langchain

0 Fork 0

mirror of https://github.com/langchain-ai/langchain.git synced 2026-10-05 09:25:14 +03:00

T

dependabot[bot] 55f7ce204c chore(deps): bump anyio from 4.11.0 to 4.14.2 in /libs/partners/xai (#40628 )

Bumps [anyio](https://github.com/agronholm/anyio) from 4.11.0 to 4.14.2.
<details>
<summary>Release notes</summary>
<p><em>Sourced from <a
href="https://github.com/agronholm/anyio/releases">anyio's
releases</a>.</em></p>
<blockquote>
<h2>4.14.2</h2>
<ul>
<li>Changed <code>ByteReceiveStream.receive()</code> implementations to
raise a <code>ValueError</code> when <code>max_bytes</code> is not a
positive integer (<a
href="https://redirect.github.com/agronholm/anyio/pull/1191">#1191</a>)</li>
<li>Fixed <code>CapacityLimiter.total_tokens</code> rejecting
<code>float(&quot;inf&quot;)</code> when the limiter was instantiated
outside of an event loop. The adapter setter checked for infinity by
identity (<code>value is math.inf</code>), so only the exact
<code>math.inf</code> singleton was accepted, while every backend setter
(using <code>math.isinf()</code>) accepts any positive infinity (<a
href="https://redirect.github.com/agronholm/anyio/pull/1189">#1189</a>;
PR by <a
href="https://github.com/greymoth-jp"><code>@greymoth-jp</code></a>).</li>
<li>Fixed <code>to_process.run_sync()</code> deadlocking when the worker
function writes enough data to <code>sys.stderr</code> to fill the
(undrained) pipe buffer. The worker process now redirects
<code>sys.stderr</code> to <code>os.devnull</code> as well, matching the
documented behavior</li>
<li>Fixed <code>TLSStream.wrap()</code> matching an internationalized
(unicode) host name against the peer certificate using IDNA 2003 (via
the standard library) instead of IDNA 2008, which could cause the host
name to be matched against the wrong certificate (<a
href="https://redirect.github.com/agronholm/anyio/pull/1208">#1208</a>)</li>
<li>Fixed <code>anyio.open_process()</code> (and
<code>run_process()</code>) ignoring the <code>extra_groups</code>
argument, as it mistakenly passed the value of the <code>group</code>
argument instead (<a
href="https://redirect.github.com/agronholm/anyio/pull/1209">#1209</a>)</li>
<li>Fixed <code>CapacityLimiter.acquire_nowait()</code> and
<code>CapacityLimiter.acquire_nowait_on_behalf_of()</code> raising
<code>trio.WouldBlock</code> instead of <code>anyio.WouldBlock</code> on
the <code>trio</code> backend when there are no tokens available (<a
href="https://redirect.github.com/agronholm/anyio/pull/1218">#1218</a>)</li>
<li>Fixed <code>CapacityLimiter</code> on the asyncio backend
over-granting tokens (<code>borrowed_tokens</code> exceeding
<code>total_tokens</code> and <code>available_tokens</code> going
negative) when a non-blocking acquire was made in the window between a
token being released and the notified waiter resuming. The freed token
is now reserved for the woken waiter right away, so the non-blocking
acquire correctly raises <code>WouldBlock</code> (<a
href="https://redirect.github.com/agronholm/anyio/issues/1170">#1170</a>;
PR by <a
href="https://github.com/gaoflow"><code>@gaoflow</code></a>)</li>
<li>Fixed unnecessary CPU spin when delivering cancellation from
<code>CancelScope</code> on asyncio under certain conditions, including
improper cancel scope nesting (<a
href="https://redirect.github.com/agronholm/anyio/issues/1111">#1111</a>)</li>
