Commit Graph

860 Commits

Author SHA1 Message Date
William Fu-Hinthorn
f7078e08da Support message trimming on single messages 2024-10-29 18:51:23 -07:00
Erick Friis
ab4dab9a0c core: fix batch race condition in FakeListChatModel (#26924)
fixed #26273
2024-10-03 23:14:31 +00:00
Bagatur
87fc5ce688 core[patch]: exclude model cache from ser (#27086) 2024-10-03 22:00:31 +00:00
Bagatur
546dc44da5 core[patch]: add UsageMetadata details (#27072) 2024-10-03 20:36:17 +00:00
Eugene Yurtsev
635c55c039 core[patch]: Release 0.3.8 (#27046)
0.3.8 release for core
2024-10-02 16:58:38 +00:00
Eugene Yurtsev
74bf620e97 core[patch]: Support injected tool args that are arbitrary types (#27045)
This adds support for inject tool args that are arbitrary types when
used with pydantic 2.

We'll need to add similar logic on the v1 path, and potentially mirror
the config from the original model when we're doing the subset.
2024-10-02 12:50:58 -04:00
ccurme
9d10151123 core[patch]: fix init of RunnableAssign (#26903)
Example in API ref currently raises ValidationError.

Resolves https://github.com/langchain-ai/langchain/issues/26862
2024-10-01 14:21:54 -04:00
Erick Friis
7ecd720120 multiple: update docs urls to latest 2 (#26837) 2024-09-30 17:37:07 -07:00
federico-pisanu
2538963945 core[patch]: improve index/aindex api when batch_size<n_docs (#25754)
- **Description:** prevent index function to re-index entire source
document even if nothing has changed.
- **Issue:** #22135

I worked on a solution to this issue that is a compromise between being
cheap and being fast.
In the previous code, when batch_size is greater than the number of docs
from a certain source almost the entire source is deleted (all documents
from that source except for the documents in the first batch)
My solution deletes documents from vector store and record manager only
if at least one document has changed for that source.

Hope this can help!

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2024-09-30 20:57:41 +00:00
Eugene Yurtsev
7fde2791dc core[patch]: Add kwargs to Runnable (#27008)
Fixes #26685

---------

Co-authored-by: Tibor Reiss <tibor.reiss@gmail.com>
2024-09-30 16:45:29 -04:00
Mohammad Mohtashim
e12f570ead Merge pull request #26794
* [chore]: Agent Observation should be casted to string to avoid errors

* Merge branch 'master' into fix_observation_type_streaming

* [chore]: Using Json.dumps

* [chore]: Exact same logic as  when casting agent oobservation to string
2024-09-30 15:54:51 -04:00
Bagatur
34bd718fe1 core[patch]: Release 0.3.7 (#27004) 2024-09-30 18:52:42 +00:00
Bagatur
248be02259 core[patch]: fix structured prompt template format (#27003)
template_format is an init argument on ChatPromptTemplate but not an
attribute on the object so was getting shoved into
StructuredPrompt.structured_ouptut_kwargs
2024-09-30 11:47:46 -07:00
Christophe Bornet
db8845a62a core: Add ruff rules for pycodestyle Warning (W) (#26964)
All auto-fixes.
2024-09-30 09:31:43 -04:00
Christophe Bornet
7809b31b95 core[patch]: Add ruff rules for flake8-simplify (SIM) (#26848)
Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2024-09-27 20:13:23 +00:00
Eugene Yurtsev
de0b48c41a docs: Upgrade examples with RunnableWithMessageHistory to langgraph memory (#26855)
This PR updates the documentation examples that used
RunnableWithMessageHistory to show how to achieve the same
implementation with langgraph memory.

