Abhisheklearn12 opened a new pull request, #25870: URL: https://github.com/apache/datafusion/pull/25870
## Which issue does this PR close? - Part of #25035. Split out from #25771 as requested, so the evaluator change there can be benchmarked against `main`. ## Rationale for this change On `main`, a left-deep `AND` chain can be several times slower than the same chain built right-deep, once the prefix drops below the pre-selection threshold (#25035). The existing `binary_op` benchmarks don't cover this. ## What changes are included in this PR? A `conjunction` group in `datafusion/physical-expr/benches/binary_op.rs`: left-deep and right-deep `AND` chains of cheap comparisons, casts, `IN` lists and arithmetic, with and without unused columns, plus nullable and regex-suffix variants. Only this file changes; no engine code. ## What is the testing strategy for this PR? Benchmark only. All `binary_op` benchmarks run on `main` (`cargo test --bench binary_op`). Run with `cargo bench --bench binary_op -- conjunction`. ## Are there any user-facing changes? No. -- 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]
