github  

Messages by Thread

• • Re: [PR] [benchmark only] PR #21621 push-down TopK through join, rebased on main [datafusion] via GitHub
  • Re: [PR] [benchmark only] PR #21621 push-down TopK through join, rebased on main [datafusion] via GitHub
  • Re: [PR] [benchmark only] PR #21621 push-down TopK through join, rebased on main [datafusion] via GitHub
  • Re: [PR] [benchmark only] PR #21621 push-down TopK through join, rebased on main [datafusion] via GitHub
  • Re: [PR] [benchmark only] PR #21621 push-down TopK through join, rebased on main [datafusion] via GitHub
  • Re: [PR] [benchmark only] PR #21621 push-down TopK through join, rebased on main [datafusion] via GitHub
• [PR] fix: reject duplicate Parquet field names before decoding [datafusion-comet] via GitHub
  • Re: [PR] fix: reject duplicate Parquet field names before decoding [datafusion-comet] via GitHub
• Re: [PR] feat: add datafusion-json crate with json_get_str scaffolding [datafusion] via GitHub
• Re: [PR] feat: run rlike natively by default for Java-equivalent literal patterns [datafusion-comet] via GitHub
• Re: [PR] fix: preserve aggregate result identity during exchange reuse [datafusion-comet] via GitHub
  • Re: [PR] fix: preserve aggregate result identity during exchange reuse [datafusion-comet] via GitHub
• Re: [PR] feat(physical-plan): add GroupColumn support for FixedSizeList<primitive> in multi-column GROUP BY [datafusion] via GitHub
• Re: [PR] Fuse multiple scalar-aggregate subqueries over the same source into a single scan [datafusion] via GitHub
• Re: [PR] chore: collapse duplicated branches that differ by one value [datafusion] via GitHub
• Re: [PR] chore(deps): bump soupsieve from 2.8.3 to 2.8.4 [datafusion-sandbox] via GitHub
• Re: [PR] fix: align Spark collect aggregate element nullability [datafusion] via GitHub
  • Re: [PR] fix: align Spark collect aggregate element nullability [datafusion] via GitHub
• [PR] feat: push TopK dynamic filters into native Parquet scans [datafusion-comet] via GitHub
• [PR] minor: make MinIO tests more stable [datafusion] via GitHub
  • Re: [PR] minor: make MinIO tests more stable [datafusion] via GitHub
• [PR] ci: stop non-gating labels from skipping Preflight [datafusion-comet] via GitHub
  • Re: [PR] ci: stop non-gating labels from skipping Preflight [datafusion-comet] via GitHub
• [I] Native Parquet scan multiplies rows for a struct with duplicate field names [datafusion-comet] via GitHub
• Re: [I] Improve CI to run enough tests when dependency upgraded [datafusion-comet] via GitHub
  • Re: [I] Improve CI to run enough tests when dependency upgraded [datafusion-comet] via GitHub
  • Re: [I] Improve CI to run enough tests when dependency upgraded [datafusion-comet] via GitHub
• [PR] chore: Improve network retry configuration for maven and artifact upload [datafusion-comet] via GitHub
  • Re: [PR] chore: Improve network retry configuration for maven and artifact upload [datafusion-comet] via GitHub
  • Re: [PR] chore: Improve network retry configuration for maven and artifact upload [datafusion-comet] via GitHub
  • Re: [PR] chore: Improve network retry configuration for maven and artifact upload [datafusion-comet] via GitHub
• Re: [I] Use arrow cast for hand-rolled temporal unit conversion loops [datafusion-comet] via GitHub
• Re: [PR] fix: rebase lambda params after projection schema changes [datafusion] via GitHub
• Re: [D] DISCUSSION: Boston Datafusion Meetup September 2026 [datafusion] via GitHub
  • Re: [D] DISCUSSION: Boston Datafusion Meetup September 2026 [datafusion] via GitHub
  • Re: [D] DISCUSSION: Boston Datafusion Meetup September 2026 [datafusion] via GitHub
  • Re: [D] DISCUSSION: Boston Datafusion Meetup September 2026 [datafusion] via GitHub
  • Re: [D] DISCUSSION: Boston Datafusion Meetup September 2026 [datafusion] via GitHub
  • Re: [D] DISCUSSION: Boston Datafusion Meetup September 2026 [datafusion] via GitHub
