Repository navigation
test: [cp3.0] Fix nested search aggregation expectations - #55244
yanliang567 wants to merge 1 commit into
Conversation
Cherry-pick the search aggregation test changes from milvus-io#55194 only; exclude the unrelated FileResource changes. Use deterministic vectors to cover four composite candidates spanning one or two parent categories while preserving filter, metrics, and top-hits checks. pr: milvus-io#55194 issue: milvus-io#55208 (cherry picked from commit 178265d) Signed-off-by: chyezh <chyezh@outlook.com> Signed-off-by: Yanliang Qiao <yanliang.qiao@zilliz.com>
|
[APPROVALNOTIFIER] This PR is NOT APPROVED This pull-request has been approved by: The full list of commands accepted by this bot can be found here. DetailsNeeds approval from an approver in each of these files:Approvers can indicate their approval by writing |
|
[ci-v2-notice] To rerun ci-v2 checks, comment with:
If you have any questions or requests, please contact @zhikunyao. |
|
[INFO] PR Label Summary by Default
[WARNING] Milestone not set
You can set milestone by commenting: Use /refresh-label to update related check and label manually |
pr: #55194
issue: #55208
Cherry-pick the search aggregation test changes from #55194 to 3.0. The unrelated FileResource changes are excluded.
The nested aggregation selects four composite
(category, brand)candidates, which can all belong to one category. Replace the random-vector expectation of exactly two parent buckets with deterministic cases covering both one and two parents. Preserve filter, metric, and top-hits ordering checks, and assert the expected parent keys and counts.Validation on
harbor.milvus.io/milvusdb/milvus:3.0-20261106-6b52c08e(standalone, Woodpecker; PyMilvus 3.0.3rc10):assert 1 == 2with random-vector seed 0 and in 3/3 runs with controlled candidates from one category.TestSearchAggregation: 27 passed, 1 existing xfail for [Bug]: Python MilvusClient.search ignores group_by_fields with search_aggregation #50960.git diff --checkpassed.