Fixes 3971: Check how to integrate vector databases via rest APIs

[vga91]

Issue https://github.com/neo4j-contrib/neo4j-apoc-procedures/issues/3971

Changes

Created procedures for chroma and qdrant.

Emulate the https://neo4j.com/docs/cypher-manual/current/indexes/semantic-indexes/vector-indexes/ commands.

• create vector index --> apoc.vectordb.qdrant.createCollection
• drop vector index --> apoc.vectordb.qdrant.deleteCollection
• add vector node/rel--> apoc.vectordb.qdrant.upsert
• db.index.vector.queryNodes / db.index.vector.queryRelationships --> apoc.vectordb.qdrant.get and apoc.vectordb.qdrant.query
• drop vector node/rel --> apoc.vectordb.qdrant.delete

And the same for the chroma procedures.

NOTE: Like the apoc.ml ones, the chroma and qdrand procedures are implemented in such a way that they have the same signature, even though under the hood they have different bodies/methods/etc.

And 2 custom procedures apoc.vectordb.qdrant.get and apoc.vectordb.custom to handle other vector databases (like Pinecone tested in PineconeTest).

Using the apoc.vectordb.*.get and apoc.vectordb.*.query procedures, we can auto-create neo4j vector indexes and entities, using the mapping config.

To evaluate

• apoc.vectordb.custom could be changed to a more generic naming, e.g. apoc.restapi.custom(<conf>), since it could be used with other rest APIs
• move RestAPIConfig to util package

---

Additional notes (after neo4j-contrib PR creation)

• Open a follow-up issue:

• Test / custom procedures with other databases (like Pinecone)

• Problem with Pinecone, create a PR after neo4j-contrib PR creation...
  We cannot execute Pinecone fetch API, due to these 2 pieces of apoc core codes:

  • setDoOutput(true)
  • http.setChunkedStreamingMode(1024 * 1024);
    In both cases, we receive a 200OK, but with no results.

[RobertoSannino]

rivedere questo score che forse non serve

[RobertoSannino]

rivedere questo score