraged to be honest and open during the process [employee_handbook-3.pdf]. The employee will also have the opportunity to discuss their goals and objectives for the upcoming year [employee_handbook-3.pdf]. A written summary of the performance review will be provided to the employee, which will include a rating of their performance, feedback, and goals and objectives for the upcoming year [employee_handbook-3.pdf].",
"role": "assistant",
},
{"content": "Is dental covered?", "role": "user"},
],
"context": {
"overrides": {"retrieval_mode": "text"},
},
},
)
assert response.status_code == 200
result = await response.get_json()
+ assert thoughts_contains_text(result["choices"][0]["context"]["thoughts"], "performance review")
- assert result["choices"][0]["context"]["thoughts"].find("performance review") != -1
snapshot.assert_match(json.dumps(result, indent=4), "result.json")
===========changed ref 2===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 4===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 5===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 6===========
+ # module: tests
+
+
===========changed ref 7===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 8===========
+ # module: tests.mocks
+
+
===========changed ref 9===========
# module: tests.test_content_file
-
-
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 11===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 12===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 23===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 26===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 27===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
|
app.backend.approaches.retrievethenread/RetrieveThenReadApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<1>:<add> self.chatgpt_deployment = chatgpt_deployment
|
<s>_client: SearchClient,
openai_client: AsyncOpenAI,
chatgpt_model: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
embedding_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
<0> self.search_client = search_client
<1> self.openai_client = openai_client
<2> self.chatgpt_model = chatgpt_model
<3> self.embedding_model = embedding_model
<4> self.chatgpt_deployment = chatgpt_deployment
<5> self.embedding_deployment = embedding_deployment
<6> self.sourcepage_field = sourcepage_field
<7> self.content_field = content_field
<8> self.query_language = query_language
<9> self.query_speller = query_speller
<10>
|
===========unchanged ref 0===========
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach
system_chat_template = (
"You are an intelligent assistant helping Contoso Inc employees with their healthcare plan questions and employee handbook questions. "
+ "Use 'you' to refer to the individual asking the questions even if they ask with 'I'. "
+ "Answer the following question using only the data provided in the sources below. "
+ "For tabular information return it as an html table. Do not return markdown format. "
+ "Each source has a name followed by colon and the actual information, always include the source name for each fact you use in the response. "
+ "If you cannot answer using the sources below, say you don't know. Use below example to answer"
)
question = """
'What is the deductible for the employee plan for a visit to Overlake in Bellevue?'
Sources:
info1.txt: deductibles depend on whether you are in-network or out-of-network. In-network deductibles are $500 for employee and $1000 for family. Out-of-network deductibles are $1000 for employee and $2000 for family.
info2.pdf: Overlake is in-network for the employee plan.
info3.pdf: Overlake is the name of the area that includes a park and ride near Bellevue.
info4.pdf: In-network institutions include Overlake, Swedish and others in the region
"""
answer = "In-network deductibles are $500 for employee and $1000 for family [info1.txt] and Overlake is in-network for the employee plan [info2.pdf][info4.pdf]."
at: approaches.approach.Approach
__init__(self, search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
===========changed ref 0===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 1===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 2===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 4===========
+ # module: tests
+
+
===========changed ref 5===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 6===========
+ # module: tests.mocks
+
+
===========changed ref 7===========
# module: tests.test_content_file
-
-
===========changed ref 8===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 10===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 20===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 25===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 26===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 27===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 28===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 29===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 32===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 33===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 35===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
|
app.backend.approaches.retrievethenread/RetrieveThenReadApproach.run
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<5>:<add> use_semantic_ranker = overrides.get("semantic_ranker") and has_text
<add>
<11>:<del> embedding = await self.openai_client.embeddings.create(
<12>:<del> # Azure Open AI takes the deployment name as the model name
<13>:<del> model=self.embedding_deployment if self.embedding_deployment else self.embedding_model,
<14>:<del> input=q,
<15>:<del> )
<16>:<del> query_vector = embedding.data[0].embedding
<17>:<del> vectors.append(RawVectorQuery(vector=query_vector, k=50, fields="embedding"))
<18>:<add> vectors.append(await self.compute_text_embedding(q))
<20>:<add> query_text = q if has_text else None
<del> query_text = q if has_text else ""
<22>:<del> # Use semantic ranker if requested and if retrieval mode is text or hybrid (vectors + text)
<23>:<del> if overrides.get("semantic_ranker") and has_text:
<24>:<del> r = await self.search_client.search(
<25>:<del> query_text,
<26>:<del> filter=filter,
|
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
<0> q = messages[-1]["content"]
<1> overrides = context.get("overrides", {})
<2> auth_claims = context.get("auth_claims", {})
<3> has_text = overrides.get("retrieval_mode") in ["text", "hybrid", None]
<4> has_vector = overrides.get("retrieval_mode") in ["vectors", "hybrid", None]
<5> use_semantic_captions = True if overrides.get("semantic_captions") and has_text else False
<6> top = overrides.get("top", 3)
<7> filter = self.build_filter(overrides, auth_claims)
<8> # If retrieval mode includes vectors, compute an embedding for the query
<9> vectors: list[VectorQuery] = []
<10> if has_vector:
<11> embedding = await self.openai_client.embeddings.create(
<12> # Azure Open AI takes the deployment name as the model name
<13> model=self.embedding_deployment if self.embedding_deployment else self.embedding_model,
<14> input=q,
<15> )
<16> query_vector = embedding.data[0].embedding
<17> vectors.append(RawVectorQuery(vector=query_vector, k=50, fields="embedding"))
<18>
<19> # Only keep the text query if the retrieval mode uses text, otherwise drop it
<20> query_text = q if has_text else ""
<21>
<22> # Use semantic ranker if requested and if retrieval mode is text or hybrid (vectors + text)
<23> if overrides.get("semantic_ranker") and has_text:
<24> r = await self.search_client.search(
<25> query_text,
<26> filter=filter,
</s>
|
===========below chunk 0===========
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
# offset: 1
query_language=self.query_language,
query_speller=self.query_speller,
semantic_configuration_name="default",
top=top,
query_caption="extractive|highlight-false" if use_semantic_captions else None,
vector_queries=vectors,
)
else:
r = await self.search_client.search(
query_text,
filter=filter,
top=top,
vector_queries=vectors,
)
if use_semantic_captions:
results = [
doc[self.sourcepage_field] + ": " + nonewlines(" . ".join([c.text for c in doc["@search.captions"]]))
async for doc in r
]
else:
results = [doc[self.sourcepage_field] + ": " + nonewlines(doc[self.content_field]) async for doc in r]
content = "\n".join(results)
message_builder = MessageBuilder(
overrides.get("prompt_template") or self.system_chat_template, self.chatgpt_model
)
# add user question
user_content = q + "\n" + f"Sources:\n {content}"
message_builder.insert_message("user", user_content)
# Add shots/samples. This helps model to mimic response and make sure they match rules laid out in system message.
message_builder.insert_message("assistant", self.answer)
message_builder.insert_message("user", self.question)
chat_completion = (
await self.openai</s>
===========below chunk 1===========
# module: app.backend.approaches.retrievethenread
class RetrieveThenReadApproach(Approach):
def run(
self,
messages: list[dict],
stream: bool = False, # Stream is not used in this approach
session_state: Any = None,
context: dict[str, Any] = {},
) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]:
# offset: 2
<s> message_builder.insert_message("user", self.question)
chat_completion = (
await self.openai_client.chat.completions.create(
# Azure Open AI takes the deployment name as the model name
model=self.chatgpt_deployment if self.chatgpt_deployment else self.chatgpt_model,
messages=message_builder.messages,
temperature=overrides.get("temperature") or 0.3,
max_tokens=1024,
n=1,
)
).model_dump()
extra_info = {
"data_points": results,
"thoughts": f"Question:<br>{query_text}<br><br>Prompt:<br>"
+ "\n\n".join([str(message) for message in message_builder.messages]),
}
chat_completion["choices"][0]["context"] = extra_info
chat_completion["choices"][0]["session_state"] = session_state
return chat_completion
===========unchanged ref 0===========
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach
system_chat_template = (
"You are an intelligent assistant helping Contoso Inc employees with their healthcare plan questions and employee handbook questions. "
+ "Use 'you' to refer to the individual asking the questions even if they ask with 'I'. "
+ "Answer the following question using only the data provided in the sources below. "
+ "For tabular information return it as an html table. Do not return markdown format. "
+ "Each source has a name followed by colon and the actual information, always include the source name for each fact you use in the response. "
+ "If you cannot answer using the sources below, say you don't know. Use below example to answer"
)
question = """
'What is the deductible for the employee plan for a visit to Overlake in Bellevue?'
Sources:
info1.txt: deductibles depend on whether you are in-network or out-of-network. In-network deductibles are $500 for employee and $1000 for family. Out-of-network deductibles are $1000 for employee and $2000 for family.
info2.pdf: Overlake is in-network for the employee plan.
info3.pdf: Overlake is the name of the area that includes a park and ride near Bellevue.
info4.pdf: In-network institutions include Overlake, Swedish and others in the region
"""
answer = "In-network deductibles are $500 for employee and $1000 for family [info1.txt] and Overlake is in-network for the employee plan [info2.pdf][info4.pdf]."
at: app.backend.approaches.retrievethenread.RetrieveThenReadApproach.__init__
self.openai_client = openai_client
self.chatgpt_model = chatgpt_model
self.chatgpt_deployment = chatgpt_deployment
===========unchanged ref 1===========
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
at: approaches.approach.Approach
build_filter(overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]
search(top: int, query_text: Optional[str], filter: Optional[str], vectors: List[VectorQuery], use_semantic_ranker: bool, use_semantic_captions: bool) -> List[Document]
get_sources_content(results: List[Document], use_semantic_captions: bool, use_image_citation: bool) -> list[str]
compute_text_embedding(q: str)
run(self, messages: list[dict], stream: bool=False, session_state: Any=None, context: dict[str, Any]={}) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]
at: core.messagebuilder
MessageBuilder(system_content: str, chatgpt_model: str)
at: core.messagebuilder.MessageBuilder
insert_message(role: str, content: Union[str, List[ChatCompletionContentPartParam]], index: int=1)
at: core.messagebuilder.MessageBuilder.__init__
self.messages: list[ChatCompletionMessageParam] = [
ChatCompletionSystemMessageParam(role="system", content=unicodedata.normalize("NFC", system_content))
]
at: typing
AsyncGenerator = _alias(collections.abc.AsyncGenerator, 2)
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
|
app.backend.app/ask
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<7>:<add> use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
<add> approach: Approach
<add> if use_gpt4v and CONFIG_ASK_VISION_APPROACH in current_app.config:
<add> approach = cast(Approach, current_app.config[CONFIG_ASK_VISION_APPROACH])
<add> else:
<add> approach = cast(Approach, current_app.config[CONFIG_ASK_APPROACH])
<del> approach = current_app.config[CONFIG_ASK_APPROACH]
|
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
<0> if not request.is_json:
<1> return jsonify({"error": "request must be json"}), 415
<2> request_json = await request.get_json()
<3> context = request_json.get("context", {})
<4> auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
<5> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<6> try:
<7> approach = current_app.config[CONFIG_ASK_APPROACH]
<8> r = await approach.run(
<9> request_json["messages"], context=context, session_state=request_json.get("session_state")
<10> )
<11> return jsonify(r)
<12> except Exception as error:
<13> return error_response(error, "/ask")
<14>
|
===========unchanged ref 0===========
at: app.backend.app
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_AUTH_CLIENT = "auth_client"
bp = Blueprint("routes", __name__, static_folder="static")
error_response(error: Exception, route: str, status_code: int=500)
===========changed ref 0===========
# module: app.backend.app
+ CONFIG_OPENAI_TOKEN = "openai_token"
+ CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
+ CONFIG_ASK_VISION_APPROACH = "ask_vision_approach"
+ CONFIG_CHAT_VISION_APPROACH = "chat_vision_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
+ CONFIG_GPT4V_DEPLOYED = "gpt4v_deployed"
CONFIG_SEARCH_CLIENT = "search_client"
CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 1===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 2===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 3===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 4===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 5===========
+ # module: tests
+
+
===========changed ref 6===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 7===========
+ # module: tests.mocks
+
+
===========changed ref 8===========
# module: tests.test_content_file
-
-
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 11===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 26===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 28===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 29===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 33===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 35===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 36===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 37===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
|
app.backend.app/format_as_ndjson
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<2>:<add> yield json.dumps(event, ensure_ascii=False, cls=JSONEncoder) + "\n"
<del> yield json.dumps(event, ensure_ascii=False) + "\n"
<3>:<add> except Exception as error:
<del> except Exception as e:
<4>:<add> logging.exception("Exception while generating response stream: %s", error)
<del> logging.exception("Exception while generating response stream: %s", e)
<5>:<add> yield json.dumps(error_dict(error))
<del> yield json.dumps(error_dict(e))
|
# module: app.backend.app
def format_as_ndjson(r: AsyncGenerator[dict, None]) -> AsyncGenerator[str, None]:
<0> try:
<1> async for event in r:
<2> yield json.dumps(event, ensure_ascii=False) + "\n"
<3> except Exception as e:
<4> logging.exception("Exception while generating response stream: %s", e)
<5> yield json.dumps(error_dict(e))
<6>
|
===========unchanged ref 0===========
at: app.backend.app
error_dict(error: Exception) -> dict
at: json
dumps(obj: Any, *, skipkeys: bool=..., ensure_ascii: bool=..., check_circular: bool=..., allow_nan: bool=..., cls: Optional[Type[JSONEncoder]]=..., indent: Union[None, int, str]=..., separators: Optional[Tuple[str, str]]=..., default: Optional[Callable[[Any], Any]]=..., sort_keys: bool=..., **kwds: Any) -> str
at: logging
exception(msg: Any, *args: Any, exc_info: _ExcInfoType=..., stack_info: bool=..., extra: Optional[Dict[str, Any]]=..., **kwargs: Any) -> None
at: typing
AsyncGenerator = _alias(collections.abc.AsyncGenerator, 2)
===========changed ref 0===========
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
if not request.is_json:
return jsonify({"error": "request must be json"}), 415
request_json = await request.get_json()
context = request_json.get("context", {})
auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
try:
+ use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
+ approach: Approach
+ if use_gpt4v and CONFIG_ASK_VISION_APPROACH in current_app.config:
+ approach = cast(Approach, current_app.config[CONFIG_ASK_VISION_APPROACH])
+ else:
+ approach = cast(Approach, current_app.config[CONFIG_ASK_APPROACH])
- approach = current_app.config[CONFIG_ASK_APPROACH]
r = await approach.run(
request_json["messages"], context=context, session_state=request_json.get("session_state")
)
return jsonify(r)
except Exception as error:
return error_response(error, "/ask")
===========changed ref 1===========
# module: app.backend.app
+ CONFIG_OPENAI_TOKEN = "openai_token"
+ CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
+ CONFIG_ASK_VISION_APPROACH = "ask_vision_approach"
+ CONFIG_CHAT_VISION_APPROACH = "chat_vision_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
+ CONFIG_GPT4V_DEPLOYED = "gpt4v_deployed"
CONFIG_SEARCH_CLIENT = "search_client"
CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 2===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 4===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 5===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 6===========
+ # module: tests
+
+
===========changed ref 7===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 8===========
+ # module: tests.mocks
+
+
===========changed ref 9===========
# module: tests.test_content_file
-
-
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 11===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 12===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 23===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 26===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 27===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 28===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 29===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 30===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
|
app.backend.app/chat
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<6>:<add>
<7>:<add> use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
<add> approach: Approach
<add> if use_gpt4v and CONFIG_CHAT_VISION_APPROACH in current_app.config:
<add> approach = cast(Approach, current_app.config[CONFIG_CHAT_VISION_APPROACH])
<add> else:
<add> approach = cast(Approach, current_app.config[CONFIG_CHAT_APPROACH])
<del> approach = current_app.config[CONFIG_CHAT_APPROACH]
<8>:<add>
|
# module: app.backend.app
@bp.route("/chat", methods=["POST"])
async def chat():
<0> if not request.is_json:
<1> return jsonify({"error": "request must be json"}), 415
<2> request_json = await request.get_json()
<3> context = request_json.get("context", {})
<4> auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
<5> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<6> try:
<7> approach = current_app.config[CONFIG_CHAT_APPROACH]
<8> result = await approach.run(
<9> request_json["messages"],
<10> stream=request_json.get("stream", False),
<11> context=context,
<12> session_state=request_json.get("session_state"),
<13> )
<14> if isinstance(result, dict):
<15> return jsonify(result)
<16> else:
<17> response = await make_response(format_as_ndjson(result))
<18> response.timeout = None # type: ignore
<19> response.mimetype = "application/json-lines"
<20> return response
<21> except Exception as error:
<22> return error_response(error, "/chat")
<23>
|
===========unchanged ref 0===========
at: app.backend.app
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_AUTH_CLIENT = "auth_client"
bp = Blueprint("routes", __name__, static_folder="static")
error_response(error: Exception, route: str, status_code: int=500)
format_as_ndjson(r: AsyncGenerator[dict, None]) -> AsyncGenerator[str, None]
===========changed ref 0===========
# module: app.backend.app
def format_as_ndjson(r: AsyncGenerator[dict, None]) -> AsyncGenerator[str, None]:
try:
async for event in r:
+ yield json.dumps(event, ensure_ascii=False, cls=JSONEncoder) + "\n"
- yield json.dumps(event, ensure_ascii=False) + "\n"
+ except Exception as error:
- except Exception as e:
+ logging.exception("Exception while generating response stream: %s", error)
- logging.exception("Exception while generating response stream: %s", e)
+ yield json.dumps(error_dict(error))
- yield json.dumps(error_dict(e))
===========changed ref 1===========
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
if not request.is_json:
return jsonify({"error": "request must be json"}), 415
request_json = await request.get_json()
context = request_json.get("context", {})
auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
try:
+ use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
+ approach: Approach
+ if use_gpt4v and CONFIG_ASK_VISION_APPROACH in current_app.config:
+ approach = cast(Approach, current_app.config[CONFIG_ASK_VISION_APPROACH])
+ else:
+ approach = cast(Approach, current_app.config[CONFIG_ASK_APPROACH])
- approach = current_app.config[CONFIG_ASK_APPROACH]
r = await approach.run(
request_json["messages"], context=context, session_state=request_json.get("session_state")
)
return jsonify(r)
except Exception as error:
return error_response(error, "/ask")
===========changed ref 2===========
# module: app.backend.app
+ CONFIG_OPENAI_TOKEN = "openai_token"
+ CONFIG_CREDENTIAL = "azure_credential"
CONFIG_ASK_APPROACH = "ask_approach"
+ CONFIG_ASK_VISION_APPROACH = "ask_vision_approach"
+ CONFIG_CHAT_VISION_APPROACH = "chat_vision_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_BLOB_CONTAINER_CLIENT = "blob_container_client"
CONFIG_AUTH_CLIENT = "auth_client"
+ CONFIG_GPT4V_DEPLOYED = "gpt4v_deployed"
CONFIG_SEARCH_CLIENT = "search_client"
CONFIG_OPENAI_CLIENT = "openai_client"
ERROR_MESSAGE = """The app encountered an error processing your request.
