langchain

Commit Graph

Author	SHA1	Message	Date
Bagatur	feb351737c	core[patch]: fix empty OpenAI tools when strict=True (#26287 ) Fix #26232 --------- Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>	2 weeks ago
ccurme	398718e1cb	core[patch]: fix regression in convert_to_openai_tool with instances of Tool (#26327 ) ```python from langchain_core.tools import Tool from langchain_core.utils.function_calling import convert_to_openai_tool def my_function(x: int) -> int: return x + 2 tool = Tool( name="tool_name", func=my_function, description="test description", ) convert_to_openai_tool(tool) ``` Current: ``` {'type': 'function', 'function': {'name': 'tool_name', 'description': 'test description', 'parameters': {'type': 'object', 'properties': {'args': {'type': 'array', 'items': {}}, 'config': {'type': 'object', 'properties': {'tags': {'type': 'array', 'items': {'type': 'string'}}, 'metadata': {'type': 'object'}, 'callbacks': {'anyOf': [{'type': 'array', 'items': {}}, {}]}, 'run_name': {'type': 'string'}, 'max_concurrency': {'type': 'integer'}, 'recursion_limit': {'type': 'integer'}, 'configurable': {'type': 'object'}, 'run_id': {'type': 'string', 'format': 'uuid'}}}, 'kwargs': {'type': 'object'}}, 'required': ['config']}}} ``` Here: ``` {'type': 'function', 'function': {'name': 'tool_name', 'description': 'test description', 'parameters': {'properties': {'__arg1': {'title': '__arg1', 'type': 'string'}}, 'required': ['__arg1'], 'type': 'object'}}} ```	2 weeks ago
이규민	7feae62ad7	core[patch]: Support non ASCII characters in tool output if user doesn't output string (#26319 ) ### simple modify core: add supporting non english character target issue is #26315 same issue on langgraph - https://github.com/langchain-ai/langgraph/issues/1504	2 weeks ago
William FH	b993172702	Keyword-like runnable config (#26295 )	2 weeks ago
Nuno Campos	212c688ee0	core[minor]: Remove serialized manifest from tracing requests for non-llm runs (#26270 ) - This takes a long time to compute, isn't used, and currently called on every invocation of every chain/retriever/etc	2 weeks ago
Vadym Barda	bab9de581c	core[patch]: wrap mermaid node names w/ markdown in <p> tag (#26235 ) This fixes the issue where `__start__` and `__end__` node labels are being interpreted as markdown, as of the most recent Mermaid update	2 weeks ago
Vadym Barda	1b3bd52e0e	core[patch]: fix edge labels for mermaid graphs (#26201 )	2 weeks ago
Bagatur	dba308447d	fmt	3 weeks ago
Bagatur	576574c82c	fmt	3 weeks ago
Eugene Yurtsev	fa8402ea09	core[minor]: Add support for multiple env keys for secrets_from_env (#25971 ) - Add support to look up secret using more than one env variable - Add overload to help mypy Needed for https://github.com/langchain-ai/langchain/pull/25491	3 weeks ago
Bagatur	5b99bb2437	docs: fix bullet list spacing (#25950 ) Fix #25935	3 weeks ago
Bagatur	933bc0d6ff	core[patch]: support additional kwargs on StructuredPrompt (#25645 )	3 weeks ago
Nuno Campos	464dae8ac2	core: Include global variables in variables found by get_function_nonlocals (#25936 ) Thank you for contributing to LangChain! - [ ] PR title: "package: description" - Where "package" is whichever of langchain, community, core, experimental, etc. is being modified. Use "docs: ..." for purely docs changes, "templates: ..." for template changes, "infra: ..." for CI changes. - Example: "community: add foobar LLM" - [ ] PR message: *Delete this entire checklist* and replace with - Description: a description of the change - Issue: the issue # it fixes, if applicable - Dependencies: any dependencies required for this change - Twitter handle: if your PR gets announced, and you'd like a mention, we'll gladly shout you out! - [ ] Add tests and docs: If you're adding a new integration, please include 1. a test for the integration, preferably unit tests that do not rely on network access, 2. an example notebook showing its use. It lives in `docs/docs/integrations` directory. - [ ] Lint