</ul>
<h2>4.14.1</h2>
<ul>
<li>Fixed teardown of higher-scoped async fixtures failing on asyncio
with <code>RuntimeError: Attempted to exit cancel scope in a different
task than it was entered in</code> when an async test raise an outcome
exception (e.g., <code>pytest.skip()</code>,
<code>pytest.xfail()</code>, or <code>pytest.fail()</code>) (<a
href="https://redirect.github.com/agronholm/anyio/issues/1179">#1179</a>;
PR by <a
href="https://github.com/EmmanuelNiyonshuti"><code>@EmmanuelNiyonshuti</code></a>)</li>
<li>Fixed <code>CapacityLimiter.total_tokens</code> rejecting a value of
<code>0</code> when the limiter was instantiated outside of an event
loop, contradicting the documented behavior of allowing 0 total tokens
(<a
href="https://redirect.github.com/agronholm/anyio/pull/1183">#1183</a>;
PR by <a
href="https://github.com/nyxst4ck"><code>@nyxst4ck</code></a>)</li>
</ul>
<h2>4.14.0</h2>
<ul>
<li>
<p>Added support for Python 3.15</p>
</li>
<li>
<p>Added an asynchronous implementation of the <code>itertools</code>
module (<a
href="https://redirect.github.com/agronholm/anyio/issues/998">#998</a>;
PR by <a href="https://github.com/11kkw"><code>@11kkw</code></a>)</p>
</li>
<li>
<p>Added the <code>local_port</code> parameter to
<code>connect_tcp()</code> to allow binding to a specific local port
before connecting (<a
href="https://redirect.github.com/agronholm/anyio/issues/1067">#1067</a>;
PR by <a
href="https://github.com/nullwiz"><code>@nullwiz</code></a>)</p>
</li>
<li>
<p>Added support for custom capacity limiters in async path and file I/O
functions and classes</p>
</li>
<li>
<p>Added the <code>create_task()</code> task group method for easier
asyncio migration (returns a <code>TaskHandle</code>) (<a
href="https://redirect.github.com/agronholm/anyio/pull/1098">#1098</a>)</p>
</li>
<li>
<p>Changed <code>TaskGroup.start_soon()</code> to return a
<code>TaskHandle</code></p>
</li>
<li>
<p>Added an option for <code>TaskGroup.start()</code> to return a
<code>TaskHandle</code> (which then contains the start value in the
<code>start_value</code> property)</p>
</li>
<li>
<p>Added the <code>cancel()</code> convenience method to
<code>TaskGroup</code> as a shortcut for cancelling the task group's
cancel scope</p>
</li>
<li>
<p>Improved the error message when a known backend is not installed to
suggest the install command (<a
href="https://redirect.github.com/agronholm/anyio/pull/1115">#1115</a>;
PR by <a
href="https://github.com/EmmanuelNiyonshuti"><code>@EmmanuelNiyonshuti</code></a>)</p>
</li>
<li>
<p>Improved <code>anyio.Path</code> to preserve subclass types by
returning <code>Self</code> in methods that return path objects (<a
href="https://redirect.github.com/agronholm/anyio/issues/1130">#1130</a>;
PR by <a
href="https://github.com/EmmanuelNiyonshuti"><code>@EmmanuelNiyonshuti</code></a>)</p>
</li>
<li>
<p>Changed the parameter type annotation in
<code>anyio.Path.write_bytes()</code> to accept any
<code>ReadableBuffer</code>, thus allowing it to accept
<code>bytearray</code> and <code>memoryview</code> to match
<code>pathlib.Path.write_bytes()</code> (<a
href="https://redirect.github.com/agronholm/anyio/issues/1135">#1135</a>;
PR by <a href="https://github.com/SAY-5"><code>@SAY-5</code></a>)</p>
</li>
<li>
<p>Changed several type annotations to only accept callables returning
