mirror of
https://gitee.com/milvus-io/milvus.git
synced 2026-01-07 19:31:51 +08:00
a1721bb47b
Issue: #46627 add one more test case to cover duplicate pk partial update On branch feature/partial-update Changes to be committed: modified: milvus_client/test_milvus_client_partial_update.py <!-- This is an auto-generated comment: release notes by coderabbit.ai --> - Core invariant: upserts with partial_update=True consolidate records by primary key (PK) rather than creating duplicate rows; this test verifies the partial-update upsert path preserves PK identity and merge semantics. - Change: adds test test_milvus_client_partial_update_duplicate_pk_partial_update which inserts duplicate-PK batches, then calls client.upsert(..., partial_update=True) on a subset of fields and asserts final row count equals default_nb, exercising the partial-update code path (upsert → partial update handling → query) not previously covered. - No production logic removed/simplified: this PR only adds test coverage (no code paths removed or altered); nothing in production code is changed or simplified by the PR. - No data loss or regression introduced: the test validates concrete code paths — upsert with partial_update True followed by query(out_fields/with_vec, pk checks) — and asserts deduplication (2×default_nb → default_nb). Because the PR only adds assertions against existing behavior and does not modify runtime logic, it cannot cause data loss or behavioral regressions. <!-- end of auto-generated comment: release notes by coderabbit.ai --> Signed-off-by: Eric Hou <eric.hou@zilliz.com> Co-authored-by: Eric Hou <eric.hou@zilliz.com>