infra: add -p to mkdir in lint steps (langchain-ai#17013)

Previously, if this did not find a mypy cache then it wouldnt run

this makes it always run

adding mypy ignore comments with existing uncaught issues to unblock other prs

---------

Co-authored-by: Erick Friis <[email protected]>
Co-authored-by: Bagatur <[email protected]>

.github/workflows/_lint.yml

@@ -86,7 +86,7 @@ jobs:
         with:
           path: |
             ${{ env.WORKDIR }}/.mypy_cache
-          key: mypy-lint-${{ runner.os }}-${{ runner.arch }}-py${{ matrix.python-version }}-${{ inputs.working-directory }}-${{ hashFiles(format('{0}/poetry.lock', env.WORKDIR)) }}
+          key: mypy-lint-${{ runner.os }}-${{ runner.arch }}-py${{ matrix.python-version }}-${{ inputs.working-directory }}-${{ hashFiles(format('{0}/poetry.lock', inputs.working-directory)) }}
 
 
       - name: Analysing the code with our lint
@@ -105,7 +105,7 @@ jobs:
         # It doesn't matter how you change it, any change will cause a cache-bust.
         working-directory: ${{ inputs.working-directory }}
         run: |
-          poetry install --with test,test_integration
+          poetry install --with test
 
       - name: Get .mypy_cache_test to speed up mypy
         uses: actions/cache@v3
@@ -114,7 +114,7 @@ jobs:
         with:
           path: |
             ${{ env.WORKDIR }}/.mypy_cache_test
-          key: mypy-test-${{ runner.os }}-${{ runner.arch }}-py${{ matrix.python-version }}-${{ inputs.working-directory }}-${{ hashFiles(format('{0}/poetry.lock', env.WORKDIR)) }}
+          key: mypy-test-${{ runner.os }}-${{ runner.arch }}-py${{ matrix.python-version }}-${{ inputs.working-directory }}-${{ hashFiles(format('{0}/poetry.lock', inputs.working-directory)) }}
 
       - name: Analysing the code with our lint
         working-directory: ${{ inputs.working-directory }}

libs/community/Makefile

@@ -41,7 +41,7 @@ lint lint_diff lint_package lint_tests:
 	poetry run ruff .
 	[ "$(PYTHON_FILES)" = "" ] || poetry run ruff format $(PYTHON_FILES) --diff
 	[ "$(PYTHON_FILES)" = "" ] || poetry run ruff --select I $(PYTHON_FILES)
-	[ "$(PYTHON_FILES)" = "" ] || mkdir -p $(MYPY_CACHE) || poetry run mypy $(PYTHON_FILES) --cache-dir $(MYPY_CACHE)
+	[ "$(PYTHON_FILES)" = "" ] || mkdir -p $(MYPY_CACHE) && poetry run mypy $(PYTHON_FILES) --cache-dir $(MYPY_CACHE)
 
 format format_diff:
 	poetry run ruff format $(PYTHON_FILES)

libs/community/langchain_community/agent_toolkits/openapi/planner.py

@@ -84,7 +84,7 @@ def _run(self, text: str) -> str:
             raise e
         data_params = data.get("params")
         response = self.requests_wrapper.get(data["url"], params=data_params)
-        response = response[: self.response_length]
+        response = response[: self.response_length]  # type: ignore[index]
         return self.llm_chain.predict(
             response=response, instructions=data["output_instructions"]
         ).strip()
@@ -115,7 +115,7 @@ def _run(self, text: str) -> str:
         except json.JSONDecodeError as e:
             raise e
         response = self.requests_wrapper.post(data["url"], data["data"])
-        response = response[: self.response_length]
+        response = response[: self.response_length]  # type: ignore[index]
         return self.llm_chain.predict(
             response=response, instructions=data["output_instructions"]
         ).strip()
@@ -146,7 +146,7 @@ def _run(self, text: str) -> str:
         except json.JSONDecodeError as e:
             raise e
         response = self.requests_wrapper.patch(data["url"], data["data"])
-        response = response[: self.response_length]
+        response = response[: self.response_length]  # type: ignore[index]
         return self.llm_chain.predict(
             response=response, instructions=data["output_instructions"]
         ).strip()
@@ -177,7 +177,7 @@ def _run(self, text: str) -> str:
         except json.JSONDecodeError as e:
             raise e
         response = self.requests_wrapper.put(data["url"], data["data"])
-        response = response[: self.response_length]
+        response = response[: self.response_length]  # type: ignore[index]
         return self.llm_chain.predict(
             response=response, instructions=data["output_instructions"]
         ).strip()
@@ -209,7 +209,7 @@ def _run(self, text: str) -> str:
         except json.JSONDecodeError as e:
             raise e
         response = self.requests_wrapper.delete(data["url"])
-        response = response[: self.response_length]
+        response = response[: self.response_length]  # type: ignore[index]
         return self.llm_chain.predict(
             response=response, instructions=data["output_instructions"]
         ).strip()

