Rich-T-kid opened a new pull request, #11744: URL: https://github.com/apache/arrow-rs/pull/11744
# Which issue does this PR close? - Closes #11720 # Rationale for this change `RunEndEncoded(Int32, Int32)` columns currently use `GenericInProgressArray`. Filtering therefore materializes an intermediate array before the coalescer assembles the output with `concat`. For very sparse filters, resolving the selected logical rows directly into a run builder avoids most of that work. # What changes are included in this PR? This adds `InProgressInt32RunArray` for `RunEndEncoded(Int32, Int32)`. Very sparse filters represented as indices append selected values directly to a run builder, while denser filters continue to use the existing materialized filter path. # Are these changes tested? The new tests cover sparse filtering across multiple runs, null values, reconstructed run ends, and ordering when sparse and materialized inputs are mixed. # Are there any significant 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]