and test: Run `make format`, `make lint` and `make test` from the root of the package(s) you've modified. See contribution guidelines for more: https://python.langchain.com/docs/contributing/ Additional guidelines: - Make sure optional dependencies are imported within a function. - Please do not add dependencies to pyproject.toml files (even optional ones) unless they are required for unit tests. - Most PRs should not touch more than one package. - Changes should be backwards compatible. - If you are adding something to community, do not re-import it in langchain. If no one reviews your PR within a few days, please @-mention one of baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.	3 weeks ago
Bagatur	d19e074374	core[patch]: handle serializable fields that cant be converted to bool (#25903 )	3 weeks ago
Christophe Bornet	0a752a74cc	community[patch], docs: Add API reference doc for GraphVectorStore (#25751 )	3 weeks ago
Bagatur	fabd3295fa	core[patch]: dont mutate merged lists/dicts (#25858 ) Update merging utils to - not mutate objects - have special handling to 'type' keys in dicts	4 weeks ago
Erick Friis	c8b8335b82	core: prompt variable error msg (#25787 )	4 weeks ago
Christophe Bornet	ff0df5ea15	core[patch]: Add B(bugbear) ruff rules (#25520 ) Co-authored-by: Bagatur <baskaryan@gmail.com>	4 weeks ago
ccurme	3c784e10a8	docs: improve docs for InMemoryVectorStore (#25786 ) Closes https://github.com/langchain-ai/langchain/issues/25775	4 weeks ago
James Espichan Vilca	080741d336	core[patch]: Fix type for inner input in base prompts (#25713 ) Thank you for contributing to LangChain! - [ ] PR title: "langchain-core: Fix type" - The file to modify is located in /libs/core/langchain_core/prompts/base.py - [ ] PR message: - Description: The change is a type for the inner input variable, the type go from dict to Any. This change is required since the method _validate input expects a type that is not only a dictionary. - Dependencies: There are no dependencies for this change - [ ] Add tests and docs: 1. A test is not needed. This error occurs because I overrode a portion of the _validate_input method, which is causing a 'beartype' to raise an error.	1 month ago
Christophe Bornet	7f1e444efa	partners: Use simsimd types (#25299 ) The simsimd package [now has types](https://github.com/ashvardanian/SimSIMD/releases/tag/v5.0.0)	1 month ago
Erick Friis	b28bc252c4	core[patch]: mmr util (#25689 )	1 month ago
Erick Friis	6096c80b71	core: pydantic output parser streaming fix (#24415 ) Co-authored-by: Bagatur <baskaryan@gmail.com>	1 month ago
Eugene Yurtsev	c316361115	core[patch]: Add _api.rename_parameter to support renaming of parameters in functions (#25101 ) Add ability to rename paramerters in function signatures ```python @rename_parameter(since="2.0.0", removal="3.0.0", old="old_name", new="new_name") def foo(new_name: str) -> str: """original doc""" return new_name ``` --------- Co-authored-by: Bagatur <baskaryan@gmail.com> Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>	1 month ago
Yusuke Fukasawa	0258cb96fa	core[patch]: add additionalProperties recursively to oai function if strict (#25169 ) Hello. First of all, thank you for maintaining such a great project. ## Description In https://github.com/langchain-ai/langchain/pull/25123, support for structured_output is added. However, `"additionalProperties": false` needs to be included at all levels when a nested object is generated. error from current code: https://gist.github.com/fufufukakaka/e9b475300e6934853d119428e390f204 ``` BadRequestError: Error code: 400 - {'error': {'message': "Invalid schema for response_format 'JokeWithEvaluation': In context=('properties', 'self_evaluation'), 'additionalProperties' is required to be supplied and to be false", 'type': 'invalid_request_error', 'param': 'response_format', 'code': None}} ``` Reference: [Introducing Structured Outputs in the API](https://openai.com/index/introducing-structured-outputs-in-the-api/) ```json { "model": "gpt-4o-2024-08-06", "messages": [ { "role": "system", "content": "You are a helpful math tutor." }, { "role": "user", "content": "solve 8x + 31 = 2" } ], "response_format": { "type": "json_schema", "json_schema": { "name": "math_response", "strict": true, "schema": { "type": "object", "properties": { "steps": { "type": "array", "items": { "type": "object", "properties": { "explanation": { "type": "string" }, "output": { "type": "string" } }, "required": ["explanation", "output"], "additionalProperties": false } }, "final_answer": { "type": "string" } }, "required": ["steps", "final_answer"], "additionalProperties": false } } } } ``` In the current code, `"additionalProperties": false` is only added at the last level. This PR introduces the `_add_additional_properties_key` function, which recursively adds `"additionalProperties": false` to the entire JSON schema for the request. Twitter handle: `@fukkaa1225` Thank you! --------- Co-authored-by: Bagatur <baskaryan@gmail.com>	1 month ago
Christophe Bornet	ee98da4f4e	core[patch]: Add UP(upgrade) ruff rules (#25358 )	1 month ago
William FH	294f7fcb38	core[patch]: Remove different parent run id warning (#25683 )	1 month ago
Vadym Barda	46d344c33d	core[patch]: support drawing nested subgraphs in draw_mermaid (#25581 ) Previously the code was able to only handle a single level of nesting for subgraphs in mermaid. This change adds support for arbitrary nesting of subgraphs.	1 month ago
CastaChick	7d13a2f958	core[patch]: add option to specify the chunk separator in `merge_message_runs` (#24783 ) Description: LLM will stop generating text even in the middle of a sentence if `finish_reason` is `length` (for OpenAI) or `stop_reason` is `max_tokens` (for Anthropic). To obtain longer outputs from LLM, we should call the message generation API multiple times and merge the results into the text to circumvent the API's output token limit. The extra line breaks forced by the `merge_message_runs` function when seamlessly merging messages can be annoying, so I added the option to specify the chunk separator. Issue: No corresponding issues. Dependencies: No dependencies required. Twitter handle: @hanama_chem https://x.com/hanama_chem --------- Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com> Co-authored-by: Bagatur <baskaryan@gmail.com>	1 month ago
Ivan	5b9290a449	Fix UnionType type var replacement (#25566 ) [langchain_core] Fix UnionType type var replacement - Added types.UnionType to typing.Union mapping Type replacement cause `TypeError: 'type' object is not subscriptable` if any of union type comes as function `_py_38_safe_origin` return `types.UnionType` instead of `typing.Union` ```python >>> from types import UnionType >>> from typing import Union, get_origin >>> type_ = get_origin(str \| None) >>> type_ <class 'types.UnionType'> >>> UnionType[(str, None)] Traceback (most recent call last): File "<stdin>", line 1, in <module> TypeError: 'type' object is not subscriptable >>> Union[(str, None)] typing.Optional[str] ``` --------- Co-authored-by: Chester Curme <chester.curme@gmail.com>	1 month ago
William FH	8230ba47f3	core[patch]: Improve some error messages and add another test for checking RunnableWithMessageHistory (#25209 ) Also add more useful error messages. --------- Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>	1 month ago
Bagatur	0bc3845e1e	core[patch]: support oai dicts as messages (#25621 ) and update langsmtih example selector docs	1 month ago
Erick Friis	e37caa9b9a	core: fix fallback context overwriting (#25550 ) fixes #25337	1 month ago
Bagatur	8a71f1b41b	core[minor]: add langsmith document loader (#25493 ) needs tests	1 month ago
Bagatur	4bd005adb6	core[patch]: Allow bound models as token_counter in trim_messages (#25563 )	1 month ago
Erick Friis	e01c6789c4	core,community: add beta decorator to missed GraphVectorStore extensions (#25562 )	1 month ago
