Skip to main content

​
New Features

  • Support ORDER BY indexed expressions in the TopN scan path.
  • Add Czech stemming and stopwords support.
  • (beta) Snapshot all indexes via ParallelScanState for consistent parallel JoinScan DocAddresses.
  • (beta) Add a sort-merge-join variant of semi_join_filter.

​
Stability Improvements

  • Fix bypass of min_rows_per_worker for sorted/TopK scans.
  • Fix rebuild of custom scan heap state on rescan.
  • Use upstream tantivy-jieba ordinal position support.
The full changelog is available on the GitHub Release.