If you are an administrator of the app, view the full error in the logs. See aka.ms/appservice-logs for more information.
Error type: {error_type}
"""
ERROR_MESSAGE_FILTER = """Your message contains content that was flagged by the OpenAI content filter."""
bp = Blueprint("routes", __name__, static_folder="static")
# Fix Windows registry issue with mimetypes
mimetypes.add_type("application/javascript", ".js")
mimetypes.add_type("text/css", ".css")
===========changed ref 3===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 4===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 5===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 6===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 7===========
+ # module: tests
+
+
===========changed ref 8===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 9===========
+ # module: tests.mocks
+
+
===========changed ref 10===========
# module: tests.test_content_file
-
-
===========changed ref 11===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 12===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 13===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 23===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 24===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 26===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 27===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 28===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 29===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 30===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 31===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
|
scripts.prepdocs/setup_file_strategy
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<5>:<add> store_page_images=args.searchimages,
|
# module: scripts.prepdocs
+ def setup_file_strategy(credential: AsyncTokenCredential, args: Any) -> FileStrategy:
- def setup_file_strategy(credential: AsyncTokenCredential, args: Any) -> FileStrategy:
<0> storage_creds = credential if is_key_empty(args.storagekey) else args.storagekey
<1> blob_manager = BlobManager(
<2> endpoint=f"https://{args.storageaccount}.blob.core.windows.net",
<3> container=args.container,
<4> credential=storage_creds,
<5> verbose=args.verbose,
<6> )
<7>
<8> pdf_parser: PdfParser
<9> if args.localpdfparser:
<10> pdf_parser = LocalPdfParser()
<11> else:
<12> # check if Azure Document Intelligence credentials are provided
<13> if args.formrecognizerservice is None:
<14> print(
<15> "Error: Azure Document Intelligence service is not provided. Please provide --formrecognizerservice or use --localpdfparser for local pypdf parser."
<16> )
<17> exit(1)
<18> formrecognizer_creds: Union[AsyncTokenCredential, AzureKeyCredential] = (
<19> credential if is_key_empty(args.formrecognizerkey) else AzureKeyCredential(args.formrecognizerkey)
<20> )
<21> pdf_parser = DocumentAnalysisPdfParser(
<22> endpoint=f"https://{args.formrecognizerservice}.cognitiveservices.azure.com/",
<23> credential=formrecognizer_creds,
<24> verbose=args.verbose,
<25> )
<26>
<27> use_vectors = not args.novectors
<28> embeddings: Optional[OpenAIEmbeddings] = None
<29> if use_vectors and args.openaihost != "openai":
<30> azure_open_ai_credential: Union[AsyncTokenCredential, AzureKeyCredential] = (
<31> credential if is_key_empty(args.openaikey) else AzureKeyCredential(args.openaikey)
<32> )
<33> embeddings = AzureOpenAIEmbeddingService(
<34> open_ai_</s>
|
===========below chunk 0===========
# module: scripts.prepdocs
+ def setup_file_strategy(credential: AsyncTokenCredential, args: Any) -> FileStrategy:
- def setup_file_strategy(credential: AsyncTokenCredential, args: Any) -> FileStrategy:
# offset: 1
open_ai_deployment=args.openaideployment,
open_ai_model_name=args.openaimodelname,
credential=azure_open_ai_credential,
disable_batch=args.disablebatchvectors,
verbose=args.verbose,
)
elif use_vectors:
embeddings = OpenAIEmbeddingService(
open_ai_model_name=args.openaimodelname,
credential=args.openaikey,
organization=args.openaiorg,
disable_batch=args.disablebatchvectors,
verbose=args.verbose,
)
print("Processing files...")
list_file_strategy: ListFileStrategy
if args.datalakestorageaccount:
adls_gen2_creds = credential if is_key_empty(args.datalakekey) else args.datalakekey
print(f"Using Data Lake Gen2 Storage Account {args.datalakestorageaccount}")
list_file_strategy = ADLSGen2ListFileStrategy(
data_lake_storage_account=args.datalakestorageaccount,
data_lake_filesystem=args.datalakefilesystem,
data_lake_path=args.datalakepath,
credential=adls_gen2_creds,
verbose=args.verbose,
)
else:
print(f"Using local files in {args.files}")
list_file_strategy = LocalListFileStrategy(path_pattern=args.files, verbose=args.verbose)
if args.removeall:
document_action = DocumentAction.RemoveAll
elif args.remove:
document_action = DocumentAction.Remove
else:
document_action = DocumentAction.Add
return FileStrategy(
list_file_strategy=list_file_strategy,
blob_manager=blob_manager</s>
===========below chunk 1===========
# module: scripts.prepdocs
+ def setup_file_strategy(credential: AsyncTokenCredential, args: Any) -> FileStrategy:
- def setup_file_strategy(credential: AsyncTokenCredential, args: Any) -> FileStrategy:
# offset: 2
<s>
return FileStrategy(
list_file_strategy=list_file_strategy,
blob_manager=blob_manager,
pdf_parser=pdf_parser,
text_splitter=TextSplitter(),
document_action=document_action,
embeddings=embeddings,
search_analyzer_name=args.searchanalyzername,
use_acls=args.useacls,
category=args.category,
)
===========unchanged ref 0===========
at: prepdocslib.blobmanager
BlobManager(endpoint: str, container: str, credential: Union[AsyncTokenCredential, str], store_page_images: bool=False, verbose: bool=False)
at: prepdocslib.embeddings
OpenAIEmbeddings(open_ai_model_name: str, disable_batch: bool=False, verbose: bool=False)
AzureOpenAIEmbeddingService(open_ai_service: str, open_ai_deployment: str, open_ai_model_name: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], disable_batch: bool=False, verbose: bool=False)
OpenAIEmbeddingService(open_ai_model_name: str, credential: str, organization: Optional[str]=None, disable_batch: bool=False, verbose: bool=False)
ImageEmbeddings(credential: str, endpoint: str, verbose: bool=False)
at: prepdocslib.filestrategy
FileStrategy(list_file_strategy: ListFileStrategy, blob_manager: BlobManager, pdf_parser: PdfParser, text_splitter: TextSplitter, document_action: DocumentAction=DocumentAction.Add, embeddings: Optional[OpenAIEmbeddings]=None, image_embeddings: Optional[ImageEmbeddings]=None, search_analyzer_name: Optional[str]=None, use_acls: bool=False, category: Optional[str]=None)
at: prepdocslib.listfilestrategy
ListFileStrategy()
ADLSGen2ListFileStrategy(data_lake_storage_account: str, data_lake_filesystem: str, data_lake_path: str, credential: Union[AsyncTokenCredential, str], verbose: bool=False)
at: prepdocslib.pdfparser
PdfParser()
LocalPdfParser()
DocumentAnalysisPdfParser(endpoint: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], model_id="prebuilt-layout", verbose: bool=False)
at: scripts.prepdocs
is_key_empty(key)
args = parser.parse_args()
===========changed ref 0===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 1===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 2===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 3===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 4===========
+ # module: tests
+
+
===========changed ref 5===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 6===========
+ # module: tests.mocks
+
+
===========changed ref 7===========
# module: tests.test_content_file
-
-
===========changed ref 8===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 10===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 11===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 20===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
|
scripts.prepdocslib.filestrategy/FileStrategy.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<6>:<add> self.image_embeddings = image_embeddings
|
<s> list_file_strategy: ListFileStrategy,
blob_manager: BlobManager,
pdf_parser: PdfParser,
text_splitter: TextSplitter,
document_action: DocumentAction = DocumentAction.Add,
embeddings: Optional[OpenAIEmbeddings] = None,
+ image_embeddings: Optional[ImageEmbeddings] = None,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
category: Optional[str] = None,
):
<0> self.list_file_strategy = list_file_strategy
<1> self.blob_manager = blob_manager
<2> self.pdf_parser = pdf_parser
<3> self.text_splitter = text_splitter
<4> self.document_action = document_action
<5> self.embeddings = embeddings
<6> self.search_analyzer_name = search_analyzer_name
<7> self.use_acls = use_acls
<8> self.category = category
<9>
|
===========unchanged ref 0===========
at: scripts.prepdocslib.blobmanager
BlobManager(endpoint: str, container: str, credential: Union[AsyncTokenCredential, str], store_page_images: bool=False, verbose: bool=False)
at: scripts.prepdocslib.embeddings
OpenAIEmbeddings(open_ai_model_name: str, disable_batch: bool=False, verbose: bool=False)
ImageEmbeddings(credential: str, endpoint: str, verbose: bool=False)
at: scripts.prepdocslib.filestrategy
DocumentAction()
at: scripts.prepdocslib.listfilestrategy
ListFileStrategy()
at: scripts.prepdocslib.pdfparser
PdfParser()
at: scripts.prepdocslib.textsplitter
TextSplitter(has_image_embeddings, verbose: bool=False)
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ """
+ Class for using image embeddings from Azure AI Vision
+ To learn more, please visit https://learn.microsoft.com/azure/ai-services/computer-vision/how-to/image-retrieval#call-the-vectorize-image-api
+ """
+
===========changed ref 1===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 2===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 3===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 4===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 5===========
+ # module: tests
+
+
===========changed ref 6===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 7===========
+ # module: tests.mocks
+
+
===========changed ref 8===========
# module: tests.test_content_file
-
-
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 11===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 26===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 28===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 29===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 33===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 35===========
# module: app.backend.app
+ @bp.route("/config", methods=["GET"])
+ def config():
+ return jsonify({"showGPT4VOptions": current_app.config[CONFIG_GPT4V_DEPLOYED]})
+
===========changed ref 36===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 37===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 38===========
# module: app.backend.app
+ class JSONEncoder(json.JSONEncoder):
+ def default(self, o):
+ if dataclasses.is_dataclass(o):
+ return dataclasses.asdict(o)
+ return super().default(o)
+
===========changed ref 39===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
|
scripts.prepdocslib.filestrategy/FileStrategy.setup
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<0>:<add> search_manager = SearchManager(
<add> search_info,
<add> self.search_analyzer_name,
<add> self.use_acls,
<add> self.embeddings,
<add> search_images=self.image_embeddings is not None,
<add> )
<del> search_manager = SearchManager(search_info, self.search_analyzer_name, self.use_acls, self.embeddings)
|
# module: scripts.prepdocslib.filestrategy
class FileStrategy(Strategy):
def setup(self, search_info: SearchInfo):
<0> search_manager = SearchManager(search_info, self.search_analyzer_name, self.use_acls, self.embeddings)
<1> await search_manager.create_index()
<2>
|
===========unchanged ref 0===========
at: scripts.prepdocslib.strategy
SearchInfo(endpoint: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], index_name: str, verbose: bool=False)
at: scripts.prepdocslib.strategy.Strategy
setup(self, search_info: SearchInfo)
===========changed ref 0===========
<s> list_file_strategy: ListFileStrategy,
blob_manager: BlobManager,
pdf_parser: PdfParser,
text_splitter: TextSplitter,
document_action: DocumentAction = DocumentAction.Add,
embeddings: Optional[OpenAIEmbeddings] = None,
+ image_embeddings: Optional[ImageEmbeddings] = None,
search_analyzer_name: Optional[str] = None,
use_acls: bool = False,
category: Optional[str] = None,
):
self.list_file_strategy = list_file_strategy
self.blob_manager = blob_manager
self.pdf_parser = pdf_parser
self.text_splitter = text_splitter
self.document_action = document_action
self.embeddings = embeddings
+ self.image_embeddings = image_embeddings
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.category = category
===========changed ref 1===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 2===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 3===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 4===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 5===========
+ # module: tests
+
+
===========changed ref 6===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 7===========
+ # module: tests.mocks
+
+
===========changed ref 8===========
# module: tests.test_content_file
-
-
===========changed ref 9===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 11===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 12===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 13===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 14===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 15===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 21===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 22===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 26===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 28===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 29===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 30===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 31===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 32===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 33===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 34===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 35===========
# module: app.backend.app
+ @bp.route("/config", methods=["GET"])
+ def config():
+ return jsonify({"showGPT4VOptions": current_app.config[CONFIG_GPT4V_DEPLOYED]})
+
===========changed ref 36===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 37===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 38===========
# module: app.backend.app
+ class JSONEncoder(json.JSONEncoder):
+ def default(self, o):
+ if dataclasses.is_dataclass(o):
+ return dataclasses.asdict(o)
+ return super().default(o)
+
===========changed ref 39===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return self.data.pop(0) # This should be a list of dictionaries.