libs/community/langchain_community/agent_toolkits/sql/base.py

@@ -177,12 +177,12 @@ def create_sql_agent(
     elif agent_type == AgentType.OPENAI_FUNCTIONS:
         if prompt is None:
             messages = [
-                SystemMessage(content=prefix),
+                SystemMessage(content=prefix),  # type: ignore[arg-type]
                 HumanMessagePromptTemplate.from_template("{input}"),
                 AIMessage(content=suffix or SQL_FUNCTIONS_SUFFIX),
                 MessagesPlaceholder(variable_name="agent_scratchpad"),
             ]
-            prompt = ChatPromptTemplate.from_messages(messages)
+            prompt = ChatPromptTemplate.from_messages(messages)  # type: ignore[arg-type]
         agent = RunnableAgent(
             runnable=create_openai_functions_agent(llm, tools, prompt),
             input_keys_arg=["input"],
@@ -191,12 +191,12 @@ def create_sql_agent(
     elif agent_type == "openai-tools":
         if prompt is None:
             messages = [
-                SystemMessage(content=prefix),
+                SystemMessage(content=prefix),  # type: ignore[arg-type]
                 HumanMessagePromptTemplate.from_template("{input}"),
                 AIMessage(content=suffix or SQL_FUNCTIONS_SUFFIX),
                 MessagesPlaceholder(variable_name="agent_scratchpad"),
             ]
-            prompt = ChatPromptTemplate.from_messages(messages)
+            prompt = ChatPromptTemplate.from_messages(messages)  # type: ignore[arg-type]
         agent = RunnableMultiActionAgent(
             runnable=create_openai_tools_agent(llm, tools, prompt),
             input_keys_arg=["input"],

libs/community/langchain_community/callbacks/mlflow_callback.py

@@ -723,7 +723,7 @@ def _create_session_analysis_df(self) -> Any:
         )
         return session_analysis_df
 
-    def _contain_llm_records(self):
+    def _contain_llm_records(self):  # type: ignore[no-untyped-def]
         return bool(self.records["on_llm_start_records"])
 
     def flush_tracker(self, langchain_asset: Any = None, finish: bool = False) -> None:

libs/community/langchain_community/chat_message_histories/elasticsearch.py

@@ -47,7 +47,7 @@ def __init__(
     ):
         self.index: str = index
         self.session_id: str = session_id
-        self.ensure_ascii: bool = esnsure_ascii
+        self.ensure_ascii: bool = esnsure_ascii  # type: ignore[assignment]
 
         # Initialize Elasticsearch client from passed client arg or connection info
         if es_connection is not None:

libs/community/langchain_community/chat_message_histories/tidb.py

@@ -40,7 +40,7 @@ def __init__(
         self.session_id = session_id
         self.table_name = table_name
         self.earliest_time = earliest_time
-        self.cache = []
+        self.cache = []  # type: ignore[var-annotated]
 
         # Set up SQLAlchemy engine and session
         self.engine = create_engine(connection_string)
@@ -102,7 +102,7 @@ def _load_messages_to_cache(self) -> None:
             logger.error(f"Error loading messages to cache: {e}")
 