• [I] size, arrays_zip, map_from_arrays and array_append return wrong answers for a nondeterministic child [datafusion-comet] via GitHub
• [PR] fix: derive stage input/output rows from the plan's leaves and root instead of summing every operator [datafusion-ballista] via GitHub
• [PR] feat(scheduler): stage a join's build side before deciding the join [datafusion-ballista] via GitHub
  • Re: [PR] feat(scheduler): stage a join's build side before deciding the join [datafusion-ballista] via GitHub
• Re: [PR] feat(functions-aggregate): make approx_distinct HLL precision configurable [datafusion] via GitHub
• Re: [PR] perf: use columnar state for ordered `ARRAY_AGG` and `STRING_AGG` [datafusion] via GitHub
• Re: [PR] minor: validate config `default_null_ordering` when setting it [datafusion] via GitHub
  • Re: [PR] minor: validate config `default_null_ordering` when setting it [datafusion] via GitHub
• [PR] fix: Derive Substrait intersection nullability from every input [datafusion] via GitHub
  • Re: [PR] fix: Derive Substrait intersection nullability from every input [datafusion] via GitHub
• Re: [PR] Push down topk through join [datafusion] via GitHub
  • Re: [PR] Push down topk through join [datafusion] via GitHub
  • Re: [PR] Push down topk through join [datafusion] via GitHub
  • Re: [PR] Push down topk through join [datafusion] via GitHub
  • Re: [PR] Push down topk through join [datafusion] via GitHub
• [PR] fix(substrait): reject incompatible decimal output types [datafusion] via GitHub
  • Re: [PR] fix(substrait): reject incompatible decimal output types [datafusion] via GitHub
• [PR] fix: roll native Iceberg data files on iceberg-java's 1000-row grid [datafusion-comet] via GitHub
  • Re: [PR] fix: roll native Iceberg data files on iceberg-java's 1000-row grid [datafusion-comet] via GitHub
  • Re: [PR] fix: roll native Iceberg data files on iceberg-java's 1000-row grid [datafusion-comet] via GitHub
  • Re: [PR] fix: roll native Iceberg data files on iceberg-java's 1000-row grid [datafusion-comet] via GitHub
• Re: [PR] Add streaming shared subplan example [datafusion] via GitHub
• [PR] fix: match iceberg-java's exception for unclustered input to a clustered Iceberg write [datafusion-comet] via GitHub
  • Re: [PR] fix: match iceberg-java's exception for unclustered input to a clustered Iceberg write [datafusion-comet] via GitHub
  • Re: [PR] fix: match iceberg-java's exception for unclustered input to a clustered Iceberg write [datafusion-comet] via GitHub
• Re: [I] Add AQE test coverage for CometInMemoryTableScanExec [datafusion-comet] via GitHub
• [PR] fix: Set Substrait output_type on aggregate functions [datafusion] via GitHub
• [PR] perf: batch the nested-element list hash instead of slicing per element [datafusion-comet] via GitHub
  • Re: [PR] perf: batch the nested-element list hash instead of slicing per element [datafusion-comet] via GitHub
  • Re: [PR] perf: batch the nested-element list hash instead of slicing per element [datafusion-comet] via GitHub
• [I] Hashing array<struct<..>> is dominated by per-element slicing and dispatch [datafusion-comet] via GitHub
• [PR] Enable more clippy lints (part 3/3) [datafusion] via GitHub
  • Re: [PR] Enable more clippy lints (part 3/3) [datafusion] via GitHub
• [I] Iceberg TestMetadataTablesWithPartitionEvolution.testPartitionColumnNamedPartition returns rows in a different order with native Iceberg writes enabled [datafusion-comet] via GitHub
• [I] Enable TopK dynamic filter pushdown into native Parquet scans [datafusion-comet] via GitHub
  • Re: [I] Enable TopK dynamic filter pushdown into native Parquet scans [datafusion-comet] via GitHub
  • Re: [I] Enable TopK dynamic filter pushdown into native Parquet scans [datafusion-comet] via GitHub
  • Re: [I] Enable TopK dynamic filter pushdown into native Parquet scans [datafusion-comet] via GitHub
