mirror of
https://github.com/honeymoose/OpenSearch.git
synced 2025-02-07 05:28:34 +00:00
2c77c0df65
testIndexHasDuplicateData tests were failing ocassionally, due to approximate calculation of BKDReader.estimatePointCount, where if the node is Leaf, the number of points in it was (maxPointsInLeafNode + 1) / 2. As DEFAULT_MAX_POINTS_IN_LEAF_NODE = 1024, for small indexes used in tests, the estimation could be really off. This rewrites tests, to make the max points in leaf node to be a small value to control the tests. Closes #49703