docs: CrateDB: Register package langchain-cratedb, and add minimal "provider" documentation

[amotl]

Hi Erick. Coming back from a previous attempt, we now made a separate package for the CrateDB adapter, called langchain-cratedb, as advised. Other than registering the package within libs/packages.yml, this patch includes a minimal amount of documentation to accompany the advent of this new package. Let us know about any mistakes we made, or changes you would like to see. Thanks, Andreas.

About

• Description: Register a new database adapter package, langchain-cratedb, providing traditional vector store, document loader, and chat message history features for a start.
• Addressed to: @efriis, @eyurtsev
• References: community: CrateDB: Vector Store #27710
• Preview: Providers » More » CrateDB

Status

• PyPI: https://pypi.org/project/langchain-cratedb/
• GitHub: https://github.com/crate/langchain-cratedb
• Documentation (CrateDB): https://cratedb.com/docs/guide/integrate/langchain/
• Documentation (LangChain): This PR.

Backlog?

Is this applicable for this kind of patch?

• 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.

Q&A

1. Notebooks that use the LangChain CrateDB adapter are currently at CrateDB LangChain Examples, and the documentation refers to them. Because they are derived from very old blueprints coming from LangChain 0.0.x times, we guess they need a refresh before adding them to docs/docs/integrations. Is it applicable to merge this minimal package registration + documentation patch, which already includes valid code snippets in cratedb.mdx, and add corresponding notebooks on behalf of a subsequent patch later?

2. How would it work getting into the tabular list of Integration Packages enumerated on the documentation entrypoint page about Providers?

/cc Please also review, @ckurze, @wierdvanderhaar, @kneth, @simonprickett, if you can find the time. Thanks!

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
langchain	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Dec 22, 2024 5:23am

[ccurme]

Thanks very much for this. To your questions:

1. Notebooks that use the LangChain CrateDB adapter are currently at CrateDB LangChain Examples, and the documentation refers to them. Because they are derived from very old blueprints coming from LangChain 0.0.x times, we guess they need a refresh before adding them to docs/docs/integrations. Is it applicable to merge this minimal package registration + documentation patch, which already includes valid code snippets in cratedb.mdx, and add corresponding notebooks on behalf of a subsequent patch later?

Yes, fine to add notebook for the specific components later. Cursory look suggests minimal examples (e.g., instantiating and using document loaders) could work as-is.

2. How would it work getting into the tabular list of Integration Packages enumerated on the documentation entrypoint page about Providers?

We do not yet have a policy for this and are currently only including packages in the langchain-ai org. This table is built from the packages.yml file so if the policy is updated then langchain-cratedb should be picked up automatically.

[ccurme]

Consider linking to langchain-cratedb repo or pypi somewhere near the top of this page.

[ccurme]

I believe langchain-unstructured can load web pages as well.

[amotl]

Thanks for this suggestion, we are always aiming to follow most recent developments in LangChain, and already recognized langchain-unstructured to be a real powerhouse.

I just tried to bring it in, but a drawback to use it for minimal examples could be the large array of dependencies it pulls in.

$ uv pip install langchain-unstructured
Resolved 45 packages in 598ms
Prepared 5 packages in 3.44s
Uninstalled 2 packages in 19ms
Installed 16 packages in 86ms
 + aiofiles==24.1.0
 + coloredlogs==15.0.1
 + cryptography==44.0.0
 + eval-type-backport==0.2.2
 + flatbuffers==24.12.23
 + humanfriendly==10.0
 + jsonpath-python==1.0.6
 + langchain-unstructured==0.1.6
 + mpmath==1.3.0
 + onnxruntime==1.19.2
 + protobuf==5.29.2
 - pydantic==2.10.4
 + pydantic==2.9.2
 - pydantic-core==2.27.2
 + pydantic-core==2.23.4
 + pypdf==5.1.0
 + sympy==1.13.3
 + unstructured-client==0.28.1

[amotl]

Thanks a stack for the quick merge, @ccurme, and also thanks for your feedback. I will apply your suggestions on the next iteration when sending updates. 🙇