• [I] Support dynamic filter pushdown from ungrouped MIN/MAX into native Parquet scans [datafusion-comet] via GitHub
  • Re: [I] Support dynamic filter pushdown from ungrouped MIN/MAX into native Parquet scans [datafusion-comet] via GitHub
  • Re: [I] Support dynamic filter pushdown from ungrouped MIN/MAX into native Parquet scans [datafusion-comet] via GitHub
  • Re: [I] Support dynamic filter pushdown from ungrouped MIN/MAX into native Parquet scans [datafusion-comet] via GitHub
• Re: [PR] fix: normalize signed zero in nested float array comparisons [datafusion-comet] via GitHub
• Re: [I] Discussion: API for Join Access Path and Join Order Selection [datafusion] via GitHub
• Re: [PR] fix: support Utf8/LargeUtf8/Utf8View in native RLike without panicking [datafusion-comet] via GitHub
  • Re: [PR] fix: support Utf8/LargeUtf8/Utf8View in native RLike without panicking [datafusion-comet] via GitHub
  • Re: [PR] fix: support Utf8/LargeUtf8/Utf8View in native RLike without panicking [datafusion-comet] via GitHub
  • Re: [PR] fix: support Utf8/LargeUtf8/Utf8View in native RLike without panicking [datafusion-comet] via GitHub
  • Re: [PR] fix: support Utf8/LargeUtf8/Utf8View in native RLike without panicking [datafusion-comet] via GitHub
  • Re: [PR] fix: support Utf8/LargeUtf8/Utf8View in native RLike without panicking [datafusion-comet] via GitHub
• [PR] fix: null-aware anti join could emit rows after a sibling probe partition saw NULL [datafusion] via GitHub
  • Re: [PR] fix: null-aware anti join could emit rows after a sibling probe partition saw NULL [datafusion] via GitHub
  • Re: [PR] fix: null-aware anti join could emit rows after a sibling probe partition saw NULL [datafusion] via GitHub
  • Re: [PR] fix: null-aware anti join could emit rows after a sibling probe partition saw NULL [datafusion] via GitHub
  • Re: [PR] fix: null-aware anti join could emit rows after a sibling probe partition saw NULL [datafusion] via GitHub
  • Re: [PR] fix: null-aware anti join could emit rows after a sibling probe partition saw NULL [datafusion] via GitHub
• [I] `PhysicalExtensionCodec` should take a `PhysicalPlanDecodeContext` instead of a `TaskContext` [datafusion] via GitHub
  • Re: [I] `PhysicalExtensionCodec` should take a `PhysicalPlanDecodeContext` instead of a `TaskContext` [datafusion] via GitHub
• Re: [PR] perf: Optimize string split functions and avoid intermediate heap allocations [datafusion-comet] via GitHub
• [I] `map[k]` panics when key type is list and key is absent [datafusion] via GitHub
• [PR] perf: demonstrate nested struct statistics pruning [datafusion-comet] via GitHub
• [I] CI takes hours to start [datafusion] via GitHub
  • Re: [I] CI takes hours to start [datafusion] via GitHub
  • Re: [I] CI takes hours to start [datafusion] via GitHub
• [PR] chore: mark partial reduce stream as can handle back pressure (spillable) [datafusion] via GitHub
  • Re: [PR] chore: mark partial reduce stream as can handle back pressure (spillable) [datafusion] via GitHub
  • Re: [PR] chore: mark partial reduce stream as can handle back pressure (spillable) [datafusion] via GitHub
• Re: [PR] feat: RR partition with vectorized distribution [datafusion-comet] via GitHub
  • Re: [PR] feat: RR partition with vectorized distribution [datafusion-comet] via GitHub
• [PR] feat: Fix output bytes metric in hash agg [datafusion] via GitHub
  • Re: [PR] feat: Fix output bytes metric in hash agg [datafusion] via GitHub
• Re: [I] Support Apache Celeborn remote shuffle service [datafusion-comet] via GitHub
• Re: [I] [Variant] Normalize marked Variant arrays at the native Parquet boundary [datafusion-comet] via GitHub
• [I] Compute list/map cardinality from offsets for nested lists [datafusion] via GitHub
• Re: [PR] Prepare SQL runner for native benchmark migration [datafusion] via GitHub
  • Re: [PR] Prepare SQL runner for native benchmark migration [datafusion] via GitHub
• Re: [I] Native Celeborn shuffle: the remote read path costs up to 4.5x local decode and is unbenchmarked [datafusion-comet] via GitHub
• Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
  • Re: [I] CI: Merge queue rejects PR even after all tests pass [datafusion] via GitHub
• [I] Cast from `Timestamp(_, None)` to a named timezone errors on DST boundaries [datafusion] via GitHub
  • Re: [I] Cast from `Timestamp(_, None)` to a named timezone errors on DST boundaries [datafusion] via GitHub
  • Re: [I] Cast from `Timestamp(_, None)` to a named timezone errors on DST boundaries [datafusion] via GitHub
• [I] Consolidate map lookup code between `get_field` and `map_extract` [datafusion] via GitHub
• Re: [PR] fix: scope information schema to catalog [datafusion] via GitHub
• [PR] fix: dispatch Iceberg system functions wrapped as ApplyFunctionExpression [datafusion-comet] via GitHub
  • Re: [PR] fix: dispatch Iceberg system functions wrapped as ApplyFunctionExpression [datafusion-comet] via GitHub
  • Re: [PR] fix: dispatch Iceberg system functions wrapped as ApplyFunctionExpression [datafusion-comet] via GitHub
  • Re: [PR] fix: dispatch Iceberg system functions wrapped as ApplyFunctionExpression [datafusion-comet] via GitHub
  • Re: [PR] fix: dispatch Iceberg system functions wrapped as ApplyFunctionExpression [datafusion-comet] via GitHub
  • Re: [PR] fix: dispatch Iceberg system functions wrapped as ApplyFunctionExpression [datafusion-comet] via GitHub
• [PR] Enable more clippy lints (part 2/3) [datafusion] via GitHub
  • Re: [PR] Enable more clippy lints (part 2/3) [datafusion] via GitHub
• Re: [I] [EPIC] Use blocked / chunked memory management in hash aggregation [datafusion] via GitHub
• Re: [PR] Enable more clippy lints (part 1/3) [datafusion] via GitHub
• [PR] perf: skip calendar reconstruction in hour/minute/second and dayofweek/weekday (up to 23x) [datafusion-comet] via GitHub
  • Re: [PR] perf: skip calendar reconstruction in hour/minute/second and dayofweek/weekday (up to 23x) [datafusion-comet] via GitHub
  • Re: [PR] perf: skip calendar reconstruction in hour/minute/second and dayofweek/weekday (up to 23x) [datafusion-comet] via GitHub
  • Re: [PR] perf: skip calendar reconstruction in hour/minute/second and dayofweek/weekday [datafusion-comet] via GitHub
• Re: [I] `datafusion.execution.time_zone` is not used for basic time zone inference [datafusion] via GitHub
  • Re: [I] `datafusion.execution.time_zone` is not used for basic time zone inference [datafusion] via GitHub
  • Re: [I] `datafusion.execution.time_zone` is not used for basic time zone inference [datafusion] via GitHub
• [PR] perf: compute `cardinality` from list/map offsets [datafusion] via GitHub
  • Re: [PR] perf: compute `cardinality` from list/map offsets [datafusion] via GitHub
• [I] Compute list/map cardinality from offsets when possible [datafusion] via GitHub
  • Re: [I] Compute list/map cardinality from offsets when possible [datafusion] via GitHub
• Re: [I] Support row group pruning for struct field predicates [datafusion] via GitHub
• [I] HashJoinExec `probe_hit_rate` metric overcounts probe rows when a probe batch is processed in several chunks [datafusion] via GitHub
  • Re: [I] HashJoinExec `probe_hit_rate` metric overcounts probe rows when a probe batch is processed in several chunks [datafusion] via GitHub
• [PR] fix: preserve computed projections in unions_to_filter [datafusion] via GitHub
  • Re: [PR] fix: preserve computed projections in unions_to_filter [datafusion] via GitHub
  • Re: [PR] fix: preserve computed projections in unions_to_filter [datafusion] via GitHub
  • Re: [PR] fix: preserve computed projections in unions_to_filter [datafusion] via GitHub
  • Re: [PR] fix: preserve computed projections in unions_to_filter [datafusion] via GitHub
• [PR] perf: reduce invariant-check code generation [datafusion] via GitHub