     @property
-    def messages(self) -> List[BaseMessage]:
+    def messages(self) -> List[BaseMessage]:  # type: ignore[override]
         """returns all messages"""
         if len(self.cache) == 0:
             self.reload_cache()

libs/community/langchain_community/chat_message_histories/zep.py

@@ -149,7 +149,7 @@ def _get_memory(self) -> Optional[Memory]:
             return None
         return zep_memory
 
-    def add_user_message(
+    def add_user_message(  # type: ignore[override]
         self, message: str, metadata: Optional[Dict[str, Any]] = None
     ) -> None:
         """Convenience method for adding a human message string to the store.
@@ -160,7 +160,7 @@ def add_user_message(
         """
         self.add_message(HumanMessage(content=message), metadata=metadata)
 
-    def add_ai_message(
+    def add_ai_message(  # type: ignore[override]
         self, message: str, metadata: Optional[Dict[str, Any]] = None
     ) -> None:
         """Convenience method for adding an AI message string to the store.

libs/community/langchain_community/chat_models/azureml_endpoint.py

@@ -20,7 +20,7 @@
 
 
 class LlamaContentFormatter(ContentFormatterBase):
-    def __init__(self):
+    def __init__(self):  # type: ignore[no-untyped-def]
         raise TypeError(
             "`LlamaContentFormatter` is deprecated for chat models. Use "
             "`LlamaChatContentFormatter` instead."
@@ -72,7 +72,7 @@ def _convert_message_to_dict(message: BaseMessage) -> Dict:
     def supported_api_types(self) -> List[AzureMLEndpointApiType]:
         return [AzureMLEndpointApiType.realtime, AzureMLEndpointApiType.serverless]
 
-    def format_request_payload(
+    def format_request_payload(  # type: ignore[override]
         self,
         messages: List[BaseMessage],
         model_kwargs: Dict,
@@ -98,17 +98,17 @@ def format_request_payload(
             raise ValueError(
                 f"`api_type` {api_type} is not supported by this formatter"
             )
-        return str.encode(request_payload)
+        return str.encode(request_payload)  # type: ignore[return-value]
 
-    def format_response_payload(
+    def format_response_payload(  # type: ignore[override]
         self, output: bytes, api_type: AzureMLEndpointApiType
     ) -> ChatGeneration:
         """Formats response"""
         if api_type == AzureMLEndpointApiType.realtime:
             try:
                 choice = json.loads(output)["output"]
             except (KeyError, IndexError, TypeError) as e:
-                raise ValueError(self.format_error_msg.format(api_type=api_type)) from e
+                raise ValueError(self.format_error_msg.format(api_type=api_type)) from e  # type: ignore[union-attr]
             return ChatGeneration(
                 message=BaseMessage(
                     content=choice.strip(),
@@ -125,7 +125,7 @@ def format_response_payload(
                         "model. Expected `dict` but `{type(choice)}` was received."
                     )
             except (KeyError, IndexError, TypeError) as e:
-                raise ValueError(self.format_error_msg.format(api_type=api_type)) from e
+                raise ValueError(self.format_error_msg.format(api_type=api_type)) from e  # type: ignore[union-attr]
             return ChatGeneration(
                 message=BaseMessage(
                     content=choice["message"]["content"].strip(),

libs/community/langchain_community/chat_models/edenai.py

@@ -175,7 +175,7 @@ def _stream(
         """Call out to EdenAI's chat endpoint."""
         url = f"{self.edenai_api_url}/text/chat/stream"
         headers = {
-            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",
+            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",  # type: ignore[union-attr]
             "User-Agent": self.get_user_agent(),
         }
         formatted_data = _format_edenai_messages(messages=messages)
@@ -216,7 +216,7 @@ async def _astream(
     ) -> AsyncIterator[ChatGenerationChunk]:
         url = f"{self.edenai_api_url}/text/chat/stream"
         headers = {
-            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",
+            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",  # type: ignore[union-attr]
             "User-Agent": self.get_user_agent(),
         }
         formatted_data = _format_edenai_messages(messages=messages)
@@ -265,7 +265,7 @@ def _generate(
 