+
|
scripts.prepdocslib.filestrategy/FileStrategy.run
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<12>:<add> image_embeddings: Optional[List[List[float]]]
<del> await search_manager.update_content(sections)
<13>:<add> blob_sas_uris = await self.blob_manager.upload_blob(file)
<del> await self.blob_manager.upload_blob(file)
<14>:<add> if self.image_embeddings and blob_sas_uris:
<add> image_embeddings = await self.image_embeddings.create_embeddings(blob_sas_uris)
<add> await search_manager.update_content(sections, image_embeddings)
|
# module: scripts.prepdocslib.filestrategy
class FileStrategy(Strategy):
def run(self, search_info: SearchInfo):
<0> search_manager = SearchManager(search_info, self.search_analyzer_name, self.use_acls, self.embeddings)
<1> if self.document_action == DocumentAction.Add:
<2> files = self.list_file_strategy.list()
<3> async for file in files:
<4> try:
<5> pages = [page async for page in self.pdf_parser.parse(content=file.content)]
<6> if search_info.verbose:
<7> print(f"Splitting '{file.filename()}' into sections")
<8> sections = [
<9> Section(split_page, content=file, category=self.category)
<10> for split_page in self.text_splitter.split_pages(pages)
<11> ]
<12> await search_manager.update_content(sections)
<13> await self.blob_manager.upload_blob(file)
<14> finally:
<15> if file:
<16> file.close()
<17> elif self.document_action == DocumentAction.Remove:
<18> paths = self.list_file_strategy.list_paths()
<19> async for path in paths:
<20> await self.blob_manager.remove_blob(path)
<21> await search_manager.remove_content(path)
<22> elif self.document_action == DocumentAction.RemoveAll:
<23> await self.blob_manager.remove_blob()
<24> await search_manager.remove_content()
<25>
|
===========unchanged ref 0===========
at: scripts.prepdocslib.blobmanager.BlobManager
upload_blob(file: File) -> Optional[List[str]]
at: scripts.prepdocslib.embeddings.ImageEmbeddings
create_embeddings(blob_urls: List[str]) -> List[List[float]]
at: scripts.prepdocslib.filestrategy
DocumentAction()
at: scripts.prepdocslib.filestrategy.FileStrategy.__init__
self.list_file_strategy = list_file_strategy
self.blob_manager = blob_manager
self.pdf_parser = pdf_parser
self.text_splitter = text_splitter
self.document_action = document_action
self.embeddings = embeddings
self.image_embeddings = image_embeddings
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.category = category
at: scripts.prepdocslib.filestrategy.FileStrategy.setup
search_manager = SearchManager(
search_info,
self.search_analyzer_name,
self.use_acls,
self.embeddings,
search_images=self.image_embeddings is not None,
)
at: scripts.prepdocslib.listfilestrategy.ListFileStrategy
list() -> AsyncGenerator[File, None]
at: scripts.prepdocslib.pdfparser.PdfParser
parse(content: IO) -> AsyncGenerator[Page, None]
at: scripts.prepdocslib.searchmanager
Section(split_page: SplitPage, content: File, category: Optional[str]=None)
SearchManager(search_info: SearchInfo, search_analyzer_name: Optional[str]=None, use_acls: bool=False, embeddings: Optional[OpenAIEmbeddings]=None, search_images: bool=False)
at: scripts.prepdocslib.searchmanager.SearchManager
create_index()
===========unchanged ref 1===========
update_content(sections: List[Section], image_embeddings: Optional[List[List[float]]]=None)
at: scripts.prepdocslib.strategy
SearchInfo(endpoint: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], index_name: str, verbose: bool=False)
at: scripts.prepdocslib.strategy.SearchInfo.__init__
self.verbose = verbose
at: scripts.prepdocslib.strategy.Strategy
run(self, search_info: SearchInfo)
at: scripts.prepdocslib.textsplitter.TextSplitter
split_pages(pages: List[Page]) -> Generator[SplitPage, None, None]
at: typing
List = _alias(list, 1, inst=False, name='List')
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def create_embeddings(self, blob_urls: List[str]) -> List[List[float]]:
+ headers = {"Ocp-Apim-Subscription-Key": self.credential}
+ params = {"api-version": "2023-02-01-preview", "modelVersion": "latest"}
+ endpoint = urljoin(self.endpoint, "computervision/retrieval:vectorizeImage")
+ embeddings: List[List[float]] = []
+ async with aiohttp.ClientSession(headers=headers) as session:
+ for blob_url in blob_urls:
+ async for attempt in AsyncRetrying(
+ retry=retry_if_exception_type(Exception),
+ wait=wait_random_exponential(min=15, max=60),
+ stop=stop_after_attempt(15),
+ before_sleep=self.before_retry_sleep,
+ ):
+ with attempt:
+ body = {"url": blob_url}
+ async with session.post(url=endpoint, params=params, json=body) as resp:
+ resp_json = await resp.json()
+ embeddings.append(resp_json["vector"])
+
+ return embeddings
+
===========changed ref 1===========
# module: scripts.prepdocslib.searchmanager
class SearchManager:
+ def update_content(self, sections: List[Section], image_embeddings: Optional[List[List[float]]] = None):
- def update_content(self, sections: List[Section]):
MAX_BATCH_SIZE = 1000
section_batches = [sections[i : i + MAX_BATCH_SIZE] for i in range(0, len(sections), MAX_BATCH_SIZE)]
async with self.search_info.create_search_client() as search_client:
for batch_index, batch in enumerate(section_batches):
documents = [
{
"id": f"{section.content.filename_to_id()}-page-{section_index + batch_index * MAX_BATCH_SIZE}",
"content": section.split_page.text,
"category": section.category,
+ "sourcepage": BlobManager.blob_image_name_from_file_page(
+ filename=section.content.filename(), page=section.split_page.page_num
+ )
+ if image_embeddings
+ else BlobManager.sourcepage_from_file_page(
- "sourcepage": BlobManager.sourcepage_from_file_page(
filename=section.content.filename(), page=section.split_page.page_num
),
"sourcefile": section.content.filename(),
**section.content.acls,
}
for section_index, section in enumerate(batch)
]
if self.embeddings:
embeddings = await self.embeddings.create_embeddings(
texts=[section.split_page.text for section in batch]
)
for i, document in enumerate(documents):
document["embedding"] = embeddings[i]
+ if image_embeddings:
+ for i, (document, section) in enumerate(zip(documents, batch)):
+ document["imageEmbedding"] = image_embeddings[section.split_page.page_num]
await search_client.upload_documents(documents)
|
tests.conftest/mock_openai_chatcompletion
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
# module: tests.conftest
@pytest.fixture
def mock_openai_chatcompletion(monkeypatch):
<0> class AsyncChatCompletionIterator:
<1> def __init__(self, answer: str):
<2> chunk_id = "test-id"
<3> model = "gpt-35-turbo"
<4> self.responses = [
<5> {"object": "chat.completion.chunk", "choices": [], "id": chunk_id, "model": model, "created": 1},
<6> {
<7> "object": "chat.completion.chunk",
<8> "choices": [{"delta": {"role": "assistant"}, "index": 0, "finish_reason": None}],
<9> "id": chunk_id,
<10> "model": model,
<11> "created": 1,
<12> },
<13> ]
<14> # Split at << to simulate chunked responses
<15> if answer.find("<<") > -1:
<16> parts = answer.split("<<")
<17> self.responses.append(
<18> {
<19> "object": "chat.completion.chunk",
<20> "choices": [
<21> {
<22> "delta": {"role": "assistant", "content": parts[0] + "<<"},
<23> "index": 0,
<24> "finish_reason": None,
<25> }
<26> ],
<27> "id": chunk_id,
<28> "model": model,
<29> "created": 1,
<30> }
<31> )
<32> self.responses.append(
<33> {
<34> "object": "chat.completion.chunk",
<35> "choices": [
<36> {"delta": {"role": "assistant", "content": parts[1]}, "index": 0, "finish_reason": None}
<37> ],
<38> "id": chunk_id,
<39> "model": model,
<40> "created": 1,
<41> }
<42> )
<43> self.responses.append(
<44> {
<45> "object": "chat.completion.chunk",
<46> "choices": [{"delta": {"role": None, "content":</s>
|
===========below chunk 0===========
# module: tests.conftest
@pytest.fixture
def mock_openai_chatcompletion(monkeypatch):
# offset: 1
"id": chunk_id,
"model": model,
"created": 1,
}
)
else:
self.responses.append(
{
"object": "chat.completion.chunk",
"choices": [{"delta": {"content": answer}, "index": 0, "finish_reason": None}],
"id": chunk_id,
"model": model,
"created": 1,
}
)
def __aiter__(self):
return self
async def __anext__(self):
if self.responses:
return ChatCompletionChunk.model_validate(self.responses.pop(0))
else:
raise StopAsyncIteration
async def mock_acreate(*args, **kwargs):
messages = kwargs["messages"]
if messages[-1]["content"] == "Generate search query for: What is the capital of France?":
answer = "capital of France"
else:
answer = "The capital of France is Paris. [Benefit_Options-2.pdf]."
if messages[0]["content"].find("Generate 3 very brief follow-up questions") > -1:
answer = "The capital of France is Paris. [Benefit_Options-2.pdf]. <<What is the capital of Spain?>>"
if "stream" in kwargs and kwargs["stream"] is True:
return AsyncChatCompletionIterator(answer)
else:
return ChatCompletion(
object="chat.completion",
choices=[
Choice(
message=ChatCompletionMessage(role="assistant", content=answer), finish_reason="stop", index=0
)
],
id="test-123",
created=0,
model="test-model",
)
def patch(openai_client):
monkeypatch.setattr(openai_client.chat.completions, "create", mock_acreate)
</s>
===========below chunk 1===========
# module: tests.conftest
@pytest.fixture
def mock_openai_chatcompletion(monkeypatch):
# offset: 2
<s>client):
monkeypatch.setattr(openai_client.chat.completions, "create", mock_acreate)
return patch
===========changed ref 0===========
# module: tests.conftest
- class MockAzureCredential(AsyncTokenCredential):
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 1===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_get_secret(monkeypatch):
+ monkeypatch.setattr(SecretClient, "get_secret", MockKeyVaultSecretClient().get_secret)
+
===========changed ref 2===========
# module: tests.conftest
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 3===========
# module: tests.conftest
+ def mock_search(self, *args, **kwargs):
+ self.filter = kwargs.get("filter")
+ return MockAsyncSearchResultsIterator(kwargs.get("search_text"), kwargs.get("vector_queries"))
+
===========changed ref 4===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_compute_embeddings_call(monkeypatch):
+ def mock_post(*args, **kwargs):
+ if kwargs.get("url").endswith("computervision/retrieval:vectorizeText"):
+ return mock_computervision_response()
+ else:
+ raise Exception("Unexpected URL for mock call to ClientSession.post()")
+
+ monkeypatch.setattr(aiohttp.ClientSession, "post", mock_post)
+
===========changed ref 5===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 6===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 7===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 8===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 9===========
+ # module: tests
+
+
===========changed ref 10===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 11===========
+ # module: tests.mocks
+
+
===========changed ref 12===========
# module: tests.test_content_file
-
-
===========changed ref 13===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 14===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 15===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 25===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 26===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 28===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 29===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 30===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 31===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 32===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 33===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 34===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
|
|
tests.conftest/mock_acs_search
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<0>:<del> class Caption:
<1>:<del> def __init__(self, text):
<2>:<del> self.text = text
<3>:<del>
<4>:<del> class AsyncSearchResultsIterator:
<5>:<del> def __init__(self):
<6>:<del> self.num = 1
<7>:<del>
<8>:<del> def __aiter__(self):
<9>:<del> return self
<10>:<del>
<11>:<del> async def __anext__(self):
<12>:<del> if self.num == 1:
<13>:<del> self.num = 0
<14>:<del> return {
<15>:<del> "sourcepage": "Benefit_Options-2.pdf",
<16>:<del> "sourcefile": "Benefit_Options.pdf",
<17>:<del> "content": "There is a whistleblower policy.",
<18>:<del> "embeddings": [],
<19>:<del> "category": None,
<20>:<del> "id": "file-Benefit_Options_pdf-42656E656669745F4F7074696F6E732E706466-page-2",
<21>:<del> "@search.score": 0.03279569745063782,
<22>:<del> "@search.reranker_score": 3.4577205181121826,
<23>:<del> "@search.highlights": None,
<24>:<del> "@search.captions": [Caption("Caption: A whistleblower policy.")],
<25>:<del> }
<26>:<del> else:
<27>:<del> raise StopAsyncIteration
<28>:<del>
<29>:<del> async def mock_search(*args, **kwargs):
<30>:<del> return AsyncSearchResultsIterator()
<31>:<del>
<32>:<add> monkeypatch.setattr(SearchClient, "search", mock_search)
|
# module: tests.conftest
@pytest.fixture
def mock_acs_search(monkeypatch):
<0> class Caption:
<1> def __init__(self, text):
<2> self.text = text
<3>
<4> class AsyncSearchResultsIterator:
<5> def __init__(self):
<6> self.num = 1
<7>
<8> def __aiter__(self):
<9> return self
<10>
<11> async def __anext__(self):
<12> if self.num == 1:
<13> self.num = 0
<14> return {
<15> "sourcepage": "Benefit_Options-2.pdf",
<16> "sourcefile": "Benefit_Options.pdf",
<17> "content": "There is a whistleblower policy.",
<18> "embeddings": [],
<19> "category": None,
<20> "id": "file-Benefit_Options_pdf-42656E656669745F4F7074696F6E732E706466-page-2",
<21> "@search.score": 0.03279569745063782,
<22> "@search.reranker_score": 3.4577205181121826,
<23> "@search.highlights": None,
<24> "@search.captions": [Caption("Caption: A whistleblower policy.")],
<25> }
<26> else:
<27> raise StopAsyncIteration
<28>
<29> async def mock_search(*args, **kwargs):
<30> return AsyncSearchResultsIterator()
<31>
<32> monkeypatch.setattr(SearchClient, "search", mock_search)
<33>
|
===========changed ref 0===========
# module: tests.conftest
- class MockAzureCredential(AsyncTokenCredential):
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 1===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_get_secret(monkeypatch):
+ monkeypatch.setattr(SecretClient, "get_secret", MockKeyVaultSecretClient().get_secret)
+
===========changed ref 2===========
# module: tests.conftest
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 3===========
# module: tests.conftest
+ def mock_search(self, *args, **kwargs):
+ self.filter = kwargs.get("filter")
+ return MockAsyncSearchResultsIterator(kwargs.get("search_text"), kwargs.get("vector_queries"))
+
===========changed ref 4===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_compute_embeddings_call(monkeypatch):
+ def mock_post(*args, **kwargs):
+ if kwargs.get("url").endswith("computervision/retrieval:vectorizeText"):
+ return mock_computervision_response()
+ else:
+ raise Exception("Unexpected URL for mock call to ClientSession.post()")
+
+ monkeypatch.setattr(aiohttp.ClientSession, "post", mock_post)
+
===========changed ref 5===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 6===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 7===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 8===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 9===========
+ # module: tests
+
+
===========changed ref 10===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 11===========
+ # module: tests.mocks
+
+
===========changed ref 12===========
# module: tests.test_content_file
-
-
===========changed ref 13===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 14===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 15===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 16===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 25===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 26===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 27===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 28===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 29===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 30===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 31===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 32===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 33===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 34===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 35===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 36===========
# module: app.backend.approaches.approach
+ @dataclass
+ class ThoughtStep:
+ title: str
+ description: Optional[Any]
+ props: Optional[dict[str, Any]] = None
+
===========changed ref 37===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
===========changed ref 38===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ def extract_followup_questions(self, content: str):
+ return content.split("<<")[0], re.findall(r"<<([^>>]+)>>", content)
+
===========changed ref 39===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __anext__(self):
+ if not self.data:
+ raise StopAsyncIteration
+ return MockAsyncPageIterator(self.data.pop(0))
+
===========changed ref 40===========
# module: app.backend.app
+ @bp.route("/config", methods=["GET"])
+ def config():
+ return jsonify({"showGPT4VOptions": current_app.config[CONFIG_GPT4V_DEPLOYED]})
+
===========changed ref 41===========
+ # module: tests.mocks
+ class MockCaption:
+ def __init__(self, text, highlights=None, additional_properties=None):
+ self.text = text
+ self.highlights = highlights or []
+ self.additional_properties = additional_properties or {}
+
===========changed ref 42===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
+ @classmethod
+ def blob_image_name_from_file_page(cls, filename, page=0) -> str:
+ return os.path.splitext(os.path.basename(filename))[0] + f"-{page}" + ".png"
+
===========changed ref 43===========
# module: app.backend.app
+ class JSONEncoder(json.JSONEncoder):
+ def default(self, o):
+ if dataclasses.is_dataclass(o):
+ return dataclasses.asdict(o)
+ return super().default(o)
+
|
tests.conftest/mock_acs_search_filter
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<0>:<del> class AsyncSearchResultsIterator:
<1>:<del> def __init__(self):
<2>:<del> self.num = 1
<3>:<del>
<4>:<del> def __aiter__(self):
<5>:<del> return self
<6>:<del>
<7>:<del> async def __anext__(self):
<8>:<del> raise StopAsyncIteration
<9>:<del>
<10>:<del> async def mock_search(self, *args, **kwargs):
<11>:<del> self.filter = kwargs.get("filter")
<12>:<del> return AsyncSearchResultsIterator()
<13>:<del>
|
# module: tests.conftest
@pytest.fixture
def mock_acs_search_filter(monkeypatch):
<0> class AsyncSearchResultsIterator:
<1> def __init__(self):
<2> self.num = 1
<3>
<4> def __aiter__(self):
<5> return self
<6>
<7> async def __anext__(self):
<8> raise StopAsyncIteration
<9>
<10> async def mock_search(self, *args, **kwargs):
<11> self.filter = kwargs.get("filter")
<12> return AsyncSearchResultsIterator()
<13>
<14> monkeypatch.setattr(SearchClient, "search", mock_search)
<15>
|
===========changed ref 0===========
# module: tests.conftest
- class MockAzureCredential(AsyncTokenCredential):
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 1===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_get_secret(monkeypatch):
+ monkeypatch.setattr(SecretClient, "get_secret", MockKeyVaultSecretClient().get_secret)
+
===========changed ref 2===========
# module: tests.conftest
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 3===========
# module: tests.conftest
+ def mock_search(self, *args, **kwargs):
+ self.filter = kwargs.get("filter")
+ return MockAsyncSearchResultsIterator(kwargs.get("search_text"), kwargs.get("vector_queries"))
+
===========changed ref 4===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_compute_embeddings_call(monkeypatch):
+ def mock_post(*args, **kwargs):
+ if kwargs.get("url").endswith("computervision/retrieval:vectorizeText"):
+ return mock_computervision_response()
+ else:
+ raise Exception("Unexpected URL for mock call to ClientSession.post()")
+
+ monkeypatch.setattr(aiohttp.ClientSession, "post", mock_post)
+
===========changed ref 5===========
# module: tests.conftest
@pytest.fixture
def mock_acs_search(monkeypatch):
- class Caption:
- def __init__(self, text):
- self.text = text
-
- class AsyncSearchResultsIterator:
- def __init__(self):
- self.num = 1
-
- def __aiter__(self):
- return self
-
- async def __anext__(self):
- if self.num == 1:
- self.num = 0
- return {
- "sourcepage": "Benefit_Options-2.pdf",
- "sourcefile": "Benefit_Options.pdf",
- "content": "There is a whistleblower policy.",
- "embeddings": [],
- "category": None,
- "id": "file-Benefit_Options_pdf-42656E656669745F4F7074696F6E732E706466-page-2",
- "@search.score": 0.03279569745063782,
- "@search.reranker_score": 3.4577205181121826,
- "@search.highlights": None,
- "@search.captions": [Caption("Caption: A whistleblower policy.")],
- }
- else:
- raise StopAsyncIteration
-
- async def mock_search(*args, **kwargs):
- return AsyncSearchResultsIterator()
-
+ monkeypatch.setattr(SearchClient, "search", mock_search)
monkeypatch.setattr(SearchClient, "search", mock_search)
===========changed ref 6===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 7===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 8===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 9===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 10===========
+ # module: tests
+
+
===========changed ref 11===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 12===========
+ # module: tests.mocks
+
+
===========changed ref 13===========
# module: tests.test_content_file
-
-
===========changed ref 14===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 15===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 16===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 17===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 18===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 19===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
===========changed ref 26===========
+ # module: tests.test_chatvisionapproach
+ @pytest.fixture
+ def openai_client():
+ return MockOpenAIClient()
+
===========changed ref 27===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def __init__(self):
+ self.embeddings = self
+
===========changed ref 28===========
+ # module: tests.mocks
+ class MockResponse:
+ def json(self):
+ return json.loads(self.text)
+
===========changed ref 29===========
+ # module: tests.mocks
+ class MockKeyVaultSecretClient:
+ def get_secret(self, secret_name):
+ return MockKeyVaultSecret("mysecret")
+
===========changed ref 30===========
+ # module: tests.mocks
+ class MockAzureCredential(AsyncTokenCredential):
+ def get_token(self, uri):
+ return MockToken("", 9999999999, "")
+
===========changed ref 31===========
# module: tests.test_content_file
- class MockAzureCredential:
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 32===========
+ # module: tests.mocks
+ class MockResponse:
+ def __init__(self, text, status):
+ self.text = text
+ self.status = status
+
===========changed ref 33===========
# module: tests.test_content_file
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 34===========
+ # module: tests.mocks
+ MockToken = namedtuple("MockToken", ["token", "expires_on", "value"])
+
===========changed ref 35===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def __init__(self, credential: str, endpoint: str, verbose: bool = False):
+ self.credential = credential
+ self.endpoint = endpoint
+ self.verbose = verbose
+
===========changed ref 36===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_client(self) -> AsyncOpenAI:
- return AsyncOpenAI(api_key=self.credential, organization=self.organization)
-
===========changed ref 37===========
# module: app.backend.approaches.approach
+ @dataclass
+ class ThoughtStep:
+ title: str
+ description: Optional[Any]
+ props: Optional[dict[str, Any]] = None
+
===========changed ref 38===========
# module: scripts.prepdocslib.embeddings
+ class ImageEmbeddings:
+ def before_retry_sleep(self, retry_state):
+ if self.verbose:
+ print("Rate limited on the Vision embeddings API, sleeping before retrying...")
