Classic298 ea1b58eb6a fix: apply the shared metadata size cap to the last four vector DB backends (#29502) ... Chunk metadata inserted into the vector DB carries arbitrary client- supplied fields (e.g. custom metadata from file uploads), so it needs the same size cap, type coercion and null-byte sanitization every other backend already applies through process_metadata(). Four backends never called it: Qdrant, Qdrant multitenancy, Milvus multitenancy and Oracle23ai, so an oversized or malformed metadata blob went into those unfiltered. Wire process_metadata() in at each backend's single insert/upsert chokepoint, matching the pattern already used by the other eleven backends.		2026-09-04 11:42:20 -04:00
..
loaders	refac	2026-08-25 15:47:55 -04:00
models	fix: repair two broken logging calls, one of which makes VECTOR_DB=opengauss unusable (#27838)	2026-08-10 22:19:41 -06:00
vector	fix: apply the shared metadata size cap to the last four vector DB backends (#29502)	2026-09-04 11:42:20 -04:00
web	refac	2026-08-31 00:40:15 -04:00
external.py	refac	2026-07-27 01:59:17 -04:00
utils.py	refac	2026-08-24 20:34:59 -04:00