[Feature]: Any plan to support Sparse vectors and bm25?

[yourchanges]

Describe your use case

[Feature]: Any plan to support Sparse vectors and bm25?

Describe the solution you'd like

[Feature]: Any plan to support Sparse vectors and bm25?

Describe alternatives you've considered

No response

Additional context

No response

[GITHUBear]

@yourchanges The Sparse Vector feature depends on the update of OceanBase vector store, and this feature will be updated later this year. The BM25 feature is included in the Full Text Index and is expected to be supported this week.

[yourchanges]

@GITHUBear that's great， if the driver with bm25 ready， we can update the dify oceanbase vector implements to support the fulltext search.

[GITHUBear]

@yourchanges We are pleased to announce that the latest version of pyobvector now supports BM25 algorithm-based full-text indexing. You can find usage examples at: https://github.com/oceanbase/pyobvector/blob/main/tests/test_fts_index.py

[GITHUBear]

@yourchanges By the way, you should deploy a V4.3.5-bp1 oceanbase cluster to enable this feature. A fast method to use is deploy a standalone oceanbase docker container with following command:

docker run --name=ob435 -e MODE=slim -p 2881:2881 -d quay.io/oceanbase/oceanbase-ce:4.3.5.1-101000042025031818

[yourchanges]

@GITHUBear It's cool , but because the version conflict of deps such as pydantic and etc before the v0.2.3, we have already do a implements based on the v0.1.18, here is the commit
langgenius/dify@6157f57

if we have more time later, we can do a upgrade to use the v0.2.3 or above

[whhe]

Sparse vector is supported in #19.