+
|
tests.conftest/client
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
36015895f14c8569e3a3989fcf5744a9b5bd4845
|
Integrate GPT4-vision support (#1056)
|
<3>:<add> test_app.app.config.update({"TESTING": True})
<del> quart_app.config.update({"TESTING": True})
|
<s>ftest
@pytest_asyncio.fixture()
+ async def client(
+ monkeypatch,
+ mock_env,
+ mock_openai_chatcompletion,
+ mock_openai_embedding,
+ mock_acs_search,
+ mock_blob_container_client,
+ mock_compute_embeddings_call,
+ ):
- async def client(monkeypatch, mock_env, mock_openai_chatcompletion, mock_openai_embedding, mock_acs_search, request):
<0> quart_app = app.create_app()
<1>
<2> async with quart_app.test_app() as test_app:
<3> quart_app.config.update({"TESTING": True})
<4> mock_openai_chatcompletion(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
<5> mock_openai_embedding(test_app.app.config[app.CONFIG_OPENAI_CLIENT])
<6> yield test_app.test_client()
<7>
|
===========changed ref 0===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_blob_container_client(monkeypatch):
+ monkeypatch.setattr(ContainerClient, "get_blob_client", lambda *args, **kwargs: MockBlobClient())
+
===========changed ref 1===========
# module: tests.conftest
@pytest.fixture
def mock_acs_search_filter(monkeypatch):
- class AsyncSearchResultsIterator:
- def __init__(self):
- self.num = 1
-
- def __aiter__(self):
- return self
-
- async def __anext__(self):
- raise StopAsyncIteration
-
- async def mock_search(self, *args, **kwargs):
- self.filter = kwargs.get("filter")
- return AsyncSearchResultsIterator()
-
monkeypatch.setattr(SearchClient, "search", mock_search)
===========changed ref 2===========
# module: tests.conftest
- class MockAzureCredential(AsyncTokenCredential):
- def get_token(self, uri):
- return MockToken("mock_token", 9999999999)
-
===========changed ref 3===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_get_secret(monkeypatch):
+ monkeypatch.setattr(SecretClient, "get_secret", MockKeyVaultSecretClient().get_secret)
+
===========changed ref 4===========
# module: tests.conftest
- MockToken = namedtuple("MockToken", ["token", "expires_on"])
-
===========changed ref 5===========
# module: tests.conftest
+ def mock_search(self, *args, **kwargs):
+ self.filter = kwargs.get("filter")
+ return MockAsyncSearchResultsIterator(kwargs.get("search_text"), kwargs.get("vector_queries"))
+
===========changed ref 6===========
# module: tests.conftest
+ @pytest.fixture
+ def mock_compute_embeddings_call(monkeypatch):
+ def mock_post(*args, **kwargs):
+ if kwargs.get("url").endswith("computervision/retrieval:vectorizeText"):
+ return mock_computervision_response()
+ else:
+ raise Exception("Unexpected URL for mock call to ClientSession.post()")
+
+ monkeypatch.setattr(aiohttp.ClientSession, "post", mock_post)
+
===========changed ref 7===========
# module: tests.conftest
envs = [
{
"OPENAI_HOST": "openai",
"OPENAI_API_KEY": "secretkey",
"OPENAI_ORGANIZATION": "organization",
},
{
"OPENAI_HOST": "azure",
"AZURE_OPENAI_SERVICE": "test-openai-service",
"AZURE_OPENAI_CHATGPT_DEPLOYMENT": "test-chatgpt",
"AZURE_OPENAI_EMB_DEPLOYMENT": "test-ada",
+ "AZURE_OPENAI_GPT4V_MODEL": "gpt-4",
+ "VISION_SECRET_NAME": "mysecret",
+ "VISION_ENDPOINT": "https://testvision.cognitiveservices.azure.com/",
+ "AZURE_KEY_VAULT_NAME": "mykeyvault",
},
]
auth_envs = [
{
"OPENAI_HOST": "azure",
"AZURE_OPENAI_SERVICE": "test-openai-service",
"AZURE_OPENAI_CHATGPT_DEPLOYMENT": "test-chatgpt",
"AZURE_OPENAI_EMB_DEPLOYMENT": "test-ada",
"AZURE_USE_AUTHENTICATION": "true",
"AZURE_SERVER_APP_ID": "SERVER_APP",
"AZURE_SERVER_APP_SECRET": "SECRET",
"AZURE_CLIENT_APP_ID": "CLIENT_APP",
"AZURE_TENANT_ID": "TENANT_ID",
},
]
===========changed ref 8===========
# module: tests.conftest
@pytest.fixture
def mock_acs_search(monkeypatch):
- class Caption:
- def __init__(self, text):
- self.text = text
-
- class AsyncSearchResultsIterator:
- def __init__(self):
- self.num = 1
-
- def __aiter__(self):
- return self
-
- async def __anext__(self):
- if self.num == 1:
- self.num = 0
- return {
- "sourcepage": "Benefit_Options-2.pdf",
- "sourcefile": "Benefit_Options.pdf",
- "content": "There is a whistleblower policy.",
- "embeddings": [],
- "category": None,
- "id": "file-Benefit_Options_pdf-42656E656669745F4F7074696F6E732E706466-page-2",
- "@search.score": 0.03279569745063782,
- "@search.reranker_score": 3.4577205181121826,
- "@search.highlights": None,
- "@search.captions": [Caption("Caption: A whistleblower policy.")],
- }
- else:
- raise StopAsyncIteration
-
- async def mock_search(*args, **kwargs):
- return AsyncSearchResultsIterator()
-
+ monkeypatch.setattr(SearchClient, "search", mock_search)
monkeypatch.setattr(SearchClient, "search", mock_search)
===========changed ref 9===========
+ # module: app.backend.approaches.chatreadretrievereadvision
+
+
===========changed ref 10===========
+ # module: app.backend.approaches.chatapproach
+
+
===========changed ref 11===========
+ # module: tests.test_chatvisionapproach
+
+
===========changed ref 12===========
+ # module: app.backend.approaches.retrievethenreadvision
+
+
===========changed ref 13===========
+ # module: tests
+
+
===========changed ref 14===========
+ # module: app.backend.core.imageshelper
+
+
===========changed ref 15===========
+ # module: tests.mocks
+
+
===========changed ref 16===========
# module: tests.test_content_file
-
-
===========changed ref 17===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @abstractmethod
+ async def run_until_final_call(self, history, overrides, auth_claims, should_stream) -> tuple:
+ pass
+
===========changed ref 18===========
+ # module: app.backend.approaches.chatapproach
+ class ChatApproach(Approach, ABC):
+ @property
+ @abstractmethod
+ def system_message_chat_conversation(self) -> str:
+ pass
+
===========changed ref 19===========
+ # module: tests.test_chatvisionapproach
+ class MockOpenAIClient:
+ def create(self, *args, **kwargs):
+ pass
+
===========changed ref 20===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aexit__(self, exc_type, exc, tb):
+ pass
+
===========changed ref 21===========
+ # module: tests.mocks
+ class MockResponse:
+ def __aenter__(self):
+ return self
+
===========changed ref 22===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def by_page(self):
+ return self
+
===========changed ref 23===========
+ # module: tests.mocks
+ class MockAsyncSearchResultsIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 24===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __aiter__(self):
+ return self
+
===========changed ref 25===========
+ # module: tests.mocks
+ class MockResponse:
+ def text(self):
+ return self._text
+
===========changed ref 26===========
+ # module: tests.mocks
+ class MockBlobClient:
+ def download_blob(self):
+ return MockBlob()
+
===========changed ref 27===========
+ # module: tests.mocks
+ class MockAsyncPageIterator:
+ def __init__(self, data):
+ self.data = data
+
===========changed ref 28===========
+ # module: tests.mocks
+ class MockKeyVaultSecret:
+ def __init__(self, value):
+ self.value = value
+
|
tests.test_chatvisionapproach/chat_approach
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> auth_helper=AuthenticationHelper(
<add> search_index=MockSearchIndex,
<add> use_authentication=True,
<add> server_app_id="SERVER_APP",
<add> server_app_secret="SERVER_SECRET",
<add> client_app_id="CLIENT_APP",
<add> tenant_id="TENANT_ID",
<add> require_access_control=None,
<add> ),
|
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
<0> return ChatReadRetrieveReadVisionApproach(
<1> search_client=None,
<2> openai_client=openai_client,
<3> blob_container_client=None,
<4> vision_endpoint="endpoint",
<5> vision_key="key",
<6> gpt4v_deployment="gpt-4v",
<7> gpt4v_model="gpt-4v",
<8> embedding_deployment="embeddings",
<9> embedding_model="text-",
<10> sourcepage_field="",
<11> content_field="",
<12> query_language="en-us",
<13> query_speller="lexicon",
<14> )
<15>
|
===========unchanged ref 0===========
at: _pytest.fixtures
fixture(fixture_function: FixtureFunction, *, scope: "Union[_ScopeName, Callable[[str, Config], _ScopeName]]"=..., params: Optional[Iterable[object]]=..., autouse: bool=..., ids: Optional[
Union[Sequence[Optional[object]], Callable[[Any], Optional[object]]]
]=..., name: Optional[str]=...) -> FixtureFunction
fixture(fixture_function: None=..., *, scope: "Union[_ScopeName, Callable[[str, Config], _ScopeName]]"=..., params: Optional[Iterable[object]]=..., autouse: bool=..., ids: Optional[
Union[Sequence[Optional[object]], Callable[[Any], Optional[object]]]
]=..., name: Optional[str]=None) -> FixtureFunctionMarker
at: tests.conftest
mock_confidential_client_success(monkeypatch)
at: tests.test_chatvisionapproach
MockOpenAIClient()
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 3===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 5===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
|
app.backend.core.authentication/AuthenticationHelper.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<8>:<del> self.token_cache_path = token_cache_path
<9>:<del> if not self.token_cache_path:
<10>:<del> self.temporary_directory = TemporaryDirectory()
<11>:<del> self.token_cache_path = os.path.join(self.temporary_directory.name, "token_cache.bin")
<12>:<del> try:
<13>:<del> persistence = build_encrypted_persistence(location=self.token_cache_path)
<14>:<del> except Exception:
<15>:<del> logging.exception("Encryption unavailable. Opting in to plain text.")
<16>:<del> persistence = FilePersistence(location=self.token_cache_path)
<17>:<add> field_names = [field.name for field in search_index.fields] if search_index else []
<add> self.has_auth_fields = "oids" in field_names and "groups" in field_names
<add> self.require_access_control = require_access_control
<18>:<del> server_app_id,
<19>:<del> authority=self.authority,
<20>:<del> client_credential=server_app_secret,
<21>:<del> token_cache=PersistedTokenCache(persistence),
<22>:<add> server_app_id, authority=self.authority, client_credential=server_app_secret, token_cache=TokenCache()
<23>:<add> else:
<add> self.has_auth_fields = False
<add> self.require_access_control = False
|
# module: app.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
<0> self.use_authentication = use_authentication
<1> self.server_app_id = server_app_id
<2> self.server_app_secret = server_app_secret
<3> self.client_app_id = client_app_id
<4> self.tenant_id = tenant_id
<5> self.authority = f"https://login.microsoftonline.com/{tenant_id}"
<6>
<7> if self.use_authentication:
<8> self.token_cache_path = token_cache_path
<9> if not self.token_cache_path:
<10> self.temporary_directory = TemporaryDirectory()
<11> self.token_cache_path = os.path.join(self.temporary_directory.name, "token_cache.bin")
<12> try:
<13> persistence = build_encrypted_persistence(location=self.token_cache_path)
<14> except Exception:
<15> logging.exception("Encryption unavailable. Opting in to plain text.")
<16> persistence = FilePersistence(location=self.token_cache_path)
<17> self.confidential_client = ConfidentialClientApplication(
<18> server_app_id,
<19> authority=self.authority,
<20> client_credential=server_app_secret,
<21> token_cache=PersistedTokenCache(persistence),
<22> )
<23>
|
===========changed ref 0===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 1===========
+ # module: scripts.auth_common
+
+
===========changed ref 2===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 3===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 4===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 5===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 6===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 7===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 8===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
|
app.backend.core.authentication/AuthenticationHelper.get_auth_setup_for_client
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> "requireAccessControl": self.require_access_control, # Whether or not access control is required to use the application
|
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_setup_for_client(self) -> dict[str, Any]:
<0> # returns MSAL.js settings used by the client app
<1> return {
<2> "useLogin": self.use_authentication, # Whether or not login elements are enabled on the UI
<3> "msalConfig": {
<4> "auth": {
<5> "clientId": self.client_app_id, # Client app id used for login
<6> "authority": self.authority, # Directory to use for login https://learn.microsoft.com/azure/active-directory/develop/msal-client-application-configuration#authority
<7> "redirectUri": "/redirect", # Points to window.location.origin. You must register this URI on Azure Portal/App Registration.
<8> "postLogoutRedirectUri": "/", # Indicates the page to navigate after logout.
<9> "navigateToLoginRequestUrl": False, # If "true", will navigate back to the original request location before processing the auth code response.
<10> },
<11> "cache": {
<12> "cacheLocation": "sessionStorage",
<13> "storeAuthStateInCookie": False,
<14> }, # Configures cache location. "sessionStorage" is more secure, but "localStorage" gives you SSO between tabs. # Set this to "true" if you are having issues on IE11 or Edge
<15> },
<16> "loginRequest": {
<17> # Scopes you add here will be prompted for user consent during sign-in.
<18> # By default, MSAL.js will add OIDC scopes (openid, profile, email) to any login request.
<19> # For more information about OIDC scopes, visit:
<20> # https://docs.microsoft.com/azure/active-directory/develop/v2-permissions-and-consent#openid-connect-scopes
<21> "scopes": [".default"],
<22> # Uncomment the following line to cause a consent dialog to appear on every login
<23> # For more information, please visit https://learn.microsoft.com/azure/active-directory/develop/v2-oauth2-auth-code-</s>
|
===========below chunk 0===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_setup_for_client(self) -> dict[str, Any]:
# offset: 1
# "prompt": "consent"
},
"tokenRequest": {
"scopes": [f"api://{self.server_app_id}/access_as_user"],
},
}
===========changed ref 0===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 1===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
self.use_authentication = use_authentication
self.server_app_id = server_app_id
self.server_app_secret = server_app_secret
self.client_app_id = client_app_id
self.tenant_id = tenant_id
self.authority = f"https://login.microsoftonline.com/{tenant_id}"
if self.use_authentication:
- self.token_cache_path = token_cache_path
- if not self.token_cache_path:
- self.temporary_directory = TemporaryDirectory()
- self.token_cache_path = os.path.join(self.temporary_directory.name, "token_cache.bin")
- try:
- persistence = build_encrypted_persistence(location=self.token_cache_path)
- except Exception:
- logging.exception("Encryption unavailable. Opting in to plain text.")