         url = f"{self.edenai_api_url}/text/chat"
         headers = {
-            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",
+            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",  # type: ignore[union-attr]
             "User-Agent": self.get_user_agent(),
         }
         formatted_data = _format_edenai_messages(messages=messages)
@@ -323,7 +323,7 @@ async def _agenerate(
 
         url = f"{self.edenai_api_url}/text/chat"
         headers = {
-            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",
+            "Authorization": f"Bearer {self.edenai_api_key.get_secret_value()}",  # type: ignore[union-attr]
             "User-Agent": self.get_user_agent(),
         }
         formatted_data = _format_edenai_messages(messages=messages)

libs/community/langchain_community/chat_models/ernie.py

@@ -214,7 +214,7 @@ def _create_chat_result(self, response: Mapping[str, Any]) -> ChatResult:
         generations = [
             ChatGeneration(
                 message=AIMessage(
-                    content=response.get("result"),
+                    content=response.get("result"),  # type: ignore[arg-type]
                     additional_kwargs={**additional_kwargs},
                 )
             )

libs/community/langchain_community/chat_models/gpt_router.py

@@ -56,7 +56,7 @@ class GPTRouterModel(BaseModel):
     provider_name: str
 
 
-def get_ordered_generation_requests(
+def get_ordered_generation_requests(  # type: ignore[no-untyped-def, no-untyped-def]
     models_priority_list: List[GPTRouterModel], **kwargs
 ):
     """
@@ -100,7 +100,7 @@ def completion_with_retry(
     models_priority_list: List[GPTRouterModel],
     run_manager: Optional[CallbackManagerForLLMRun] = None,
     **kwargs: Any,
-) -> Union[GenerationResponse, Generator[ChunkedGenerationResponse]]:
+) -> Union[GenerationResponse, Generator[ChunkedGenerationResponse]]:  # type: ignore[type-arg]
     """Use tenacity to retry the completion call."""
     retry_decorator = _create_retry_decorator(llm, run_manager=run_manager)
 
@@ -122,7 +122,7 @@ async def acompletion_with_retry(
     models_priority_list: List[GPTRouterModel],
     run_manager: Optional[AsyncCallbackManagerForLLMRun] = None,
     **kwargs: Any,
-) -> Union[GenerationResponse, AsyncGenerator[ChunkedGenerationResponse]]:
+) -> Union[GenerationResponse, AsyncGenerator[ChunkedGenerationResponse]]:  # type: ignore[type-arg]
     """Use tenacity to retry the async completion call."""
 
     retry_decorator = _create_retry_decorator(llm, run_manager=run_manager)
@@ -282,7 +282,7 @@ async def _agenerate(
         )
         return self._create_chat_result(response)
 
-    def _create_chat_generation_chunk(
+    def _create_chat_generation_chunk(  # type: ignore[no-untyped-def, no-untyped-def]
         self, data: Mapping[str, Any], default_chunk_class
     ):
         chunk = _convert_delta_to_message_chunk(
@@ -293,7 +293,7 @@ def _create_chat_generation_chunk(
             dict(finish_reason=finish_reason) if finish_reason is not None else None
         )
         default_chunk_class = chunk.__class__
-        chunk = ChatGenerationChunk(message=chunk, generation_info=generation_info)
+        chunk = ChatGenerationChunk(message=chunk, generation_info=generation_info)  # type: ignore[assignment]
         return chunk, default_chunk_class
 
     def _stream(

libs/community/langchain_community/chat_models/huggingface.py

@@ -144,7 +144,7 @@ def _resolve_model_id(self) -> None:
 
         elif isinstance(self.llm, HuggingFaceHub):
             # no need to look up model_id for HuggingFaceHub LLM
-            self.model_id = self.llm.repo_id
+            self.model_id = self.llm.repo_id  # type: ignore[assignment]
             return
 
         else:

libs/community/langchain_community/chat_models/konko.py

@@ -169,7 +169,7 @@ def get_available_models(
         }
 
         if openai_api_key:
-            headers["X-OpenAI-Api-Key"] = openai_api_key.get_secret_value()
+            headers["X-OpenAI-Api-Key"] = openai_api_key.get_secret_value()  # type: ignore[union-attr]
 
         models_response = requests.get(models_url, headers=headers)
 

libs/community/langchain_community/chat_models/ollama.py

@@ -74,10 +74,10 @@ def _format_message_as_text(self, message: BaseMessage) -> str:
         if isinstance(message, ChatMessage):
             message_text = f"\n\n{message.role.capitalize()}: {message.content}"
         elif isinstance(message, HumanMessage):
-            if message.content[0].get("type") == "text":
-                message_text = f"[INST] {message.content[0]['text']} [/INST]"
-            elif message.content[0].get("type") == "image_url":
-                message_text = message.content[0]["image_url"]["url"]
+            if message.content[0].get("type") == "text":  # type: ignore[union-attr]
+                message_text = f"[INST] {message.content[0]['text']} [/INST]"  # type: ignore[index]
+            elif message.content[0].get("type") == "image_url":  # type: ignore[union-attr]
+                message_text = message.content[0]["image_url"]["url"]  # type: ignore[index, index]
         elif isinstance(message, AIMessage):
             message_text = f"{message.content}"
         elif isinstance(message, SystemMessage):
@@ -112,11 +112,11 @@ def _convert_messages_to_ollama_messages(
                 content = message.content
             else:
                 for content_part in message.content:
-                    if content_part.get("type") == "text":
-                        content += f"\n{content_part['text']}"
-                    elif content_part.get("type") == "image_url":
-                        if isinstance(content_part.get("image_url"), str):
-                            image_url_components = content_part["image_url"].split(",")
+                    if content_part.get("type") == "text":  # type: ignore[union-attr]
+                        content += f"\n{content_part['text']}"  # type: ignore[index]
+                    elif content_part.get("type") == "image_url":  # type: ignore[union-attr]
+                        if isinstance(content_part.get("image_url"), str):  # type: ignore[union-attr]
+                            image_url_components = content_part["image_url"].split(",")  # type: ignore[index]
                             # Support data:image/jpeg;base64,<image> format
                             # and base64 strings
                             if len(image_url_components) > 1:
@@ -142,7 +142,7 @@ def _convert_messages_to_ollama_messages(
                 }
             )
 
-        return ollama_messages
+        return ollama_messages  # type: ignore[return-value]
 
     def _create_chat_stream(
         self,
@@ -337,7 +337,7 @@ async def _astream(
                             verbose=self.verbose,
                         )
         except OllamaEndpointNotFoundError:
-            async for chunk in self._legacy_astream(messages, stop, **kwargs):
+            async for chunk in self._legacy_astream(messages, stop, **kwargs):  # type: ignore[attr-defined]
                 yield chunk
 
     @deprecated("0.0.3", alternative="_stream")

libs/community/langchain_community/chat_models/tongyi.py

@@ -197,7 +197,7 @@ def _default_params(self) -> Dict[str, Any]:
         return {
             "model": self.model_name,
             "top_p": self.top_p,
-            "api_key": self.dashscope_api_key.get_secret_value(),
+            "api_key": self.dashscope_api_key.get_secret_value(),  # type: ignore[union-attr]
             "result_format": "message",
             **self.model_kwargs,
         }

libs/community/langchain_community/chat_models/vertexai.py

@@ -121,7 +121,7 @@ def _convert_to_prompt(part: Union[str, Dict]) -> Part:
             elif path.startswith("data:image/"):
                 # extract base64 component from image uri
                 try:
-                    encoded = re.search(r"data:image/\w{2,4};base64,(.*)", path).group(
+                    encoded = re.search(r"data:image/\w{2,4};base64,(.*)", path).group(  # type: ignore[union-attr]
                         1
                     )
                 except AttributeError: