[HIVE-23877] Hive on Spark incorrect partition pruning ANALYZE TABLE

• Type: Bug
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Labels: pull-request-available
• Target Version/s: 3.1.2
• Assignee: Han
• Reporter: Han
• Created: 19/Jul/20 06:46
• Updated: 29/Sep/20 00:49

Description

Partitions are pruned based on the partition specification in ANALYZE TABLE command and cached in TableSpec.
When compiling, It's unnecessary to use PartitionPruner.prune() to get partitions again. And PartitionPruner can not prune partitions for ANALYZE TABLE command, so it will get all partitions.

Issue Links

• links to: GitHub Pull Request #1278