coroutine-like objects instead of arbitrary awaitables:</p>
<ul>
<li><code>TaskGroup.start_soon()</code></li>
<li><code>TaskGroup.start()</code></li>
<li><code>anyio.from_thread.run()</code></li>
</ul>
<p>This reverts an earlier change from v3.7.0 which was made in error.
(<a
href="https://redirect.github.com/agronholm/anyio/pull/1153">#1153</a>)</p>
</li>
<li>
<p>Changed <code>anyio.run</code> to support callables returning
arbitrary awaitables at runtime on all backends. Previously, this only
worked on asyncio (<a
href="https://redirect.github.com/agronholm/anyio/pull/1171">#1171</a>;
PR by <a
href="https://github.com/gschaffner"><code>@gschaffner</code></a>)</p>
</li>
<li>
<p>Changed several classes (and their subclasses) to have
<code>__slots__</code> (with <code>__weakref__</code>):</p>
<ul>
<li><code>anyio.CancelScope</code></li>
</ul>
</li>
</ul>
<!-- raw HTML omitted -->
</blockquote>
<p>... (truncated)</p>
</details>
<details>
<summary>Commits</summary>
<ul>
<li><a
href="https://github.com/agronholm/anyio/commit/c384f99687c64c59ed8a11c3a0f11a2d57daff71"><code>c384f99</code></a>
Bumped up the version</li>
<li><a
href="https://github.com/agronholm/anyio/commit/dbba29d1ade7936f18fb71ba24aa92978673482a"><code>dbba29d</code></a>
Fixed 100% CPU spin on cancel scope misuse (<a
href="https://redirect.github.com/agronholm/anyio/issues/1217">#1217</a>)</li>
<li><a
href="https://github.com/agronholm/anyio/commit/6bbc6c33caabc13af5bc4256f745027cf8d5d7b8"><code>6bbc6c3</code></a>
Fix CapacityLimiter over-granting tokens on asyncio (<a
href="https://redirect.github.com/agronholm/anyio/issues/1172">#1172</a>)</li>
<li><a
href="https://github.com/agronholm/anyio/commit/6f82b2537cbbe98f3df3f295499056ab7de0b15b"><code>6f82b25</code></a>
Refactored TestTLSStream.test_receive_invalid_max_bytes() to be less
flaky</li>
<li><a
href="https://github.com/agronholm/anyio/commit/be24b0414f67f604bcbdd5ea3bcc56ab920d872e"><code>be24b04</code></a>
Relaxed timeouts to fix test flakiness</li>
<li><a
href="https://github.com/agronholm/anyio/commit/81135065749b4f60c06619b9caaf0a11871c1ddf"><code>8113506</code></a>
Fix test flakiness caused by slow callback duration logging</li>
<li><a
href="https://github.com/agronholm/anyio/commit/1e988b617b69588e33fecb75e36a9837245f562f"><code>1e988b6</code></a>
Fixed CapacityLimiter raising trio.WouldBlock instead of
anyio.WouldBlock (<a
href="https://redirect.github.com/agronholm/anyio/issues/1">#1</a>...</li>
<li><a
href="https://github.com/agronholm/anyio/commit/44713f345cd29dd4e7d76553c134543a1296cc62"><code>44713f3</code></a>
Pin setup-uv to a commit sha across downstream jobs (<a
href="https://redirect.github.com/agronholm/anyio/issues/1213">#1213</a>)</li>
<li><a
href="https://github.com/agronholm/anyio/commit/f1b7301c8264b0d2e8d24a5788fd29e93dea4040"><code>f1b7301</code></a>
Fixed stderr writes in a worker subprocess causing a deadlock (<a
href="https://redirect.github.com/agronholm/anyio/issues/1207">#1207</a>)</li>
<li><a
href="https://github.com/agronholm/anyio/commit/212be93c2cf2c841e753e95e5e2c543ee7feca90"><code>212be93</code></a>
Fix flaky test_tcp_listener_same_port using a hardcoded port (<a
href="https://redirect.github.com/agronholm/anyio/issues/1206">#1206</a>)</li>
<li>Additional commits viewable in <a
href="https://github.com/agronholm/anyio/compare/4.11.0...4.14.2">compare
view</a></li>
</ul>
</details>
<br />