ccurme	b83f1eb0d5	core, partners: implement standard tracing params for LLMs (#25410 )	1 month ago
Eugene Yurtsev	e18511bb22	core[minor], anthropic[patch]: Upgrade @root_validator usage to be consistent with pydantic 2 (#25457 ) anthropic: Upgrade `@root_validator` usage to be consistent with pydantic 2 core: support looking up multiple keys from env in from_env factory	1 month ago
Eugene Yurtsev	eb3870e9d8	fireworks[patch]: Upgrade @root_validators to be pydantic 2 compliant (#25443 ) Update @root_validators to be pydantic 2 compliant	1 month ago
William FH	75ae585deb	Merge support for group manager (#25360 )	1 month ago
Bagatur	2494cecabf	core[patch]: tool import fix (#25419 )	1 month ago
Chengyu Yan	d0ad713937	core: fix issue#24660, slove error messages about `ValueError` when use model with history (#25183 ) - Description: This PR will slove error messages about `ValueError` when use model with history. Detail in #24660. #22933 causes that `langchain_core.runnables.history.RunnableWithMessageHistory._get_output_messages` miss type check of `output_val` if `output_val` is `False`. After running `RunnableWithMessageHistory._is_not_async`, `output` is `False`. `249945a572/libs/core/langchain_core/runnables/history.py (L323-L334)` `15a36dd0a2/libs/core/langchain_core/runnables/history.py (L461-L471)` ~~I suggest that `_get_output_messages` return empty list when `output_val == False`.~~ - Issue: - #24660 - Dependencies:: No Change. --------- Co-authored-by: Bagatur <baskaryan@gmail.com> Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>	1 month ago
Bagatur	493e474063	docs: udpated api reference (#25172 ) - Move the API reference into the vercel build - Update api reference organization and styling	1 month ago
Eugene Yurtsev	dc51cc5690	core[minor]: Prevent PydanticOutputParser from encoding schema as ASCII (#25386 ) This allows users to provide parameter descriptions in the pydantic models in other languages. Continuing this PR: https://github.com/langchain-ai/langchain/pull/24809	1 month ago
ccurme	27690506d0	multiple: update removal targets (#25361 )	1 month ago
Eugene Yurtsev	056c7c2983	core[patch]: Update API reference for fake embeddings (#25313 ) Issue: https://github.com/langchain-ai/langchain/issues/24856 Using the same template for the fake embeddings in langchain_core as used in the integrations.	1 month ago
ccurme	e77eeee6ee	core[patch]: add standard tracing params for retrievers (#25240 )	1 month ago
gbaian10	aa2722cbe2	docs: update numbering of items in docstring (#25267 ) A problem similar to #25093 . Co-authored-by: ccurme <chester.curme@gmail.com>	1 month ago
Eugene Yurtsev	6dd9f053e3	core[patch]: Deprecating beta upsert APIs in vectorstore (#25069 ) This PR deprecates the beta upsert APIs in vectorstore. We'll introduce them in a V2 abstraction instead to keep the existing vectorstore implementations lighter weight. The main problem with the existing APIs is that it's a bit more challenging to implement the correct behavior w/ respect to IDs since ID can be present in both the function signature and as an optional attribute on the document object. But VectorStores that pass the standard tests should have implemented the semantics properly! --------- Co-authored-by: Erick Friis <erick@langchain.dev>	1 month ago
blueoom	c3ced4c6ce	core[patch]: use time.monotonic() instead time.time() in InMemoryRateLimiter Description: The get time point method in the _consume() method of core.rate_limiters.InMemoryRateLimiter uses time.time(), which can be affected by system time backwards. Therefore, it is recommended to use the monotonically increasing monotonic() to obtain the time ```python with self._consume_lock: now = time.time() # time.time() -> time.monotonic() # initialize on first call to avoid a burst if self.last is None: self.last = now elapsed = now - self.last # when use time.time(), elapsed may be negative when system time backwards ```	1 month ago

1 2 3 4 5 ...

630 Commits (feb351737c55a45344b9fed0d22a98996ee18362)