- persistence = FilePersistence(location=self.token_cache_path)
+ field_names = [field.name for field in search_index.fields] if search_index else []
+ self.has_auth_fields = "oids" in field_names and "groups" in field_names
+ self.require_access_control = require_access_control
self.confidential_client = ConfidentialClientApplication(
- server_app_id,
- authority=self.authority,
- client_credential=server_app_secret,
- token_cache=PersistedTokenCache(persistence),
+ server_app_id, authority=self.authority,</s>
===========changed ref 2===========
<s>.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
# offset: 1
<s>,
- token_cache=PersistedTokenCache(persistence),
+ server_app_id, authority=self.authority, client_credential=server_app_secret, token_cache=TokenCache()
)
+ else:
+ self.has_auth_fields = False
+ self.require_access_control = False
===========changed ref 3===========
+ # module: scripts.auth_common
+
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 5===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 8===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 9===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 10===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
|
app.backend.core.authentication/AuthenticationHelper.get_token_auth_header
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<2>:<add> if auth:
<del> if not auth:
<3>:<del> raise AuthError(
<4>:<del> {"code": "authorization_header_missing", "description": "Authorization header is expected"}, 401
<5>:<del> )
<6>:<add> parts = auth.split()
<7>:<add> if parts[0].lower() != "bearer":
<add> raise AuthError(error="Authorization header must start with Bearer", status_code=401)
<add> elif len(parts) == 1:
<add> raise AuthError(error="Token not found", status_code=401)
<add> elif len(parts) > 2:
<add> raise AuthError(error="Authorization header must be Bearer token", status_code=401)
<del> parts = auth.split()
<9>:<del> if parts[0].lower() != "bearer":
<10>:<del> raise AuthError(
<11>:<del> {"code": "invalid_header", "description": "Authorization header must start with Bearer"}, 401
<12>:<del> )
<13>:<del> elif len(parts) == 1:
<14>:<del> raise AuthError({"code": "invalid_header", "description": "Token not found"}, 401)
<15>:<del> elif len(parts) > 2:
<16>:<del> raise AuthError({"code": "invalid_header", "description": "Authorization header must be Bearer token"}, 401)
<17>:<add> token = parts[1]
<add> return token
<18>:<add> # App services built-in authentication passes the access token directly as a header
<add> # To learn more, please visit https://learn.microsoft.com/azure/app-service/configure-authentication-oauth-tokens
<add> token = headers.get("x-ms-token-aad-access-token", None)
<add> if token:
<del> token = parts[1]
<19>:<add> return token
<del> return token
|
# module: app.backend.core.authentication
class AuthenticationHelper:
@staticmethod
def get_token_auth_header(headers: dict) -> str:
<0> # Obtains the Access Token from the Authorization Header
<1> auth = headers.get("Authorization", None)
<2> if not auth:
<3> raise AuthError(
<4> {"code": "authorization_header_missing", "description": "Authorization header is expected"}, 401
<5> )
<6>
<7> parts = auth.split()
<8>
<9> if parts[0].lower() != "bearer":
<10> raise AuthError(
<11> {"code": "invalid_header", "description": "Authorization header must start with Bearer"}, 401
<12> )
<13> elif len(parts) == 1:
<14> raise AuthError({"code": "invalid_header", "description": "Token not found"}, 401)
<15> elif len(parts) > 2:
<16> raise AuthError({"code": "invalid_header", "description": "Authorization header must be Bearer token"}, 401)
<17>
<18> token = parts[1]
<19> return token
<20>
|
===========unchanged ref 0===========
at: app.backend.core.authentication
AuthError(error, status_code)
===========changed ref 0===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 1===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
self.use_authentication = use_authentication
self.server_app_id = server_app_id
self.server_app_secret = server_app_secret
self.client_app_id = client_app_id
self.tenant_id = tenant_id
self.authority = f"https://login.microsoftonline.com/{tenant_id}"
if self.use_authentication:
- self.token_cache_path = token_cache_path
- if not self.token_cache_path:
- self.temporary_directory = TemporaryDirectory()
- self.token_cache_path = os.path.join(self.temporary_directory.name, "token_cache.bin")
- try:
- persistence = build_encrypted_persistence(location=self.token_cache_path)
- except Exception:
- logging.exception("Encryption unavailable. Opting in to plain text.")
- persistence = FilePersistence(location=self.token_cache_path)
+ field_names = [field.name for field in search_index.fields] if search_index else []
+ self.has_auth_fields = "oids" in field_names and "groups" in field_names
+ self.require_access_control = require_access_control
self.confidential_client = ConfidentialClientApplication(
- server_app_id,
- authority=self.authority,
- client_credential=server_app_secret,
- token_cache=PersistedTokenCache(persistence),
+ server_app_id, authority=self.authority,</s>
===========changed ref 2===========
<s>.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
# offset: 1
<s>,
- token_cache=PersistedTokenCache(persistence),
+ server_app_id, authority=self.authority, client_credential=server_app_secret, token_cache=TokenCache()
)
+ else:
+ self.has_auth_fields = False
+ self.require_access_control = False
===========changed ref 3===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_setup_for_client(self) -> dict[str, Any]:
# returns MSAL.js settings used by the client app
return {
"useLogin": self.use_authentication, # Whether or not login elements are enabled on the UI
+ "requireAccessControl": self.require_access_control, # Whether or not access control is required to use the application
"msalConfig": {
"auth": {
"clientId": self.client_app_id, # Client app id used for login
"authority": self.authority, # Directory to use for login https://learn.microsoft.com/azure/active-directory/develop/msal-client-application-configuration#authority
"redirectUri": "/redirect", # Points to window.location.origin. You must register this URI on Azure Portal/App Registration.
"postLogoutRedirectUri": "/", # Indicates the page to navigate after logout.
"navigateToLoginRequestUrl": False, # If "true", will navigate back to the original request location before processing the auth code response.
},
"cache": {
"cacheLocation": "sessionStorage",
"storeAuthStateInCookie": False,
}, # Configures cache location. "sessionStorage" is more secure, but "localStorage" gives you SSO between tabs. # Set this to "true" if you are having issues on IE11 or Edge
},
"loginRequest": {
# Scopes you add here will be prompted for user consent during sign-in.
# By default, MSAL.js will add OIDC scopes (openid, profile, email) to any login request.
# For more information about OIDC scopes, visit:
# https://docs.microsoft.com/azure/active-directory/develop/v2-permissions-and-consent#openid-connect-scopes
"scopes": [".default"],
# Uncomment the following line to cause a consent dialog to appear on every login
# For more information, please visit https://learn.microsoft.com/azure/active-directory/develop/v2-oauth2-</s>
===========changed ref 4===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_setup_for_client(self) -> dict[str, Any]:
# offset: 1
<s> # For more information, please visit https://learn.microsoft.com/azure/active-directory/develop/v2-oauth2-auth-code-flow#request-an-authorization-code
# "prompt": "consent"
},
"tokenRequest": {
"scopes": [f"api://{self.server_app_id}/access_as_user"],
},
}
===========changed ref 5===========
+ # module: scripts.auth_common
+
+
===========changed ref 6===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 7===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 8===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 9===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 10===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
|
app.backend.core.authentication/AuthenticationHelper.build_security_filters
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> use_oid_security_filter = self.require_access_control or overrides.get("use_oid_security_filter")
<del> use_oid_security_filter = overrides.get("use_oid_security_filter")
<4>:<add> use_groups_security_filter = self.require_access_control or overrides.get("use_groups_security_filter")
<del> use_groups_security_filter = overrides.get("use_groups_security_filter")
<5>:<add>
<add> if (use_oid_security_filter or use_oid_security_filter) and not self.has_auth_fields:
<add> raise AuthError(
<add> error="oids and groups must be defined in the search index to use authentication", status_code=400
<add> )
|
# module: app.backend.core.authentication
class AuthenticationHelper:
- @staticmethod
+ def build_security_filters(self, overrides: dict[str, Any], auth_claims: dict[str, Any]):
- def build_security_filters(overrides: dict[str, Any], auth_claims: dict[str, Any]):
<0> # Build different permutations of the oid or groups security filter using OData filters
<1> # https://learn.microsoft.com/azure/search/search-security-trimming-for-azure-search
<2> # https://learn.microsoft.com/azure/search/search-query-odata-filter
<3> use_oid_security_filter = overrides.get("use_oid_security_filter")
<4> use_groups_security_filter = overrides.get("use_groups_security_filter")
<5>
<6> oid_security_filter = (
<7> "oids/any(g:search.in(g, '{}'))".format(auth_claims.get("oid") or "") if use_oid_security_filter else None
<8> )
<9> groups_security_filter = (
<10> "groups/any(g:search.in(g, '{}'))".format(", ".join(auth_claims.get("groups") or []))
<11> if use_groups_security_filter
<12> else None
<13> )
<14>
<15> # If only one security filter is specified, return that filter
<16> # If both security filters are specified, combine them with "or" so only 1 security filter needs to pass
<17> # If no security filters are specified, don't return any filter
<18> if oid_security_filter and not groups_security_filter:
<19> return oid_security_filter
<20> elif groups_security_filter and not oid_security_filter:
<21> return groups_security_filter
<22> elif oid_security_filter and groups_security_filter:
<23> return f"({oid_security_filter} or {groups_security_filter})"
<24> else:
<25> return None
<26>
|
===========unchanged ref 0===========
at: app.backend.core.authentication
AuthError(error, status_code)
===========changed ref 0===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 1===========
# module: app.backend.core.authentication
class AuthenticationHelper:
@staticmethod
def get_token_auth_header(headers: dict) -> str:
# Obtains the Access Token from the Authorization Header
auth = headers.get("Authorization", None)
+ if auth:
- if not auth:
- raise AuthError(
- {"code": "authorization_header_missing", "description": "Authorization header is expected"}, 401
- )
+ parts = auth.split()
+ if parts[0].lower() != "bearer":
+ raise AuthError(error="Authorization header must start with Bearer", status_code=401)
+ elif len(parts) == 1:
+ raise AuthError(error="Token not found", status_code=401)
+ elif len(parts) > 2:
+ raise AuthError(error="Authorization header must be Bearer token", status_code=401)
- parts = auth.split()
- if parts[0].lower() != "bearer":
- raise AuthError(
- {"code": "invalid_header", "description": "Authorization header must start with Bearer"}, 401
- )
- elif len(parts) == 1:
- raise AuthError({"code": "invalid_header", "description": "Token not found"}, 401)
- elif len(parts) > 2:
- raise AuthError({"code": "invalid_header", "description": "Authorization header must be Bearer token"}, 401)
+ token = parts[1]
+ return token
+ # App services built-in authentication passes the access token directly as a header
+ # To learn more, please visit https://learn.microsoft.com/azure/app-service/configure-authentication-oauth-tokens
+ token = headers.get("x-ms-token-aad-access-token", None)
+ if token:
- token = parts[1]
+ return token
- return token
===========changed ref 2===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
self.use_authentication = use_authentication
self.server_app_id = server_app_id
self.server_app_secret = server_app_secret
self.client_app_id = client_app_id
self.tenant_id = tenant_id
self.authority = f"https://login.microsoftonline.com/{tenant_id}"
if self.use_authentication:
- self.token_cache_path = token_cache_path
- if not self.token_cache_path:
- self.temporary_directory = TemporaryDirectory()
- self.token_cache_path = os.path.join(self.temporary_directory.name, "token_cache.bin")
- try:
- persistence = build_encrypted_persistence(location=self.token_cache_path)
- except Exception:
- logging.exception("Encryption unavailable. Opting in to plain text.")
- persistence = FilePersistence(location=self.token_cache_path)
+ field_names = [field.name for field in search_index.fields] if search_index else []
+ self.has_auth_fields = "oids" in field_names and "groups" in field_names
+ self.require_access_control = require_access_control
self.confidential_client = ConfidentialClientApplication(
- server_app_id,
- authority=self.authority,
- client_credential=server_app_secret,
- token_cache=PersistedTokenCache(persistence),
+ server_app_id, authority=self.authority,</s>
===========changed ref 3===========
<s>.backend.core.authentication
class AuthenticationHelper:
def __init__(
self,
+ search_index: Optional[SearchIndex],
use_authentication: bool,
server_app_id: Optional[str],
server_app_secret: Optional[str],
client_app_id: Optional[str],
tenant_id: Optional[str],
+ require_access_control: bool = False,
- token_cache_path: Optional[str] = None,
):
# offset: 1
<s>,
- token_cache=PersistedTokenCache(persistence),
+ server_app_id, authority=self.authority, client_credential=server_app_secret, token_cache=TokenCache()
)
+ else:
+ self.has_auth_fields = False
+ self.require_access_control = False
===========changed ref 4===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_setup_for_client(self) -> dict[str, Any]:
# returns MSAL.js settings used by the client app
return {
"useLogin": self.use_authentication, # Whether or not login elements are enabled on the UI
+ "requireAccessControl": self.require_access_control, # Whether or not access control is required to use the application
"msalConfig": {
"auth": {
"clientId": self.client_app_id, # Client app id used for login
"authority": self.authority, # Directory to use for login https://learn.microsoft.com/azure/active-directory/develop/msal-client-application-configuration#authority
"redirectUri": "/redirect", # Points to window.location.origin. You must register this URI on Azure Portal/App Registration.
"postLogoutRedirectUri": "/", # Indicates the page to navigate after logout.
"navigateToLoginRequestUrl": False, # If "true", will navigate back to the original request location before processing the auth code response.
},
"cache": {
"cacheLocation": "sessionStorage",
"storeAuthStateInCookie": False,
}, # Configures cache location. "sessionStorage" is more secure, but "localStorage" gives you SSO between tabs. # Set this to "true" if you are having issues on IE11 or Edge
},
"loginRequest": {
# Scopes you add here will be prompted for user consent during sign-in.
# By default, MSAL.js will add OIDC scopes (openid, profile, email) to any login request.
# For more information about OIDC scopes, visit:
# https://docs.microsoft.com/azure/active-directory/develop/v2-permissions-and-consent#openid-connect-scopes
"scopes": [".default"],
# Uncomment the following line to cause a consent dialog to appear on every login
# For more information, please visit https://learn.microsoft.com/azure/active-directory/develop/v2-oauth2-</s>
===========changed ref 5===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_setup_for_client(self) -> dict[str, Any]:
# offset: 1
<s> # For more information, please visit https://learn.microsoft.com/azure/active-directory/develop/v2-oauth2-auth-code-flow#request-an-authorization-code
# "prompt": "consent"
},
"tokenRequest": {
"scopes": [f"api://{self.server_app_id}/access_as_user"],
},
}
===========changed ref 6===========
+ # module: scripts.auth_common
+
+
===========changed ref 7===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
|
app.backend.core.authentication/AuthenticationHelper.get_auth_claims_if_enabled
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_claims_if_enabled(self, headers: dict) -> dict[str, Any]:
<0> if not self.use_authentication:
<1> return {}
<2> try:
<3> # Read the authentication token from the authorization header and exchange it using the On Behalf Of Flow
<4> # The scope is set to the Microsoft Graph API, which may need to be called for more authorization information
<5> # https://learn.microsoft.com/en-us/azure/active-directory/develop/v2-oauth2-on-behalf-of-flow
<6> auth_token = AuthenticationHelper.get_token_auth_header(headers)
<7> graph_resource_access_token = self.confidential_client.acquire_token_on_behalf_of(
<8> user_assertion=auth_token, scopes=["https://graph.microsoft.com/.default"]
<9> )
<10> if "error" in graph_resource_access_token:
<11> raise AuthError(error=str(graph_resource_access_token), status_code=401)
<12>
<13> # Read the claims from the response. The oid and groups claims are used for security filtering
<14> # https://learn.microsoft.com/azure/active-directory/develop/id-token-claims-reference
<15> id_token_claims = graph_resource_access_token["id_token_claims"]
<16> auth_claims = {"oid": id_token_claims["oid"], "groups": id_token_claims.get("groups") or []}
<17>
<18> # A groups claim may have been omitted either because it was not added in the application manifest for the API application,
<19> # or a groups overage claim may have been emitted.