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

2026-09-18 14:02:03 -04:00

.devcontainer

fix(infra): use langchain_v1 for dev container deps (#34534 )

2025-12-29 18:10:40 -06:00

.github

chore(deps): bump aws-actions/configure-aws-credentials from 6.2.3 to 6.2.4 in the minor-and-patch group (#40564 )

2026-09-17 14:08:20 -04:00

.vscode

chore(infra): remove jupyter recommended extensions (#34430 )

2025-12-19 20:24:58 -06:00

libs

chore(deps): bump anyio from 4.11.0 to 4.14.2 in /libs/partners/xai (#40628 )

2026-09-18 14:02:03 -04:00

openwiki

docs: update OpenWiki (#40318 )

2026-09-09 17:02:42 -04:00

.dockerignore

feat(infra): add .dockerignore for codespaces (#34533 )

2025-12-29 17:58:28 -06:00

.editorconfig

chore: add .editorconfig for consistent coding styles across files (#32261 )

2025-07-27 23:25:30 -04:00

.gitattributes

Update dev container (#6189 )

2023-06-16 15:42:14 -07:00

.gitignore

chore(infra): expand credential coverage in root .gitignore (#39147 )

2026-07-30 11:27:08 -04:00

.markdownlint.json

chore: formatting across codebase (#32466 )

2025-08-08 10:20:10 -04:00

.mcp.json

chore: add reference docs MCP (#35737 )

2026-03-10 23:21:19 -04:00

.pre-commit-config.yaml

chore(infra): wire up per-partner version-consistency pre-commit hooks (#38121 )

2026-06-12 22:08:22 -04:00

AGENTS.md

docs: add openwiki (#40183 )

2026-09-04 11:03:34 +02:00

CITATION.cff

rename repo namespace to langchain-ai (#11259 )

2023-10-01 15:30:58 -04:00

CLAUDE.md

docs: update OpenWiki (#40318 )

2026-09-09 17:02:42 -04:00

LICENSE

Library Licenses (#13300 )

2023-11-28 17:34:27 -08:00

README.md

fix: typo (#39684 )

2026-08-17 15:03:26 -04:00

README.md

The agent engineering platform.

LangChain is a framework for building agents and LLM-powered applications. It helps you chain together interoperable components and third-party integrations to simplify AI application development — all while future-proofing decisions as the underlying technology evolves.

Tip

Just getting started? Check out Deep Agents — a higher-level package built on LangChain for agents that have built-in capabilities for common usage patterns such as planning, subagents, file system usage, and more.

Quickstart

uv add langchain

from langchain.chat_models import init_chat_model

model = init_chat_model("openai:gpt-5.5")
result = model.invoke("Hello, world!")

If you're looking for more advanced customization or agent orchestration, check out LangGraph, our framework for building controllable agent workflows.

For an equivalent JS/TS library, check out LangChain.js.

Tip

For developing, debugging, and deploying AI agents and LLM applications, see LangSmith.

LangChain ecosystem

While the LangChain framework can be used standalone, it also integrates seamlessly with any LangChain product, giving developers a full suite of tools when building LLM applications.

Deep Agents — Build agents that can plan, use subagents, and leverage file systems for complex tasks
LangGraph — Build agents that can reliably handle complex tasks with our low-level agent orchestration framework
Integrations — Chat & embedding models, tools & toolkits, and more
LangSmith — Agent evals, observability, and debugging for LLM apps
LangSmith Deployment — Deploy and scale agents with a purpose-built platform for long-running, stateful workflows

Why use LangChain?

LangChain helps developers build applications powered by LLMs through a standard interface for models, embeddings, vector stores, and more.

Real-time data augmentation — Easily connect LLMs to diverse data sources and external/internal systems, drawing from LangChain's vast library of integrations with model providers, tools, vector stores, retrievers, and more
Model interoperability — Swap models in and out as your engineering team experiments to find the best choice for your application's needs. As the industry frontier evolves, adapt quickly — LangChain's abstractions keep you moving without losing momentum
Rapid prototyping — Quickly build and iterate on LLM applications with LangChain's modular, component-based architecture. Test different approaches and workflows without rebuilding from scratch, accelerating your development cycle
Production-ready features — Deploy reliable applications with built-in support for monitoring, evaluation, and debugging through integrations like LangSmith. Scale with confidence using battle-tested patterns and best practices
Vibrant community and ecosystem — Leverage a rich ecosystem of integrations, templates, and community-contributed components. Benefit from continuous improvements and stay up-to-date with the latest AI developments through an active open-source community
Flexible abstraction layers — Work at the level of abstraction that suits your needs — from high-level chains for quick starts to low-level components for fine-grained control. LangChain grows with your application's complexity

Resources

Documentation — conceptual overviews and guides
LangChain ecosystem overview — how LangChain, LangGraph, and Deep Agents fit together
API reference — complete reference for all public classes, functions, and types
Discussions — community forum for technical questions, ideas, and feedback
LangChain Academy — comprehensive, free courses on LangChain libraries and products, made by the LangChain team
Contributing Guide — how to contribute and find good first issues
Code of Conduct — community guidelines and standards