github  

Messages by Thread

• Re: [I] [EPIC] ANSI mode audit follow-ups (vs Spark 4.1.1) [datafusion-comet] via GitHub
• [PR] fix: Native shuffle fails with a 2GB task serialization OOM on jobs with many partitions [datafusion-comet] via GitHub
• [I] Native shuffle fails with a 2GB task serialization OOM on jobs with very many partitions [datafusion-comet] via GitHub
• Re: [I] Use arrow dictionary casts instead of hand-rolled dictionary handling in cast paths [datafusion-comet] via GitHub
• Re: [I] Guard against silent fail_on_error loss in scalar function wiring [datafusion-comet] via GitHub
• [PR] ci: Check datafusion-spark feature combinations to catch publish errors [datafusion] via GitHub
  • Re: [PR] ci: Check datafusion-spark feature combinations to catch publish errors [datafusion] via GitHub
  • Re: [PR] ci: Check datafusion-spark feature combinations to catch publish errors [datafusion] via GitHub
• [I] Error publishing datafusion-spark in datafusion 55.0.0 release "cannot find module or crate `datafusion` in this scope" [datafusion] via GitHub
• [PR] Update version to 55.0.0, add changelog [datafusion] via GitHub
  • Re: [PR] Update version to 55.0.0, add changelog [datafusion] via GitHub
• [PR] Alamb/update release instructions2 [datafusion] via GitHub
  • Re: [PR] docs: Update dev/release/README.md for new crate `datafusion-proto-models` [datafusion] via GitHub
  • Re: [PR] docs: Update dev/release/README.md for new crate `datafusion-proto-models` [datafusion] via GitHub
• Re: [I] Physical planning: cast low-medium cardinality columns to dictionary arrays before aggregation [datafusion] via GitHub
  • Re: [I] Physical planning: cast low-medium cardinality columns to dictionary arrays before aggregation [datafusion] via GitHub
• [PR] docs: Document missing SQL data types [datafusion] via GitHub
  • Re: [PR] docs: Document missing SQL data types [datafusion] via GitHub
• [I] PartitionedTopKExec does not expose any metrics (`metrics()` returns `None`) [datafusion] via GitHub
• [I] UnnestExec `output_batches` metric does not match the number of emitted batches [datafusion] via GitHub
• [I] SortExec (TopK) `output_batches` metric does not match the number of emitted batches [datafusion] via GitHub
• [PR] Fix `cargo audit` by updating `h2` to 0.4.16 [datafusion] via GitHub
  • Re: [PR] Fix `cargo audit` by updating `h2` to 0.4.16 [datafusion] via GitHub
  • Re: [PR] Fix `cargo audit` by updating `h2` to 0.4.16 [datafusion] via GitHub
• Re: [I] TUI's duplicated API response types have drifted from the scheduler, breaking the stages popup [datafusion-ballista] via GitHub
• [PR] Parser: fix IS [NOT] DISTINCT FROM right-operand precedence [datafusion-sqlparser-rs] via GitHub
• [PR] Enable eight more pedantic lints [datafusion] via GitHub
  • Re: [PR] Enable eight more pedantic lints [datafusion] via GitHub
• [PR] Enable clippy::pedantic and clippy::all, opting out rather than in [datafusion] via GitHub
  • Re: [PR] Enable clippy::pedantic and clippy::all, opting out rather than in [datafusion] via GitHub
  • Re: [PR] Enable clippy::pedantic and clippy::all, opting out rather than in [datafusion] via GitHub
• Re: [I] Native Parquet writer derives schema nullability and field IDs from Arrow rather than Catalyst [datafusion-comet] via GitHub
• [PR] fix(physical-expr): stop false infeasibility report from Eq+FALSE in cp_solver [datafusion] via GitHub
• Re: [PR] Add `EmitTo::FirstBlock` for grouped aggregate emission [datafusion] via GitHub
  • Re: [PR] Add `EmitTo::FirstBlock` for grouped aggregate emission [datafusion] via GitHub
• [I] Release DataFusion `56.1.0` (Nov 2026) (minor/patch) Release [datafusion] via GitHub
• [I] Release DataFusion `56.0.0` (Sep / Oct 2026) [datafusion] via GitHub
• [I] Release DataFusion `55.1.0` (Sep 2026) (minor/patch) Release [datafusion] via GitHub
• [I] RelationPlannerContext exposes no CTE scope, so extension relation planners silently shadow CTEs [datafusion] via GitHub
• [I] Audit physical operators for elapsed_compute double-counting of child compute time (follow-up to #24271) [datafusion] via GitHub
• [PR] fix(chaos): harden k8s backend — supervise port-forward, guard fixture dir [datafusion-ballista] via GitHub
• Re: [PR] feat: prototype allocator-level OOM protection (OomGuard breaker + cooperative real-usage accounting) [datafusion-comet] via GitHub
  • Re: [PR] feat: prototype allocator-level OOM protection (OomGuard breaker + cooperative real-usage accounting) [datafusion-comet] via GitHub
• [I] Decide and document the delta-spark version pin for the Delta contrib [datafusion-comet] via GitHub
• Re: [I] [substrait] [sqllogictest] Cannot convert Exists { subquery: <subquery>, negated: true } to Substrait [datafusion] via GitHub
• Re: [I] Support null-aware RightAnti hash join (build on the subquery side) for NOT IN [datafusion] via GitHub
• Re: [I] Remove the buffering of an extra Rows in ReusableRows [datafusion] via GitHub
• Re: [PR] feat(optimizer): logical rewrite Filter(row_number() = 1) over PARTITION BY → Aggregate(FIRST_VALUE(... ORDER BY)) [datafusion] via GitHub
• [PR] bench: pwmj right semi/anti existence joins [datafusion] via GitHub
  • Re: [PR] bench: pwmj right semi/anti existence joins [datafusion] via GitHub
  • Re: [PR] bench: pwmj right semi/anti existence joins [datafusion] via GitHub
  • Re: [PR] bench: pwmj right semi/anti existence joins [datafusion] via GitHub
• [PR] feat(pwmj): support RightSemi/RightAnti existence joins [datafusion] via GitHub
  • Re: [PR] feat(pwmj): support RightSemi/RightAnti existence joins [datafusion] via GitHub
  • Re: [PR] feat(pwmj): support RightSemi/RightAnti existence joins [datafusion] via GitHub
  • Re: [PR] feat(pwmj): support RightSemi/RightAnti existence joins [datafusion] via GitHub
  • Re: [PR] feat(pwmj): support RightSemi/RightAnti existence joins [datafusion] via GitHub
• [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
  • Re: [PR] feat: cost-based join order enumeration in JoinSelection [datafusion] via GitHub
• Re: [PR] Add nested struct schema evolution support for Map types [datafusion] via GitHub
  • Re: [PR] Add nested struct schema evolution support for Map types [datafusion] via GitHub
• [PR] Fix unparse union all nested distinct [datafusion] via GitHub
  • Re: [PR] Fix unparse union all nested distinct [datafusion] via GitHub
• [PR] perf(scheduler): run one stage when the plan asks for the same stage twice [datafusion-ballista] via GitHub
  • Re: [PR] perf(scheduler): run one stage when the plan asks for the same stage twice [datafusion-ballista] via GitHub
  • Re: [PR] perf(scheduler): run one stage when the plan asks for the same stage twice [datafusion-ballista] via GitHub
• Re: [PR] DRAFT - 21566 [datafusion] via GitHub
• [PR] perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: slice instead of take when probe indices are a contiguous run [datafusion] via GitHub
• [PR] perf(scheduler): push filters once, not on every replan [datafusion-ballista] via GitHub
  • Re: [PR] perf(scheduler): push filters once, not on every replan [datafusion-ballista] via GitHub
• Re: [PR] perf: improve ArrowWriter performance for fixed-length vectors [datafusion-comet] via GitHub
  • Re: [PR] perf: improve ArrowWriter performance for fixed-length vectors [datafusion-comet] via GitHub
  • Re: [PR] perf: improve ArrowWriter performance for fixed-length vectors [datafusion-comet] via GitHub
• [PR] perf: avoid processing discarded rows for negative `nth_value` [datafusion] via GitHub
  • Re: [PR] perf: avoid processing discarded rows for negative `nth_value` [datafusion] via GitHub
• Re: [PR] perf: reuse Arrow IPC compression context across shuffle blocks [datafusion-comet] via GitHub
• Re: [PR] refactor: use arrow make_comparator for nested structural equality in arrays_overlap and array_position [2/2] [datafusion-comet] via GitHub
• Re: [PR] fix: preserve Spark semantics for dictionary-encoded Parquet inputs and reject dictionary targets [datafusion-comet] via GitHub
  • Re: [PR] fix: preserve Spark semantics for dictionary-encoded Parquet inputs and reject dictionary targets [datafusion-comet] via GitHub
• [PR] chore(deps): bump astral-sh/setup-uv from 9.0.0 to 10.0.1 [datafusion] via GitHub
  • Re: [PR] chore(deps): bump astral-sh/setup-uv from 9.0.0 to 10.0.1 [datafusion] via GitHub
• [PR] chore(deps): bump taiki-e/install-action from 2.85.11 to 2.86.1 [datafusion] via GitHub
  • Re: [PR] chore(deps): bump taiki-e/install-action from 2.85.11 to 2.86.1 [datafusion] via GitHub
• [PR] chore(deps): bump the codeql-actions group across 1 directory with 2 updates [datafusion] via GitHub
  • Re: [PR] chore(deps): bump the codeql-actions group across 1 directory with 2 updates [datafusion] via GitHub
• [PR] chore(deps): bump tomlkit from 0.14.0 to 0.15.1 [datafusion] via GitHub
  • Re: [PR] chore(deps): bump tomlkit from 0.14.0 to 0.15.1 [datafusion] via GitHub
• [I] FuzzDataGenerator silently drops nulls for Boolean/Byte/Short/Integer columns [datafusion-comet] via GitHub
  • Re: [I] FuzzDataGenerator silently drops nulls for Boolean/Byte/Short/Integer columns [datafusion-comet] via GitHub
• [PR] test(scheduler): pin the expected static and distributed plans [datafusion-ballista] via GitHub
• [PR] perf(scheduler): only give a broadcast build side its own stage when the probe is partitioned [datafusion-ballista] via GitHub
  • Re: [PR] perf(scheduler): only give a broadcast build side its own stage when the probe is partitioned [datafusion-ballista] via GitHub
• [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
  • Re: [PR] perf: raise perfect hash join small build threshold to 256K [datafusion] via GitHub
• [I] Correctness issue with unparsing UNION & UNION ALL [datafusion] via GitHub
• [PR] perf: reuse the projected equivalence group when only child orderings change [datafusion] via GitHub
  • Re: [PR] perf: reuse the projected equivalence group when only child orderings change [datafusion] via GitHub
  • Re: [PR] perf: reuse the projected equivalence group when only child orderings change [datafusion] via GitHub
  • Re: [PR] perf: reuse the projected equivalence group when only child orderings change [datafusion] via GitHub
  • Re: [PR] perf: reuse the projected equivalence group when only child orderings change [datafusion] via GitHub
• [PR] fix: evaluate grouped aggregate arguments after FILTER [datafusion] via GitHub
  • Re: [PR] fix: evaluate grouped aggregate arguments after FILTER [datafusion] via GitHub
• [I] Grouped aggregate FILTER evaluates arguments for rejected rows [datafusion] via GitHub
  • Re: [I] Grouped aggregate FILTER evaluates arguments for rejected rows [datafusion] via GitHub
• [PR] perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
  • Re: [PR] (benchmaxx) perf: choose hash join build side at execution time [datafusion] via GitHub
• [PR] chore(deps): bump uuid from 1.24.0 to 1.24.1 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump uuid from 1.24.0 to 1.24.1 [datafusion-ballista] via GitHub
• [PR] chore(ci): bump taiki-e/install-action from 2.85.11 to 2.86.1 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump taiki-e/install-action from 2.85.11 to 2.86.1 [datafusion-ballista] via GitHub
• Re: [PR] minor(test): strengthen bitwise sort-merge join spill coverage [datafusion] via GitHub
  • Re: [PR] minor(test): strengthen bitwise sort-merge join spill coverage [datafusion] via GitHub
• [PR] perf: don't manufacture an identity projection in ParquetSource [datafusion] via GitHub
  • Re: [PR] perf: don't manufacture an identity projection in ParquetSource [datafusion] via GitHub
  • Re: [PR] perf: don't manufacture an identity projection in ParquetSource [datafusion] via GitHub
• Re: [PR] Doris SQL: create table column options [datafusion-sqlparser-rs] via GitHub
• Re: [PR] Experimental: support parquet partition write [datafusion-comet] via GitHub
  • Re: [PR] Experimental: support parquet partition write [datafusion-comet] via GitHub
• Re: [PR] feat: support max_by and min_by aggregate expressions [datafusion-comet] via GitHub
• [PR] feat: stream aggregates when group keys are partition-disjoint [datafusion] via GitHub
  • Re: [PR] feat: stream aggregates when group keys are partition-disjoint [datafusion] via GitHub
• [PR] docs: add Streamling to known users [datafusion] via GitHub
  • Re: [PR] docs: add Streamling to known users [datafusion] via GitHub
  • Re: [PR] docs: add Streamling to known users [datafusion] via GitHub
  • Re: [PR] docs: add Streamling to known users [datafusion] via GitHub
• Re: [PR] [Tracking] feat(contrib): Native Delta Lake scan via delta-kernel-rs (Iceberg-style contrib) [datafusion-comet] via GitHub
  • Re: [PR] [Tracking] feat(contrib): Native Delta Lake scan via delta-kernel-rs (Iceberg-style contrib) [datafusion-comet] via GitHub
• [PR] perf: make CometShuffleBenchmark completable and fast [datafusion-comet] via GitHub
  • Re: [PR] perf: make CometShuffleBenchmark completable and fast [datafusion-comet] via GitHub
• Re: [I] Materialize Dictionaries in Group Keys [datafusion] via GitHub
  • Re: [I] Materialize Dictionaries in Group Keys [datafusion] via GitHub
• Re: [PR] feat: remove native cast from boolean to decimal [datafusion-comet] via GitHub
• [I] Support streaming aggregates when partitions are unsorted but non‑overlapping [datafusion] via GitHub

• Earlier messages