<20> # https://learn.microsoft.com/azure/active-directory/develop/id-token-claims-reference#groups-overage-claim
<21> missing_groups_claim = "groups" not in id_token_claims
<22> has_group_overage_claim = (
<23> missing_groups_claim
<24> and "_claim_names" in id_token_claims
<25> </s>
|
===========below chunk 0===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def get_auth_claims_if_enabled(self, headers: dict) -> dict[str, Any]:
# offset: 1
)
if missing_groups_claim or has_group_overage_claim:
# Read the user's groups from Microsoft Graph
auth_claims["groups"] = await AuthenticationHelper.list_groups(graph_resource_access_token)
return auth_claims
except AuthError as e:
print(e.error)
logging.exception("Exception getting authorization information - " + json.dumps(e.error))
return {}
except Exception:
logging.exception("Exception getting authorization information")
return {}
===========unchanged ref 0===========
at: app.backend.core.authentication
AuthError(error, status_code)
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
at: app.backend.core.authentication.AuthenticationHelper
get_token_auth_header(headers: dict) -> str
list_groups(graph_resource_access_token: dict) -> list[str]
at: json
dumps(obj: Any, *, skipkeys: bool=..., ensure_ascii: bool=..., check_circular: bool=..., allow_nan: bool=..., cls: Optional[Type[JSONEncoder]]=..., indent: Union[None, int, str]=..., separators: Optional[Tuple[str, str]]=..., default: Optional[Callable[[Any], Any]]=..., sort_keys: bool=..., **kwds: Any) -> str
at: logging
exception(msg: Any, *args: Any, exc_info: _ExcInfoType=..., stack_info: bool=..., extra: Optional[Dict[str, Any]]=..., **kwargs: Any) -> None
===========changed ref 0===========
# module: app.backend.core.authentication
class AuthenticationHelper:
@staticmethod
def get_token_auth_header(headers: dict) -> str:
# Obtains the Access Token from the Authorization Header
auth = headers.get("Authorization", None)
+ if auth:
- if not auth:
- raise AuthError(
- {"code": "authorization_header_missing", "description": "Authorization header is expected"}, 401
- )
+ parts = auth.split()
+ if parts[0].lower() != "bearer":
+ raise AuthError(error="Authorization header must start with Bearer", status_code=401)
+ elif len(parts) == 1:
+ raise AuthError(error="Token not found", status_code=401)
+ elif len(parts) > 2:
+ raise AuthError(error="Authorization header must be Bearer token", status_code=401)
- parts = auth.split()
- if parts[0].lower() != "bearer":
- raise AuthError(
- {"code": "invalid_header", "description": "Authorization header must start with Bearer"}, 401
- )
- elif len(parts) == 1:
- raise AuthError({"code": "invalid_header", "description": "Token not found"}, 401)
- elif len(parts) > 2:
- raise AuthError({"code": "invalid_header", "description": "Authorization header must be Bearer token"}, 401)
+ token = parts[1]
+ return token
+ # App services built-in authentication passes the access token directly as a header
+ # To learn more, please visit https://learn.microsoft.com/azure/app-service/configure-authentication-oauth-tokens
+ token = headers.get("x-ms-token-aad-access-token", None)
+ if token:
- token = parts[1]
+ return token
- return token
===========changed ref 1===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 2===========
# module: app.backend.core.authentication
class AuthenticationHelper:
- @staticmethod
+ def build_security_filters(self, overrides: dict[str, Any], auth_claims: dict[str, Any]):
- def build_security_filters(overrides: dict[str, Any], auth_claims: dict[str, Any]):
# Build different permutations of the oid or groups security filter using OData filters
# https://learn.microsoft.com/azure/search/search-security-trimming-for-azure-search
# https://learn.microsoft.com/azure/search/search-query-odata-filter
+ use_oid_security_filter = self.require_access_control or overrides.get("use_oid_security_filter")
- use_oid_security_filter = overrides.get("use_oid_security_filter")
+ use_groups_security_filter = self.require_access_control or overrides.get("use_groups_security_filter")
- use_groups_security_filter = overrides.get("use_groups_security_filter")
+
+ if (use_oid_security_filter or use_oid_security_filter) and not self.has_auth_fields:
+ raise AuthError(
+ error="oids and groups must be defined in the search index to use authentication", status_code=400
+ )
oid_security_filter = (
"oids/any(g:search.in(g, '{}'))".format(auth_claims.get("oid") or "") if use_oid_security_filter else None
)
groups_security_filter = (
"groups/any(g:search.in(g, '{}'))".format(", ".join(auth_claims.get("groups") or []))
if use_groups_security_filter
else None
)
# If only one security filter is specified, return that filter
# If both security filters are specified, combine them with "or" so only 1 security filter needs to pass
# If no security filters are specified, don't return any filter
if oid_security_filter and not groups_security_filter:
return oid</s>
===========changed ref 3===========
# module: app.backend.core.authentication
class AuthenticationHelper:
- @staticmethod
+ def build_security_filters(self, overrides: dict[str, Any], auth_claims: dict[str, Any]):
- def build_security_filters(overrides: dict[str, Any], auth_claims: dict[str, Any]):
# offset: 1
<s> security filters are specified, don't return any filter
if oid_security_filter and not groups_security_filter:
return oid_security_filter
elif groups_security_filter and not oid_security_filter:
return groups_security_filter
elif oid_security_filter and groups_security_filter:
return f"({oid_security_filter} or {groups_security_filter})"
else:
return None
|
|
app.backend.approaches.retrievethenreadvision/RetrieveThenReadVisionApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> self.auth_helper = auth_helper
|
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
<0> self.search_client = search_client
<1> self.blob_container_client = blob_container_client
<2> self.openai_client = openai_client
<3> self.embedding_model = embedding_model
<4> self.embedding_deployment = embedding_deployment
<5> self.sourcepage_field = sourcepage_field
<6> self.content_field = content_field
<7> self.gpt4v_deployment = gpt4v_deployment
<8> self.gpt4v_model = gpt4v_model
<9> self.query_language = query_language
<10> self.query_speller = query_speller
<11> self.vision_endpoint = vision_endpoint
<12> self.vision_key = vision_key
<13>
|
===========unchanged ref 0===========
at: approaches.approach.Approach
__init__(self, search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
at: core.authentication
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 7===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 8===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 9===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 10===========
+ # module: scripts.auth_update
+ def main():
+ if not test_authentication_enabled():
+ print("Not updating authentication.")
+ exit(0)
+
+ credential = AzureDeveloperCliCredential(tenant_id=os.getenv("AZURE_AUTH_TENANT_ID", os.getenv("AZURE_TENANT_ID")))
+ auth_headers = await get_auth_headers(credential)
+
+ uri = os.getenv("BACKEND_URI")
+ client_app_id = os.getenv("AZURE_CLIENT_APP_ID", None)
+ if client_app_id:
+ client_object_id = await get_application(auth_headers, client_app_id)
+ if client_object_id:
+ print(f"Updating redirect URIs for client app ID {client_app_id}...")
+ # Redirect URIs need to be relative to the deployed application
+ payload = {
+ "publicClient": {"redirectUris": []},
+ "spa": {
+ "redirectUris": [
+ "http://localhost:50505/redirect",
+ f"{uri}/redirect",
+ ]
+ },
+ "web": {
+ "redirectUris": [
+ f"{uri}/.auth/login/aad/callback",
+ ]
+ },
+ }
+ await update_application(auth_headers, client_object_id, payload)
+ print(f"Application update for client app id {client_app_id} complete.")
+
|
tests.test_chatapproach/chat_approach
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<2>:<add> auth_helper=None,
|
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
<0> return ChatReadRetrieveReadApproach(
<1> search_client=None,
<2> openai_client=None,
<3> chatgpt_model="gpt-35-turbo",
<4> chatgpt_deployment="chat",
<5> embedding_deployment="embeddings",
<6> embedding_model="text-",
<7> sourcepage_field="",
<8> content_field="",
<9> query_language="en-us",
<10> query_speller="lexicon",
<11> )
<12>
|
===========unchanged ref 0===========
at: _pytest.fixtures
fixture(fixture_function: FixtureFunction, *, scope: "Union[_ScopeName, Callable[[str, Config], _ScopeName]]"=..., params: Optional[Iterable[object]]=..., autouse: bool=..., ids: Optional[
Union[Sequence[Optional[object]], Callable[[Any], Optional[object]]]
]=..., name: Optional[str]=...) -> FixtureFunction
fixture(fixture_function: None=..., *, scope: "Union[_ScopeName, Callable[[str, Config], _ScopeName]]"=..., params: Optional[Iterable[object]]=..., autouse: bool=..., ids: Optional[
Union[Sequence[Optional[object]], Callable[[Any], Optional[object]]]
]=..., name: Optional[str]=None) -> FixtureFunctionMarker
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 7===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 8===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 9===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
===========changed ref 10===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 11===========
+ # module: scripts.auth_update
+ def main():
+ if not test_authentication_enabled():
+ print("Not updating authentication.")
+ exit(0)
+
+ credential = AzureDeveloperCliCredential(tenant_id=os.getenv("AZURE_AUTH_TENANT_ID", os.getenv("AZURE_TENANT_ID")))
+ auth_headers = await get_auth_headers(credential)
+
+ uri = os.getenv("BACKEND_URI")
+ client_app_id = os.getenv("AZURE_CLIENT_APP_ID", None)
+ if client_app_id:
+ client_object_id = await get_application(auth_headers, client_app_id)
+ if client_object_id:
+ print(f"Updating redirect URIs for client app ID {client_app_id}...")
+ # Redirect URIs need to be relative to the deployed application
+ payload = {
+ "publicClient": {"redirectUris": []},
+ "spa": {
+ "redirectUris": [
+ "http://localhost:50505/redirect",
+ f"{uri}/redirect",
+ ]
+ },
+ "web": {
+ "redirectUris": [
+ f"{uri}/.auth/login/aad/callback",
+ ]
+ },
+ }
+ await update_application(auth_headers, client_object_id, payload)
+ print(f"Application update for client app id {client_app_id} complete.")
+
|
app.backend.approaches.approach/Approach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<2>:<add> self.auth_helper = auth_helper
|
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
<0> self.search_client = search_client
<1> self.openai_client = openai_client
<2> self.query_language = query_language
<3> self.query_speller = query_speller
<4> self.embedding_deployment = embedding_deployment
<5> self.embedding_model = embedding_model
<6> self.openai_host = openai_host
<7>
|
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 7===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 8===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 9===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 10===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
===========changed ref 11===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 12===========
+ # module: scripts.auth_update
+ def main():
+ if not test_authentication_enabled():
+ print("Not updating authentication.")
+ exit(0)
+
+ credential = AzureDeveloperCliCredential(tenant_id=os.getenv("AZURE_AUTH_TENANT_ID", os.getenv("AZURE_TENANT_ID")))
+ auth_headers = await get_auth_headers(credential)
+
+ uri = os.getenv("BACKEND_URI")
+ client_app_id = os.getenv("AZURE_CLIENT_APP_ID", None)
+ if client_app_id:
+ client_object_id = await get_application(auth_headers, client_app_id)
+ if client_object_id:
+ print(f"Updating redirect URIs for client app ID {client_app_id}...")
+ # Redirect URIs need to be relative to the deployed application
+ payload = {
+ "publicClient": {"redirectUris": []},
+ "spa": {
+ "redirectUris": [
+ "http://localhost:50505/redirect",
+ f"{uri}/redirect",
+ ]
+ },
+ "web": {
+ "redirectUris": [
+ f"{uri}/.auth/login/aad/callback",
+ ]
+ },
+ }
+ await update_application(auth_headers, client_object_id, payload)
+ print(f"Application update for client app id {client_app_id} complete.")
+
|
app.backend.approaches.approach/Approach.build_filter
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<1>:<add> security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
<del> security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
|
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
<0> exclude_category = overrides.get("exclude_category") or None
<1> security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
<2> filters = []
<3> if exclude_category:
<4> filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
<5> if security_filter:
<6> filters.append(security_filter)
<7> return None if len(filters) == 0 else " and ".join(filters)
<8>
|
===========unchanged ref 0===========
at: core.authentication
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
at: core.authentication.AuthenticationHelper
scope: str = "https://graph.microsoft.com/.default"
build_security_filters(overrides: dict[str, Any], auth_claims: dict[str, Any])
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
===========changed ref 0===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 1===========
+ # module: scripts.auth_common
+
+
===========changed ref 2===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 3===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 4===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 5===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 8===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 9===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 10===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 11===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
===========changed ref 12===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
|
tests.test_authenticationhelper/create_authentication_helper
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<1>:<add> search_index=MockSearchIndex,
<6>:<add> require_access_control=require_access_control,
<del> token_cache_path=None,
|
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
<0> return AuthenticationHelper(
<1> use_authentication=True,
<2> server_app_id="SERVER_APP",
<3> server_app_secret="SERVER_SECRET",
<4> client_app_id="CLIENT_APP",
<5> tenant_id="TENANT_ID",
<6> token_cache_path=None,
<7> )
<8>
|
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 7===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 8===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 9===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 10===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 11===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
===========changed ref 12===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
===========changed ref 13===========
# module: tests.test_chatvisionapproach
@pytest.fixture
+ def chat_approach(openai_client, mock_confidential_client_success):
- def chat_approach(openai_client):
return ChatReadRetrieveReadVisionApproach(
search_client=None,
openai_client=openai_client,
+ auth_helper=AuthenticationHelper(
+ search_index=MockSearchIndex,
+ use_authentication=True,
+ server_app_id="SERVER_APP",
+ server_app_secret="SERVER_SECRET",
+ client_app_id="CLIENT_APP",
+ tenant_id="TENANT_ID",
+ require_access_control=None,
+ ),
blob_container_client=None,
vision_endpoint="endpoint",
vision_key="key",
gpt4v_deployment="gpt-4v",
gpt4v_model="gpt-4v",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
|
tests.test_authenticationhelper/test_auth_setup
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> "requireAccessControl": False,
|
# module: tests.test_authenticationhelper
def test_auth_setup(mock_confidential_client_success):
<0> helper = create_authentication_helper()
<1> assert helper.get_auth_setup_for_client() == {
<2> "useLogin": True,
<3> "msalConfig": {
<4> "auth": {
<5> "clientId": "CLIENT_APP",
<6> "authority": "https://login.microsoftonline.com/TENANT_ID",
<7> "redirectUri": "/redirect",
<8> "postLogoutRedirectUri": "/",
<9> "navigateToLoginRequestUrl": False,
<10> },
<11> "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
<12> },
<13> "loginRequest": {
<14> "scopes": [".default"],
<15> },
<16> "tokenRequest": {
<17> "scopes": ["api://SERVER_APP/access_as_user"],
<18> },
<19> }
<20>
|
===========changed ref 0===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 1===========
+ # module: scripts.auth_common
+
+
===========changed ref 2===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 3===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 4===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 5===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 8===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 9===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 10===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 11===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 12===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
===========changed ref 13===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
|
tests.test_authenticationhelper/test_get_auth_token
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<13>:<add> AuthenticationHelper.get_token_auth_header({"x-ms-token-aad-access-token": "MockToken"}) == "MockToken"
|
# module: tests.test_authenticationhelper
+ def test_get_auth_token(mock_confidential_client_success):
- def test_get_auth_token():
<0> with pytest.raises(AuthError) as exc_info:
<1> AuthenticationHelper.get_token_auth_header({})
<2> assert exc_info.value.status_code == 401
<3> with pytest.raises(AuthError) as exc_info:
<4> AuthenticationHelper.get_token_auth_header({"Authorization": ". ."})
<5> assert exc_info.value.status_code == 401
<6> with pytest.raises(AuthError) as exc_info:
<7> AuthenticationHelper.get_token_auth_header({"Authorization": "invalid"})
<8> assert exc_info.value.status_code == 401
<9> with pytest.raises(AuthError) as exc_info:
<10> AuthenticationHelper.get_token_auth_header({"Authorization": "invalid MockToken"})
<11> assert exc_info.value.status_code == 401
<12> assert AuthenticationHelper.get_token_auth_header({"Authorization": "Bearer MockToken"}) == "MockToken"
<13>
|
===========changed ref 0===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 1===========
# module: tests.test_authenticationhelper
def test_auth_setup(mock_confidential_client_success):
helper = create_authentication_helper()
assert helper.get_auth_setup_for_client() == {
"useLogin": True,
+ "requireAccessControl": False,
"msalConfig": {
"auth": {
"clientId": "CLIENT_APP",
"authority": "https://login.microsoftonline.com/TENANT_ID",
"redirectUri": "/redirect",
"postLogoutRedirectUri": "/",
"navigateToLoginRequestUrl": False,
},
"cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
},
"loginRequest": {
"scopes": [".default"],
},
"tokenRequest": {
"scopes": ["api://SERVER_APP/access_as_user"],
},
}
===========changed ref 2===========
+ # module: scripts.auth_common
+
+
===========changed ref 3===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 4===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 5===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 6===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 7===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 8===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 9===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 10===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 11===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 12===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 13===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
===========changed ref 14===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
|
tests.test_authenticationhelper/test_build_security_filters
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<0>:<add> auth_helper = create_authentication_helper()
<add> auth_helper_require_access_control = create_authentication_helper(require_access_control=True)
<add> assert auth_helper.build_security_filters(overrides={}, auth_claims={}) is None
<del> assert AuthenticationHelper.build_security_filters(overrides={}, auth_claims={}) is None
<2>:<add> auth_helper_require_access_control.build_security_filters(overrides={}, auth_claims={})
<add> == "(oids/any(g:search.in(g, '')) or groups/any(g:search.in(g, '')))"
<add> )
<add> assert (
<del> AuthenticationHelper.build_security_filters(
<3>:<add> auth_helper.build_security_filters(overrides={"use_oid_security_filter": True}, auth_claims={"oid": "OID_X"})
<del> overrides={"use_oid_security_filter": True}, auth_claims={"oid": "OID_X"}
<4>:<del> )
<8>:<add> auth_helper_require_access_control.build_security_filters(overrides={}, auth_claims={"oid": "OID_X"})
<add> == "(oids/any(g:search.in(g, 'OID_X')) or groups/any(g:search.in(g, '')))"
|
# module: tests.test_authenticationhelper
+ def test_build_security_filters(mock_confidential_client_success):
- def test_build_security_filters():
<0> assert AuthenticationHelper.build_security_filters(overrides={}, auth_claims={}) is None
<1> assert (
<2> AuthenticationHelper.build_security_filters(
<3> overrides={"use_oid_security_filter": True}, auth_claims={"oid": "OID_X"}
<4> )