Some of the underlying PRs (not all of them):

- docs[patch]: update chatbot tutorial and migration guide (#26780)
- docs[patch]: update chatbot memory how-to (#26790)
- docs[patch]: update chatbot tools how-to (#26816)
- docs: update chat history in rag how-to (#26821)
- docs: update trim messages notebook (#26793)
- docs: clean up imports in how to guide for rag qa with chat history
(#26825)
- docs[patch]: update conversational rag tutorial (#26814)

---------

Co-authored-by: ccurme <chester.curme@gmail.com>
Co-authored-by: Vadym Barda <vadym@langchain.dev>
Co-authored-by: mercyspirit <ziying.qiu@gmail.com>
Co-authored-by: aqiu7 <aqiu7@gatech.edu>
Co-authored-by: John <43506685+Coniferish@users.noreply.github.com>
Co-authored-by: Erick Friis <erick@langchain.dev>
Co-authored-by: William FH <13333726+hinthornw@users.noreply.github.com>
Co-authored-by: Subhrajyoty Roy <subhrajyotyroy@gmail.com>
Co-authored-by: Rajendra Kadam <raj.725@outlook.com>
Co-authored-by: Christophe Bornet <cbornet@hotmail.com>
Co-authored-by: Devin Gaffney <itsme@devingaffney.com>
Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
2024-09-27 20:04:30 +00:00
Christophe Bornet
f4e738bb40 core: Add ruff rules for PIE (#26939)
All auto-fixes.
2024-09-27 12:08:35 -04:00
Emmanuel Sciara
c6350d636e core[fix]: using async rate limiter methods in async code (#26914)
**Description:** Replaced blocking (sync) rate_limiter code in async
methods.

**Issue:** #26913

**Dependencies:** N/A

**Twitter handle:** no need 🤗
2024-09-26 20:44:28 +00:00
Julius Stopforth
121e79b1f0 core: Fix IndexError when trim_messages invoked with empty list (#26896)
This prevents `trim_messages` from raising an `IndexError` when invoked
with `include_system=True`, `strategy="last"`, and an empty message
list.

Fixes #26895

Dependencies: none
2024-09-26 11:29:58 -04:00
Erick Friis
ef2ab26113 core: release 0.3.6 (#26863) 2024-09-25 11:05:53 -07:00
Christophe Bornet
3a1b9259a7 core: Add ruff rules for comprehensions (C4) (#26829) 2024-09-25 09:34:17 -04:00
William FH
82b5b77940 [Core] Add more interops tests (#26841)
To test that the client propagates both ways
2024-09-24 20:18:20 -07:00
William FH
9b6ac41442 [Core] Inherit tracing metadata & tags (#26838) 2024-09-24 19:33:12 -07:00
Nuno Campos
de70a64e3a core: Run LangChainTracer inline (#26797)
- this flag ensures the tracer always runs in the same thread as the run
being traced for both sync and async runs
- pro: less chance for ordering bugs and other oddities
- blocking the event loop is not a concern given all code in the tracer
holds the GIL anyway
2024-09-24 08:31:18 -07:00
Mohammad Mohtashim
154a5ff7ca core[patch]: On Chain Start Fix for Chain Class (#26593)
- **Issue:** #26588

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2024-09-23 19:30:59 +00:00
ccurme
bba7af903b core[patch]: set default on Blob (#26787)
Resolves https://github.com/langchain-ai/langchain/issues/26781
2024-09-23 18:55:56 +00:00
Erick Friis
238a31bbd9 core: release 0.3.5 (#26737) 2024-09-21 00:26:39 +00:00
William FH
55af6fbd02 [LangChainTracer] Omit Chunk (#26602)
in events / new llm token
2024-09-20 17:10:34 -07:00
William FH
864020e592 [Tracer] add project name to run from tracer (#26736) 2024-09-20 16:48:37 -07:00
Alejandro Rodríguez
4ac9a6f52c core: fix "template" not allowed as prompt param (#26060)
- **Description:**  fix "template" not allowed as prompt param
- **Issue:** #26058
- **Dependencies:** none


- [ ] **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.

