chroma[minor]: Add id attribute to Document returned by search results

[kwei-zhang]

• Description
  We added the document id when documents are created in function similarity_search_with_score such that enable retrieve documents again from the vector store using the same id.

  Sample code:

Before change

def _results_to_docs_and_scores(results: Any) -> List[Tuple[Document, float]]:
    return [
        # TODO: Chroma can do batch querying,
        # we shouldn't hard code to the 1st result
        (Document(page_content=result[0], metadata=result[1] or {}), result[2])
        for result in zip(
            results["documents"][0],
            results["metadatas"][0],
            results["distances"][0],
        )
    ]

After change

def _results_to_docs_and_scores(results: Any) -> List[Tuple[Document, float]]:
    return [
        # TODO: Chroma can do batch querying,
        # we shouldn't hard code to the 1st result
        (Document(id =result[0], page_content=result[1], metadata=result[2] or {}), result[3])
        for result in zip(
            results["ids"][0],
            results["documents"][0],
            results["metadatas"][0],
            results["distances"][0],
        )
    ]

• Issue
  The PR is corresponding to this issue.

Test added:

test_vectorstores.py

• [✓] 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/

[vercel]

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

1 Skipped Deployment

Name	Status	Preview	Comments	Updated (UTC)
langchain	⬜️ Ignored (Inspect)	Visit Preview		Dec 13, 2024 9:52pm

[eyurtsev]

need to bump up compatibility

id on document field was only added more recently

    id: Optional[str] = None
    """An optional identifier for the document.

    Ideally this should be unique across the document collection and formatted
    as a UUID, but this will not be enforced.

    .. versionadded:: 0.2.11
    """

[kwei-zhang]

Hey @eyurtsev, we just bumped up the version for langchain-core to 0.2.11

[chaunguyenm]

@eyurtsev Hey there, sorry for the ping. Seems like one of the workflows failed because of poetry.lock so we reverted it to when it was last changed. Please approve again, thank you!

[kwei-zhang]

This looks like an issue importing Standard Tests, will try to fix it tonight

[kwei-zhang]

Hey @eyurtsev , when I run the commands poetry install --with lint,typing, poetry install --with test in master branch and it fails with the same error. It doesn’t seem to be related to our code changes. Could you please look into it or do you have any suggestion on how to fix it?
Thanks for your help!

[kwei-zhang]

Hi @eyurtsev , I believe in recent fix they downgrade the test python version to 3.12 and fix some dependencies which solves the issue. The current PR is facing some depencies issue that is hard to fix so I rebase the change in another PR. Would you please take a look? Thank you.

[chaunguyenm]

Hi @eyurtsev, just an update on this PR, we have passed all the tests. Please review it when you have some time and let us know if there is anything else to be done. Thank you for your work!

[chaunguyenm]

Hi @efriis, we noticed that one of our checks failed exactly as this push by you. We tried to recreate this error in our local machine with python3.9.20 and poetry1.7.1 but was not able to. If you know what the issue is and how we can resolve it, please let us know. Thank you!

[mspronesti]

@chaunguyenm I think it's solved now. I faced the same failure in one of my PR to the chroma package but when they merged it the workflow passed.

[eyurtsev]

Going to try and merge

[kwei-zhang]

Going to try and merge

Hi, it seems the issue with the id function has been addressed in another PR. Feel free to proceed with merging the testing components. Thank you!