Rich-T-kid opened a new pull request, #27447: URL: https://github.com/apache/datafusion/pull/27447
## Which issue does this PR close?\n\n- Part of #1178.\n\n## Rationale for this change\n\nLiteral binary regex predicates currently cast dictionary-encoded strings to flat arrays before evaluation. This repeats the same regex work for every key occurrence instead of evaluating dictionary values once.\n\nThis is a draft performance experiment. It is gated on the existing hosted TPC-H, TPC-DS, and partitioned ClickBench suites and will be closed if they do not show an attributable speedup.\n\n## What changes are included in this PR?\n\n- Preserve dictionary input through simplification for literal regex predicates.\n- Evaluate scalar regex patterns against dictionary values and reuse the original keys.\n- Compact oversized retained dictionaries before evaluation.\n- Add optimizer, physical-expression, and end-to-end plan tests.\n\nNo benchmark files or custom benchmark queries are added or changed.\n\n## What is the testing strategy?\n\n- cargo test -p datafusion-optimizer test_cast _preimage_dictionary_regex_literal --lib\n- cargo test -p datafusion-physical-expr expressions::binary --lib\n- cargo test -p datafusion --test core_integration dictionary_literal_regex_preserves_encoding\n- cargo clippy -p datafusion-optimizer -p datafusion-physical-expr -p datafusion --lib --tests -- -D warnings\n- cargo fmt --all -- --check\n- Full repository pre-push hook\n- Hosted TPC-H, TPC-DS, and partitioned ClickBench runs on the exact PR head\n\n## Are there any user-facing changes?\n\nLogical results are unchanged. Dictionary-encoded inputs can remain dictionary encoded through the physical regex predicate. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