<5> == "oids/any(g:search.in(g, 'OID_X'))"
<6> )
<7> assert (
<8> AuthenticationHelper.build_security_filters(
<9> overrides={"use_groups_security_filter": True}, auth_claims={"groups": ["GROUP_Y", "GROUP_Z"]}
<10> )
<11> == "groups/any(g:search.in(g, 'GROUP_Y, GROUP_Z'))"
<12> )
<13> assert (
<14> AuthenticationHelper.build_security_filters(
<15> overrides={"use_oid_security_filter": True, "use_groups_security_filter": True},
<16> auth_claims={"oid": "OID_X", "groups": ["GROUP_Y", "GROUP_Z"]},
<17> )
<18> == "(oids/any(g:search.in(g, 'OID_X')) or groups/any(g:search.in(g, 'GROUP_Y, GROUP_Z')))"
<19> )
<20> assert (
<21> AuthenticationHelper.build_security_filters(
<22> overrides={"use_groups_security_filter": True}, auth_claims={"oid": "OID_X"}
<23> )
<24> == "groups/any(g:search.in(g, ''))"
<25> )
<26> assert (
<27> AuthenticationHelper.build_security_filters(
<28> overrides={"use_oid_security_filter": True}, auth_claims={"groups": ["GROUP_Y", "GROUP_Z"]}
<29> )
<30> == "oids/any(g:search.in(g, ''))"
<31> )
<32>
|
===========changed ref 0===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 1===========
# module: tests.test_authenticationhelper
def test_auth_setup(mock_confidential_client_success):
helper = create_authentication_helper()
assert helper.get_auth_setup_for_client() == {
"useLogin": True,
+ "requireAccessControl": False,
"msalConfig": {
"auth": {
"clientId": "CLIENT_APP",
"authority": "https://login.microsoftonline.com/TENANT_ID",
"redirectUri": "/redirect",
"postLogoutRedirectUri": "/",
"navigateToLoginRequestUrl": False,
},
"cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
},
"loginRequest": {
"scopes": [".default"],
},
"tokenRequest": {
"scopes": ["api://SERVER_APP/access_as_user"],
},
}
===========changed ref 2===========
# module: tests.test_authenticationhelper
+ def test_auth_setup_required_access_control(mock_confidential_client_success):
+ helper = create_authentication_helper(require_access_control=True)
+ assert helper.get_auth_setup_for_client() == {
+ "useLogin": True,
+ "requireAccessControl": True,
+ "msalConfig": {
+ "auth": {
+ "clientId": "CLIENT_APP",
+ "authority": "https://login.microsoftonline.com/TENANT_ID",
+ "redirectUri": "/redirect",
+ "postLogoutRedirectUri": "/",
+ "navigateToLoginRequestUrl": False,
+ },
+ "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
+ },
+ "loginRequest": {
+ "scopes": [".default"],
+ },
+ "tokenRequest": {
+ "scopes": ["api://SERVER_APP/access_as_user"],
+ },
+ }
+
===========changed ref 3===========
# module: tests.test_authenticationhelper
+ def test_get_auth_token(mock_confidential_client_success):
- def test_get_auth_token():
with pytest.raises(AuthError) as exc_info:
AuthenticationHelper.get_token_auth_header({})
assert exc_info.value.status_code == 401
with pytest.raises(AuthError) as exc_info:
AuthenticationHelper.get_token_auth_header({"Authorization": ". ."})
assert exc_info.value.status_code == 401
with pytest.raises(AuthError) as exc_info:
AuthenticationHelper.get_token_auth_header({"Authorization": "invalid"})
assert exc_info.value.status_code == 401
with pytest.raises(AuthError) as exc_info:
AuthenticationHelper.get_token_auth_header({"Authorization": "invalid MockToken"})
assert exc_info.value.status_code == 401
assert AuthenticationHelper.get_token_auth_header({"Authorization": "Bearer MockToken"}) == "MockToken"
+ AuthenticationHelper.get_token_auth_header({"x-ms-token-aad-access-token": "MockToken"}) == "MockToken"
===========changed ref 4===========
+ # module: scripts.auth_common
+
+
===========changed ref 5===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 6===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 7===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 8===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 9===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 10===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 11===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 12===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 13===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
|
app.backend.approaches.chatreadretrieveread/ChatReadRetrieveReadApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<2>:<add> self.auth_helper = auth_helper
|
<s>_helper: AuthenticationHelper,
openai_client: AsyncOpenAI,
chatgpt_model: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
<0> self.search_client = search_client
<1> self.openai_client = openai_client
<2> self.chatgpt_model = chatgpt_model
<3> self.chatgpt_deployment = chatgpt_deployment
<4> self.embedding_deployment = embedding_deployment
<5> self.embedding_model = embedding_model
<6> self.sourcepage_field = sourcepage_field
<7> self.content_field = content_field
<8> self.query_language = query_language
<9> self.query_speller = query_speller
<10> self.chatgpt_token_limit = get_token_limit(chatgpt_model)
<11>
|
===========unchanged ref 0===========
at: approaches.approach.Approach
__init__(self, search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
at: core.authentication
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_authenticationhelper
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 8===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 9===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 10===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 11===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 12===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 13===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
===========changed ref 14===========
<s> auth_helper: AuthenticationHelper,
gpt4v_deployment: Optional[str],
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
self.search_client = search_client
self.blob_container_client = blob_container_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.embedding_model = embedding_model
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.gpt4v_deployment = gpt4v_deployment
self.gpt4v_model = gpt4v_model
self.query_language = query_language
self.query_speller = query_speller
self.vision_endpoint = vision_endpoint
self.vision_key = vision_key
|
app.backend.approaches.retrievethenread/RetrieveThenReadApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> self.auth_helper = auth_helper
|
<s>_helper: AuthenticationHelper,
openai_client: AsyncOpenAI,
chatgpt_model: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
embedding_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
<0> self.search_client = search_client
<1> self.chatgpt_deployment = chatgpt_deployment
<2> self.openai_client = openai_client
<3> self.chatgpt_model = chatgpt_model
<4> self.embedding_model = embedding_model
<5> self.chatgpt_deployment = chatgpt_deployment
<6> self.embedding_deployment = embedding_deployment
<7> self.sourcepage_field = sourcepage_field
<8> self.content_field = content_field
<9> self.query_language = query_language
<10> self.query_speller = query_speller
<11>
|
===========unchanged ref 0===========
at: approaches.approach.Approach
__init__(self, search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
at: core.authentication
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_authenticationhelper
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 8===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 9===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 10===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 11===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 12===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 13===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
===========changed ref 14===========
<s>_helper: AuthenticationHelper,
openai_client: AsyncOpenAI,
chatgpt_model: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.chatgpt_model = chatgpt_model
self.chatgpt_deployment = chatgpt_deployment
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
self.chatgpt_token_limit = get_token_limit(chatgpt_model)
|
app.backend.approaches.chatreadretrievereadvision/ChatReadRetrieveReadVisionApproach.__init__
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<3>:<add> self.auth_helper = auth_helper
|
<s>pt4v_deployment: Optional[str], # Not needed for non-Azure OpenAI
gpt4v_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
vision_endpoint: str,
vision_key: str,
):
<0> self.search_client = search_client
<1> self.blob_container_client = blob_container_client
<2> self.openai_client = openai_client
<3> self.gpt4v_deployment = gpt4v_deployment
<4> self.gpt4v_model = gpt4v_model
<5> self.embedding_deployment = embedding_deployment
<6> self.embedding_model = embedding_model
<7> self.sourcepage_field = sourcepage_field
<8> self.content_field = content_field
<9> self.query_language = query_language
<10> self.query_speller = query_speller
<11> self.vision_endpoint = vision_endpoint
<12> self.vision_key = vision_key
<13> self.chatgpt_token_limit = get_token_limit(gpt4v_model)
<14>
|
===========unchanged ref 0===========
at: approaches.approach.Approach
__init__(self, search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
at: core.authentication
AuthenticationHelper(use_authentication: bool, server_app_id: Optional[str], server_app_secret: Optional[str], client_app_id: Optional[str], tenant_id: Optional[str], token_cache_path: Optional[str]=None)
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_authenticationhelper
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 8===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 9===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 10===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 11===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 12===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 13===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
===========changed ref 14===========
<s>_helper: AuthenticationHelper,
openai_client: AsyncOpenAI,
chatgpt_model: str,
chatgpt_deployment: Optional[str], # Not needed for non-Azure OpenAI
embedding_model: str,
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
sourcepage_field: str,
content_field: str,
query_language: str,
query_speller: str,
):
self.search_client = search_client
self.chatgpt_deployment = chatgpt_deployment
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.chatgpt_model = chatgpt_model
self.embedding_model = embedding_model
self.chatgpt_deployment = chatgpt_deployment
self.embedding_deployment = embedding_deployment
self.sourcepage_field = sourcepage_field
self.content_field = content_field
self.query_language = query_language
self.query_speller = query_speller
|
app.backend.app/ask
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<5>:<del> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<7>:<add> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
|
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
<0> if not request.is_json:
<1> return jsonify({"error": "request must be json"}), 415
<2> request_json = await request.get_json()
<3> context = request_json.get("context", {})
<4> auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
<5> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<6> try:
<7> use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
<8> approach: Approach
<9> if use_gpt4v and CONFIG_ASK_VISION_APPROACH in current_app.config:
<10> approach = cast(Approach, current_app.config[CONFIG_ASK_VISION_APPROACH])
<11> else:
<12> approach = cast(Approach, current_app.config[CONFIG_ASK_APPROACH])
<13> r = await approach.run(
<14> request_json["messages"], context=context, session_state=request_json.get("session_state")
<15> )
<16> return jsonify(r)
<17> except Exception as error:
<18> return error_response(error, "/ask")
<19>
|
===========unchanged ref 0===========
at: app.backend.app
CONFIG_ASK_APPROACH = "ask_approach"
CONFIG_ASK_VISION_APPROACH = "ask_vision_approach"
CONFIG_AUTH_CLIENT = "auth_client"
bp = Blueprint("routes", __name__, static_folder="static")
error_response(error: Exception, route: str, status_code: int=500)
at: approaches.approach
Approach(search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
at: approaches.approach.Approach
run(messages: list[dict], stream: bool=False, session_state: Any=None, context: dict[str, Any]={}) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]
at: typing
cast(typ: Type[_T], val: Any) -> _T
cast(typ: str, val: Any) -> Any
cast(typ: object, val: Any) -> Any
===========changed ref 0===========
+ # module: scripts.auth_common
+
+
===========changed ref 1===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 2===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 3===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 4===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 5===========
# module: tests.test_authenticationhelper
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 6===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 8===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 9===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 10===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 11===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 12===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
===========changed ref 13===========
# module: app.backend.approaches.approach
class Approach:
def build_filter(self, overrides: dict[str, Any], auth_claims: dict[str, Any]) -> Optional[str]:
exclude_category = overrides.get("exclude_category") or None
+ security_filter = self.auth_helper.build_security_filters(overrides, auth_claims)
- security_filter = AuthenticationHelper.build_security_filters(overrides, auth_claims)
filters = []
if exclude_category:
filters.append("category ne '{}'".format(exclude_category.replace("'", "''")))
if security_filter:
filters.append(security_filter)
return None if len(filters) == 0 else " and ".join(filters)
|
app.backend.app/chat
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6392a8cee7a872d0a5d54c6d6af9b1f118bfe048
|
Automate Login Setup (#891)
|
<5>:<del> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<6>:<del>
<8>:<add> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
|
# module: app.backend.app
@bp.route("/chat", methods=["POST"])
async def chat():
<0> if not request.is_json:
<1> return jsonify({"error": "request must be json"}), 415
<2> request_json = await request.get_json()
<3> context = request_json.get("context", {})
<4> auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
<5> context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
<6>
<7> try:
<8> use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
<9> approach: Approach
<10> if use_gpt4v and CONFIG_CHAT_VISION_APPROACH in current_app.config:
<11> approach = cast(Approach, current_app.config[CONFIG_CHAT_VISION_APPROACH])
<12> else:
<13> approach = cast(Approach, current_app.config[CONFIG_CHAT_APPROACH])
<14>
<15> result = await approach.run(
<16> request_json["messages"],
<17> stream=request_json.get("stream", False),
<18> context=context,
<19> session_state=request_json.get("session_state"),
<20> )
<21> if isinstance(result, dict):
<22> return jsonify(result)
<23> else:
<24> response = await make_response(format_as_ndjson(result))
<25> response.timeout = None # type: ignore
<26> response.mimetype = "application/json-lines"
<27> return response
<28> except Exception as error:
<29> return error_response(error, "/chat")
<30>
|
===========unchanged ref 0===========
at: app.backend.app
CONFIG_CHAT_VISION_APPROACH = "chat_vision_approach"
CONFIG_CHAT_APPROACH = "chat_approach"
CONFIG_AUTH_CLIENT = "auth_client"
bp = Blueprint("routes", __name__, static_folder="static")
error_response(error: Exception, route: str, status_code: int=500)
format_as_ndjson(r: AsyncGenerator[dict, None]) -> AsyncGenerator[str, None]
at: approaches.approach
Approach(search_client: SearchClient, openai_client: AsyncOpenAI, query_language: Optional[str], query_speller: Optional[str], embedding_deployment: Optional[str], embedding_model: str, openai_host: str)
at: approaches.approach.Approach
run(messages: list[dict], stream: bool=False, session_state: Any=None, context: dict[str, Any]={}) -> Union[dict[str, Any], AsyncGenerator[dict[str, Any], None]]
at: typing
cast(typ: Type[_T], val: Any) -> _T
cast(typ: str, val: Any) -> Any
cast(typ: object, val: Any) -> Any
===========changed ref 0===========
# module: app.backend.app
@bp.route("/ask", methods=["POST"])
async def ask():
if not request.is_json:
return jsonify({"error": "request must be json"}), 415
request_json = await request.get_json()
context = request_json.get("context", {})
auth_helper = current_app.config[CONFIG_AUTH_CLIENT]
- context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
try:
+ context["auth_claims"] = await auth_helper.get_auth_claims_if_enabled(request.headers)
use_gpt4v = context.get("overrides", {}).get("use_gpt4v", False)
approach: Approach
if use_gpt4v and CONFIG_ASK_VISION_APPROACH in current_app.config:
approach = cast(Approach, current_app.config[CONFIG_ASK_VISION_APPROACH])
else:
approach = cast(Approach, current_app.config[CONFIG_ASK_APPROACH])
r = await approach.run(
request_json["messages"], context=context, session_state=request_json.get("session_state")
)
return jsonify(r)
except Exception as error:
return error_response(error, "/ask")
===========changed ref 1===========
+ # module: scripts.auth_common
+
+
===========changed ref 2===========
+ # module: scripts.auth_common
+ TIMEOUT = 60
+
===========changed ref 3===========
# module: app.backend.core.authentication
# AuthError is raised when the authentication token sent by the client UI cannot be parsed or there is an authentication error accessing the graph API
class AuthError(Exception):
+ def __str__(self) -> str:
+ return self.error or ""
+
===========changed ref 4===========
+ # module: scripts.auth_update
+ if __name__ == "__main__":
+ asyncio.run(main())
+
===========changed ref 5===========
+ # module: scripts.auth_common
+ def get_auth_headers(credential: AsyncTokenCredential):
+ token_result = await credential.get_token("https://graph.microsoft.com/.default")
+ return {"Authorization": f"Bearer {token_result.token}"}
+
===========changed ref 6===========
# module: tests.test_authenticationhelper
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 7===========
# module: tests.test_chatvisionapproach
+ MockSearchIndex = SearchIndex(
+ name="test",
+ fields=[
+ SearchField(name="oids", type="Collection(Edm.String)"),
+ SearchField(name="groups", type="Collection(Edm.String)"),
+ ],
+ )
===========changed ref 8===========
<s>.approaches.approach
class Approach:
def __init__(
self,
search_client: SearchClient,
openai_client: AsyncOpenAI,
+ auth_helper: AuthenticationHelper,
query_language: Optional[str],
query_speller: Optional[str],
embedding_deployment: Optional[str], # Not needed for non-Azure OpenAI or for retrieval_mode="text"
embedding_model: str,
openai_host: str,
):
self.search_client = search_client
self.openai_client = openai_client
+ self.auth_helper = auth_helper
self.query_language = query_language
self.query_speller = query_speller
self.embedding_deployment = embedding_deployment
self.embedding_model = embedding_model
self.openai_host = openai_host
===========changed ref 9===========
# module: tests.test_authenticationhelper
+ def create_authentication_helper(require_access_control: bool = False):
- def create_authentication_helper():
return AuthenticationHelper(
+ search_index=MockSearchIndex,
use_authentication=True,
server_app_id="SERVER_APP",
server_app_secret="SERVER_SECRET",
client_app_id="CLIENT_APP",
tenant_id="TENANT_ID",
+ require_access_control=require_access_control,
- token_cache_path=None,
)
===========changed ref 10===========
+ # module: scripts.auth_common
+ def get_application(auth_headers: Dict[str, str], app_id: str) -> Optional[str]:
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.get(f"https://graph.microsoft.com/v1.0/applications(appId='{app_id}')") as response:
+ if response.status == 200:
+ response_json = await response.json()
+ return response_json["id"]
+
+ return None
+
===========changed ref 11===========
+ # module: scripts.auth_common
+ def update_application(auth_headers: Dict[str, str], object_id: str, app_payload: object):
+ async with aiohttp.ClientSession(headers=auth_headers, timeout=aiohttp.ClientTimeout(total=TIMEOUT)) as session:
+ async with session.patch(
+ f"https://graph.microsoft.com/v1.0/applications/{object_id}", json=app_payload
+ ) as response:
+ if not response.ok:
+ response_json = await response.json()
+ raise Exception(response_json)
+
+ return True
+
===========changed ref 12===========
# module: tests.test_chatapproach
@pytest.fixture
def chat_approach():
return ChatReadRetrieveReadApproach(
search_client=None,
+ auth_helper=None,
openai_client=None,
chatgpt_model="gpt-35-turbo",
chatgpt_deployment="chat",
embedding_deployment="embeddings",
embedding_model="text-",
sourcepage_field="",
content_field="",
query_language="en-us",
query_speller="lexicon",
)
===========changed ref 13===========
+ # module: scripts.auth_common
+ def test_authentication_enabled():
+ use_authentication = os.getenv("AZURE_USE_AUTHENTICATION", "").lower() == "true"
+ require_access_control = os.getenv("AZURE_ENFORCE_ACCESS_CONTROL", "").lower() == "true"
+ if require_access_control and not use_authentication:
+ print("AZURE_ENFORCE_ACCESS_CONTROL is true, but AZURE_USE_AUTHENTICATION is false. Stopping...")