  • Re: [PR] perf: reduce invariant-check code generation [datafusion] via GitHub
• [PR] fix: partial reduce should not propegate input ordering [datafusion] via GitHub
  • Re: [PR] fix: partial reduce should not propegate input ordering [datafusion] via GitHub
• [PR] fix: preserve memory scan fetch during repartitioning [datafusion] via GitHub
  • Re: [PR] fix: preserve memory scan fetch during repartitioning [datafusion] via GitHub
  • Re: [PR] fix: preserve memory scan fetch during repartitioning [datafusion] via GitHub
• Re: [I] Accelerate DataSource V2 writes (AppendDataExec, OverwriteByExpressionExec, OverwritePartitionsDynamicExec) [datafusion-comet] via GitHub
• [I] unions_to_filter drops computed projections and returns wrong results [datafusion] via GitHub
• [PR] fix: remap projected predicates before WindowTopN [datafusion] via GitHub
  • Re: [PR] fix: remap projected predicates before WindowTopN [datafusion] via GitHub
  • Re: [PR] fix: remap projected predicates before WindowTopN [datafusion] via GitHub
• [PR] chore: update version 55.1.0 [datafusion] via GitHub
  • Re: [PR] chore: update version 55.1.0 [datafusion] via GitHub
  • Re: [PR] chore: update version 55.1.0 [datafusion] via GitHub
  • Re: [PR] chore: update version 55.1.0 [datafusion] via GitHub
• [PR] fix: preserve file scan fetch during filter pushdown [datafusion] via GitHub
  • Re: [PR] fix: preserve file scan fetch during filter pushdown [datafusion] via GitHub
  • Re: [PR] fix: preserve file scan fetch during filter pushdown [datafusion] via GitHub
• [PR] fix: preserve window inputs during TopK aggregation [datafusion] via GitHub
  • Re: [PR] fix: preserve window inputs during TopK aggregation [datafusion] via GitHub
• Re: [I] FFI `Session::create_physical_plan` could recurse infinitely [datafusion] via GitHub
• [PR] fix: preserve fetched merges during distribution enforcement [datafusion] via GitHub
  • Re: [PR] fix: preserve fetched merges during distribution enforcement [datafusion] via GitHub
• [I] perf: skip calendar reconstruction in datetime extraction (hour/minute/second, dayofweek/weekday) [datafusion-comet] via GitHub
  • Re: [I] perf: skip calendar reconstruction in datetime extraction (hour/minute/second, dayofweek/weekday) [datafusion-comet] via GitHub
• [PR] chore(deps): bump runs-on/action from 2.3.0 to 2.3.1 [datafusion] via GitHub
  • Re: [PR] chore(deps): bump runs-on/action from 2.3.0 to 2.3.1 [datafusion] via GitHub
• [PR] chore(deps): bump taiki-e/install-action from 2.87.1 to 2.87.6 [datafusion] via GitHub
  • Re: [PR] chore(deps): bump taiki-e/install-action from 2.87.1 to 2.87.6 [datafusion] via GitHub
• [PR] chore(deps): bump zstd from 0.13.3 to 0.14.0 [datafusion] via GitHub
  • Re: [PR] chore(deps): bump zstd from 0.13.3 to 0.14.0 [datafusion] via GitHub
• [PR] chore(deps): bump the all-other-cargo-deps group with 6 updates [datafusion] via GitHub
  • Re: [PR] chore(deps): bump the all-other-cargo-deps group across 1 directory with 6 updates [datafusion] via GitHub
  • Re: [PR] chore(deps): bump the all-other-cargo-deps group across 1 directory with 6 updates [datafusion] via GitHub
• [PR] chore(deps): bump the arrow-parquet group with 9 updates [datafusion] via GitHub
  • Re: [PR] chore(deps): bump the arrow-parquet group with 9 updates [datafusion] via GitHub
• [PR] chore(deps-dev): bump webpack from 5.110.1 to 5.110.3 in /datafusion/wasmtest/datafusion-wasm-app in the all-npm-deps group [datafusion] via GitHub
  • Re: [PR] chore(deps-dev): bump webpack from 5.110.1 to 5.110.3 in /datafusion/wasmtest/datafusion-wasm-app in the all-npm-deps group [datafusion] via GitHub
• [PR] fix(proto): preserve Parquet source and sink state [datafusion] via GitHub
  • Re: [PR] fix(proto): preserve Parquet source and sink state [datafusion] via GitHub
  • Re: [PR] fix(proto): preserve Parquet source and sink state [datafusion] via GitHub

• Earlier messages
• Later messages