---------

Co-authored-by: Erick Friis <erick@langchain.dev>
2024-09-20 23:33:06 +00:00
Erick Friis
4a2745064a core: release 0.3.4 (#26729) 2024-09-20 14:47:15 -07:00
Nuno Campos
345edeb1f0 core: In astream_events propagate cancellation reason to inner task (#26727)
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.
2024-09-20 14:42:10 -07:00
Erick Friis
465e43cd43 core: release 0.3.3 (#26713) 2024-09-20 13:54:19 -07:00
Eugene Yurtsev
4fc69d61ad core[patch]: Fix defusedxml import (#26718)
Fix defusedxml import. Haven't investigated what's actually going on
under the hood -- defusedxml probably does some weird things in __init__
2024-09-20 16:53:24 -04:00
Eugene Yurtsev
79b224f6f3 core/langchain: fix version used in deprecation (#26724)
in core deprecation should be version 0.3.3 instead of 0.3.4
in langchain deprecation should be version 0.3.1 instead of 0.3.4
2024-09-20 16:47:18 -04:00
Eugene Yurtsev
91f4711e53 core[patch],langchain[patch]: deprecate memory and entity abstractions and implementations (#26717)
This PR deprecates the old memory, entity abstractions and implementations
2024-09-20 15:06:25 -04:00
William FH
19ce95d3c9 Avoid copying runs (#26689)
Also, re-unify run trees. Use a single shared client.
2024-09-20 10:57:41 -07:00
Bagatur
f7bb3640f1 core[patch]: support js chat model namespaces (#26688) 2024-09-19 16:14:20 -07:00
Bagatur
c453b76579 core[patch]: Release 0.3.2 (#26686) 2024-09-19 14:58:45 -07:00
Piyush Jain
f087ab43fd core[patch]: Fix load of ChatBedrock (#26679)
Complementary PR to master for #26643.

Co-authored-by: Bagatur <22008038+baskaryan@users.noreply.github.com>
2024-09-19 21:57:20 +00:00
Bagatur
409f35363b core[patch]: support load from path for default namespaces (#26675) 2024-09-19 14:47:27 -07:00
Erick Friis
311f861547 core, community: move graph vectorstores to community (#26678)
remove beta namespace from core, add to community
2024-09-19 11:38:14 -07:00
Christophe Bornet
fd21ffe293 core: Add N(naming) ruff rules (#25362)
Public classes/functions are not renamed and rule is ignored for them.

Co-authored-by: Erick Friis <erick@langchain.dev>
2024-09-19 05:09:39 +00:00
Erick Friis
9909354cd0 core: use ruff.target-version instead (#26634)
tested on one of the replacement cases and seems to work! 
![ScreenShot 2024-09-18 at 02 02
43PM](https://github.com/user-attachments/assets/7170975a-2542-43ed-a203-d4126c6a2c81)
2024-09-18 21:06:14 +00:00
Erick Friis
84b831356c core: remove [project] tag from pyproject (#26633)
makes core incompatible with uv installs
2024-09-18 20:39:49 +00:00
Christophe Bornet
a47b332841 core: Put Python version as a project requirement so it is considered by ruff (#26608)
Ruff doesn't know about the python version in
`[tool.poetry.dependencies]`. It can get it from
`project.requires-python`.

Notes:
* poetry seems to have issues getting the python constraints from
`requires-python` and using `python` in per dependency constraints. So I
had to duplicate the info. I will open an issue on poetry.
* `inspect.isclass()` doesn't work correctly with `GenericAlias`
(`list[...]`, `dict[..., ...]`) on Python <3.11 so I added some `not
isinstance(type, GenericAlias)` checks:

Python 3.11
```pycon
>>> import inspect
>>> inspect.isclass(list)
True
>>> inspect.isclass(list[str])
False
```

Python 3.9
```pycon
>>> import inspect
>>> inspect.isclass(list)
True
>>> inspect.isclass(list[str])
True
```

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2024-09-18 14:37:57 +00:00
Christophe Bornet
3a99467ccb core[patch]: Add ruff rule UP006(use PEP585 annotations) (#26574)
* Added rules `UPD006` now that Pydantic is v2+

---------

Co-authored-by: Eugene Yurtsev <eyurtsev@gmail.com>
2024-09-17 21:22:50 +00:00
Erick Friis
194adc485c docs: pypi readme image links (#26590) 2024-09-17 20:41:34 +00:00
Bagatur
e1d113ea84 core,openai,grow,fw[patch]: deprecate bind_functions, update chat mod… (#26584)
…el api ref
2024-09-17 11:32:39 -07:00
Bagatur
145a49cca2 core[patch]: Release 0.3.1 (#26581) 2024-09-17 17:34:09 +00:00