+ return False
+
+ if not use_authentication:
+ return False
+
+ return True
+
|
scripts.prepdocslib.filestrategy/FileStrategy.run
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
c02908c2e074c79fe99563aac3d9b56e717fa19b
|
[BugFix] Fix logic in prepdocs to always update search index (#1066)
|
<12>:<add>
<del> image_embeddings: Optional[List[List[float]]]
<14>:<add> blob_image_embeddings: Optional[List[List[float]]] = None
<15>:<add> blob_image_embeddings = await self.image_embeddings.create_embeddings(blob_sas_uris)
<del> image_embeddings = await self.image_embeddings.create_embeddings(blob_sas_uris)
<16>:<add> await search_manager.update_content(sections, blob_image_embeddings)
<del> await search_manager.update_content(sections, image_embeddings)
|
# module: scripts.prepdocslib.filestrategy
class FileStrategy(Strategy):
def run(self, search_info: SearchInfo):
<0> search_manager = SearchManager(search_info, self.search_analyzer_name, self.use_acls, self.embeddings)
<1> if self.document_action == DocumentAction.Add:
<2> files = self.list_file_strategy.list()
<3> async for file in files:
<4> try:
<5> pages = [page async for page in self.pdf_parser.parse(content=file.content)]
<6> if search_info.verbose:
<7> print(f"Splitting '{file.filename()}' into sections")
<8> sections = [
<9> Section(split_page, content=file, category=self.category)
<10> for split_page in self.text_splitter.split_pages(pages)
<11> ]
<12> image_embeddings: Optional[List[List[float]]]
<13> blob_sas_uris = await self.blob_manager.upload_blob(file)
<14> if self.image_embeddings and blob_sas_uris:
<15> image_embeddings = await self.image_embeddings.create_embeddings(blob_sas_uris)
<16> await search_manager.update_content(sections, image_embeddings)
<17> finally:
<18> if file:
<19> file.close()
<20> elif self.document_action == DocumentAction.Remove:
<21> paths = self.list_file_strategy.list_paths()
<22> async for path in paths:
<23> await self.blob_manager.remove_blob(path)
<24> await search_manager.remove_content(path)
<25> elif self.document_action == DocumentAction.RemoveAll:
<26> await self.blob_manager.remove_blob()
<27> await search_manager.remove_content()
<28>
|
===========unchanged ref 0===========
at: scripts.prepdocslib.blobmanager.BlobManager
upload_blob(file: File) -> Optional[List[str]]
remove_blob(path: Optional[str]=None)
at: scripts.prepdocslib.embeddings.ImageEmbeddings
create_embeddings(blob_urls: List[str]) -> List[List[float]]
at: scripts.prepdocslib.filestrategy
DocumentAction()
at: scripts.prepdocslib.filestrategy.FileStrategy.__init__
self.list_file_strategy = list_file_strategy
self.blob_manager = blob_manager
self.pdf_parser = pdf_parser
self.text_splitter = text_splitter
self.document_action = document_action
self.embeddings = embeddings
self.image_embeddings = image_embeddings
self.search_analyzer_name = search_analyzer_name
self.use_acls = use_acls
self.category = category
at: scripts.prepdocslib.listfilestrategy.ListFileStrategy
list() -> AsyncGenerator[File, None]
list_paths() -> AsyncGenerator[str, None]
at: scripts.prepdocslib.pdfparser.PdfParser
parse(content: IO) -> AsyncGenerator[Page, None]
at: scripts.prepdocslib.searchmanager
Section(split_page: SplitPage, content: File, category: Optional[str]=None)
SearchManager(search_info: SearchInfo, search_analyzer_name: Optional[str]=None, use_acls: bool=False, embeddings: Optional[OpenAIEmbeddings]=None, search_images: bool=False)
at: scripts.prepdocslib.searchmanager.SearchManager
update_content(sections: List[Section], image_embeddings: Optional[List[List[float]]]=None)
remove_content(path: Optional[str]=None)
===========unchanged ref 1===========
at: scripts.prepdocslib.strategy
SearchInfo(endpoint: str, credential: Union[AsyncTokenCredential, AzureKeyCredential], index_name: str, verbose: bool=False)
at: scripts.prepdocslib.strategy.SearchInfo.__init__
self.verbose = verbose
at: scripts.prepdocslib.strategy.Strategy
run(self, search_info: SearchInfo)
at: scripts.prepdocslib.textsplitter.TextSplitter
split_pages(pages: List[Page]) -> Generator[SplitPage, None, None]
at: typing
List = _alias(list, 1, inst=False, name='List')
===========changed ref 0===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
+ def create_client(self) -> AsyncOpenAI:
+ return AsyncOpenAI(api_key=self.credential, organization=self.organization)
+
===========changed ref 1===========
# module: scripts.prepdocslib.embeddings
class OpenAIEmbeddingService(OpenAIEmbeddings):
- def create_embedding_arguments(self) -> dict[str, Any]:
- return {
- "model": self.open_ai_model_name,
- "api_key": self.credential,
- "api_type": "openai",
- "organization": self.organization,
- }
-
|
scripts.prepdocslib.blobmanager/BlobManager.upload_pdf_blob_images
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
6cc78369938e463085ae777bc83a82adc27e40fe
|
[BugFix] Use default font when a specific font cannot be found (#1069)
|
<7>:<add>
<add> font = None
<add> try:
<add> font = ImageFont.truetype("arial.ttf", 20)
<add> except OSError:
<add> try:
<add> font = ImageFont.truetype("/usr/share/fonts/truetype/freefont/FreeMono.ttf", 20)
<add> except OSError:
<add> print("\tUnable to find arial.ttf or FreeMono.ttf, using default font")
<26>:<del> try:
<27>:<del> font = ImageFont.truetype("arial.ttf", 20)
<28>:<del> except OSError:
<29>:<del> font = ImageFont.truetype("/usr/share/fonts/truetype/freefont/FreeMono.ttf", 20)
|
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def upload_pdf_blob_images(
self, service_client: BlobServiceClient, container_client: ContainerClient, file: File
) -> List[str]:
<0> with open(file.content.name, "rb") as reopened_file:
<1> reader = PdfReader(reopened_file)
<2> page_count = len(reader.pages)
<3> doc = fitz.open(file.content.name)
<4> sas_uris = []
<5> start_time = datetime.datetime.now(datetime.timezone.utc)
<6> expiry_time = start_time + datetime.timedelta(days=1)
<7>
<8> for i in range(page_count):
<9> blob_name = BlobManager.blob_image_name_from_file_page(file.content.name, i)
<10> if self.verbose:
<11> print(f"\tConverting page {i} to image and uploading -> {blob_name}")
<12>
<13> doc = fitz.open(file.content.name)
<14> page = doc.load_page(i)
<15> pix = page.get_pixmap()
<16> original_img = Image.frombytes("RGB", [pix.width, pix.height], pix.samples) # type: ignore
<17>
<18> # Create a new image with additional space for text
<19> text_height = 40 # Height of the text area
<20> new_img = Image.new("RGB", (original_img.width, original_img.height + text_height), "white")
<21>
<22> # Paste the original image onto the new image
<23> new_img.paste(original_img, (0, text_height))
<24>
<25> # Draw the text on the white area
<26> try:
<27> font = ImageFont.truetype("arial.ttf", 20)
<28> except OSError:
<29> font = ImageFont.truetype("/usr/share/fonts/truetype/freefont/FreeMono.ttf", 20)
<30> draw = ImageDraw.Draw(new_img)
<31> </s>
|
===========below chunk 0===========
# module: scripts.prepdocslib.blobmanager
class BlobManager:
def upload_pdf_blob_images(
self, service_client: BlobServiceClient, container_client: ContainerClient, file: File
) -> List[str]:
# offset: 1
# 10 pixels from the top and left of the image
x = 10
y = 10
draw.text((x, y), text, font=font, fill="black")
output = io.BytesIO()
new_img.save(output, format="PNG")
output.seek(0)
blob_client = await container_client.upload_blob(blob_name, output, overwrite=True)
if not self.user_delegation_key:
self.user_delegation_key = await service_client.get_user_delegation_key(start_time, expiry_time)
if blob_client.account_name is not None:
sas_token = generate_blob_sas(
account_name=blob_client.account_name,
container_name=blob_client.container_name,
blob_name=blob_client.blob_name,
user_delegation_key=self.user_delegation_key,
permission=BlobSasPermissions(read=True),
expiry=expiry_time,
start=start_time,
)
sas_uris.append(f"{blob_client.url}?{sas_token}")
return sas_uris
|
app.backend.core.authentication/AuthenticationHelper.build_security_filters
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
e518ab0589af8d8704761d85db68ec4dffbfef55
|
bugfix:in app/backend/core/authenticatio.py/119line-'use_oid_security_filter' does not work (#1086)
|
<6>:<add> if (use_oid_security_filter or use_groups_security_filter) and not self.has_auth_fields:
<del> if (use_oid_security_filter or use_oid_security_filter) and not self.has_auth_fields:
|
# module: app.backend.core.authentication
class AuthenticationHelper:
def build_security_filters(self, overrides: dict[str, Any], auth_claims: dict[str, Any]):
<0> # Build different permutations of the oid or groups security filter using OData filters
<1> # https://learn.microsoft.com/azure/search/search-security-trimming-for-azure-search
<2> # https://learn.microsoft.com/azure/search/search-query-odata-filter
<3> use_oid_security_filter = self.require_access_control or overrides.get("use_oid_security_filter")
<4> use_groups_security_filter = self.require_access_control or overrides.get("use_groups_security_filter")
<5>
<6> if (use_oid_security_filter or use_oid_security_filter) and not self.has_auth_fields:
<7> raise AuthError(
<8> error="oids and groups must be defined in the search index to use authentication", status_code=400
<9> )
<10>
<11> oid_security_filter = (
<12> "oids/any(g:search.in(g, '{}'))".format(auth_claims.get("oid") or "") if use_oid_security_filter else None
<13> )
<14> groups_security_filter = (
<15> "groups/any(g:search.in(g, '{}'))".format(", ".join(auth_claims.get("groups") or []))
<16> if use_groups_security_filter
<17> else None
<18> )
<19>
<20> # If only one security filter is specified, return that filter
<21> # If both security filters are specified, combine them with "or" so only 1 security filter needs to pass
<22> # If no security filters are specified, don't return any filter
<23> if oid_security_filter and not groups_security_filter:
<24> return oid_security_filter
<25> elif groups_security_filter and not oid_security_filter:
<26> return groups_security_filter
<27> elif oid_security_filter and groups_security_filter:
<28> return f"({oid_security_filter} or {groups_security</s>
|
===========below chunk 0===========
# module: app.backend.core.authentication
class AuthenticationHelper:
def build_security_filters(self, overrides: dict[str, Any], auth_claims: dict[str, Any]):
# offset: 1
else:
return None
===========unchanged ref 0===========
at: app.backend.core.authentication
AuthError(error, status_code)
at: app.backend.core.authentication.AuthenticationHelper
scope: str = "https://graph.microsoft.com/.default"
at: app.backend.core.authentication.AuthenticationHelper.__init__
self.has_auth_fields = "oids" in field_names and "groups" in field_names
self.has_auth_fields = False
self.require_access_control = False
self.require_access_control = require_access_control
at: typing.Mapping
get(key: _KT, default: Union[_VT_co, _T]) -> Union[_VT_co, _T]
get(key: _KT) -> Optional[_VT_co]
|
tests.test_authenticationhelper/test_auth_setup
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
0b8724ad761c1d93489ff254bc3a66ad3fceedf2
|
Update Authentication to use local storage and easy auth refresh (#1117)
|
<12>:<add> "cache": {"cacheLocation": "localStorage", "storeAuthStateInCookie": False},
<del> "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
|
# module: tests.test_authenticationhelper
def test_auth_setup(mock_confidential_client_success):
<0> helper = create_authentication_helper()
<1> assert helper.get_auth_setup_for_client() == {
<2> "useLogin": True,
<3> "requireAccessControl": False,
<4> "msalConfig": {
<5> "auth": {
<6> "clientId": "CLIENT_APP",
<7> "authority": "https://login.microsoftonline.com/TENANT_ID",
<8> "redirectUri": "/redirect",
<9> "postLogoutRedirectUri": "/",
<10> "navigateToLoginRequestUrl": False,
<11> },
<12> "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
<13> },
<14> "loginRequest": {
<15> "scopes": [".default"],
<16> },
<17> "tokenRequest": {
<18> "scopes": ["api://SERVER_APP/access_as_user"],
<19> },
<20> }
<21>
|
===========unchanged ref 0===========
at: tests.conftest
mock_confidential_client_success(monkeypatch)
at: tests.test_authenticationhelper
create_authentication_helper(require_access_control: bool=False)
|
tests.test_authenticationhelper/test_auth_setup_required_access_control
|
Modified
|
Azure-Samples~azure-search-openai-demo
|
0b8724ad761c1d93489ff254bc3a66ad3fceedf2
|
Update Authentication to use local storage and easy auth refresh (#1117)
|
<12>:<add> "cache": {"cacheLocation": "localStorage", "storeAuthStateInCookie": False},
<del> "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
|
# module: tests.test_authenticationhelper
def test_auth_setup_required_access_control(mock_confidential_client_success):
<0> helper = create_authentication_helper(require_access_control=True)
<1> assert helper.get_auth_setup_for_client() == {
<2> "useLogin": True,
<3> "requireAccessControl": True,
<4> "msalConfig": {
<5> "auth": {
<6> "clientId": "CLIENT_APP",
<7> "authority": "https://login.microsoftonline.com/TENANT_ID",
<8> "redirectUri": "/redirect",
<9> "postLogoutRedirectUri": "/",
<10> "navigateToLoginRequestUrl": False,
<11> },
<12> "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
<13> },
<14> "loginRequest": {
<15> "scopes": [".default"],
<16> },
<17> "tokenRequest": {
<18> "scopes": ["api://SERVER_APP/access_as_user"],
<19> },
<20> }
<21>
|
===========unchanged ref 0===========
at: tests.test_authenticationhelper
create_authentication_helper(require_access_control: bool=False)
===========changed ref 0===========
# module: tests.test_authenticationhelper
def test_auth_setup(mock_confidential_client_success):
helper = create_authentication_helper()
assert helper.get_auth_setup_for_client() == {
"useLogin": True,
"requireAccessControl": False,
"msalConfig": {
"auth": {
"clientId": "CLIENT_APP",
"authority": "https://login.microsoftonline.com/TENANT_ID",
"redirectUri": "/redirect",
"postLogoutRedirectUri": "/",
"navigateToLoginRequestUrl": False,
},
+ "cache": {"cacheLocation": "localStorage", "storeAuthStateInCookie": False},
- "cache": {"cacheLocation": "sessionStorage", "storeAuthStateInCookie": False},
},
"loginRequest": {
"scopes": [".default"],
},
"tokenRequest": {
"scopes": ["api://SERVER_APP/access_as_user"],
},
}
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.