github  

Messages by Thread

• • Re: [I] Support streaming aggregates when partitions are unsorted but non‑overlapping [datafusion] via GitHub
  • Re: [I] Support streaming aggregates when partitions are unsorted but non‑overlapping [datafusion] via GitHub
  • Re: [I] Support streaming aggregates when partitions are unsorted but non‑overlapping [datafusion] via GitHub
  • Re: [I] Support streaming aggregates when partitions are unsorted but non‑overlapping [datafusion] via GitHub
  • Re: [I] Support streaming aggregates when partitions are unsorted but non‑overlapping [datafusion] via GitHub
• [PR] perf: grow the native shuffle writer's reservation in steps [datafusion-comet] via GitHub
• Re: [PR] Add atomic SessionContext.with_extensions API [datafusion-python] via GitHub
• [I] Extend native Arrow UDF path to scalar Python UDFs (ArrowEvalPythonExec) [datafusion-comet] via GitHub
  • Re: [I] Extend native Arrow UDF path to scalar Python UDFs (ArrowEvalPythonExec) [datafusion-comet] via GitHub
• [PR] perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] (benchmax) perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] (benchmax) perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] (benchmax) perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
  • Re: [PR] (benchmax) perf: bound multi-column join cardinality by the composite key [datafusion] via GitHub
• [PR] perf(core): share a broadcast read between the tasks on an executor [datafusion-ballista] via GitHub
  • Re: [PR] perf(core): share a broadcast read between the tasks on an executor [datafusion-ballista] via GitHub
• [PR] fix: use to_bits() comparison for float types in eq_array (#24431) [datafusion] via GitHub
  • Re: [PR] fix: use to_bits() comparison for float types in eq_array (#24431) [datafusion] via GitHub
  • Re: [PR] fix: use to_bits() comparison for float types in eq_array (#24431) [datafusion] via GitHub
• [PR] fix: use total_cmp for grouped float MIN/MAX accumulators (#24432) [datafusion] via GitHub
  • Re: [PR] fix: use total_cmp for grouped float MIN/MAX accumulators (#24432) [datafusion] via GitHub
  • Re: [PR] fix: use total_cmp for grouped float MIN/MAX accumulators (#24432) [datafusion] via GitHub
• Re: [I] [DISCUSSION] PR Review Process / States / "How to find the PRs in the list that need review" [datafusion] via GitHub
• Re: [I] Consider PR labels to make reviewing easier [datafusion] via GitHub
• Re: [I] Add AI tooling disclosure text to contributor guide and fields to PR templates [datafusion] via GitHub
• [PR] perf(scheduler): read an unfiltered small scan inline instead of staging it [datafusion-ballista] via GitHub
  • Re: [PR] perf(scheduler): read an unfiltered small scan inline instead of staging it [datafusion-ballista] via GitHub
• [PR] docs(dataframe): improve dataframe! and from_columns documentation (#… [datafusion] via GitHub
  • Re: [PR] docs(dataframe): improve dataframe! and from_columns documentation (#… [datafusion] via GitHub
• [PR] fix: expand float fuzz coverage and correct discovered inconsistencies [datafusion] via GitHub
  • Re: [PR] fix: expand float fuzz coverage and correct discovered inconsistencies [datafusion] via GitHub
  • Re: [PR] fix: expand float fuzz coverage and correct discovered inconsistencies [datafusion] via GitHub
• Re: [I] Allow individual cast conversions to be enabled/disabled [datafusion-comet] via GitHub
• [I] Grouped floating-point `MIN` and `MAX` return order-dependent results for NaNs and signed zeros [datafusion] via GitHub
• [I] `ScalarValue::eq_array` is inconsistent with `ScalarValue` equality for floating-point values [datafusion] via GitHub
• [I] docs(dataframe): improve dataframe! and from_columns documentation [datafusion] via GitHub
  • Re: [I] docs(dataframe): improve dataframe! and from_columns documentation [datafusion] via GitHub
• [PR] feat(core): sketch the sort key alongside T-Digest in RuntimeStatsExec [datafusion-ballista] via GitHub
  • Re: [PR] feat(core): Replace TDigest with KLL [datafusion-ballista] via GitHub
• [I] Bug triage results: 2026-08-17 [datafusion-comet] via GitHub
• [PR] chore: instrument JNI traffic in the unified memory pools [datafusion-comet] via GitHub
• [PR] fix: preserve MERGE target qualifier bindings [datafusion] via GitHub
  • Re: [PR] fix: preserve MERGE target qualifier bindings [datafusion] via GitHub
  • Re: [PR] fix: preserve MERGE target qualifier bindings [datafusion] via GitHub
  • Re: [PR] fix: preserve MERGE target qualifier bindings [datafusion] via GitHub
• [PR] Visit CreateView name as a relation [datafusion-sqlparser-rs] via GitHub
• [PR] docs: refresh benchmarking.md with SF1000 results after #2315 (AQE default-on) [datafusion-ballista] via GitHub
  • Re: [PR] docs: refresh benchmarking.md with SF1000 results after #2315 (AQE default-on) [datafusion-ballista] via GitHub
  • Re: [PR] docs: refresh benchmarking.md with SF1000 results after #2315 (AQE default-on) [datafusion-ballista] via GitHub
• Re: [PR] Snowflake: Add EXTERNAL VOLUME DDL [datafusion-sqlparser-rs] via GitHub
• [PR] perf: skip building FilterRemapper when there are no parent filters [datafusion] via GitHub
  • Re: [PR] perf: skip building FilterRemapper when there are no parent filters [datafusion] via GitHub
• Re: [PR] feat: make StatisticsRegistry the default operator-statistics path [datafusion] via GitHub
• [I] Reduce JNI round-trips in the unified memory pools (batching, hysteresis, cheaper call path) [datafusion-comet] via GitHub
  • Re: [I] Reduce JNI round-trips in the unified memory pools (batching, hysteresis, cheaper call path) [datafusion-comet] via GitHub
• [PR] Add reservation-backed memory accounting to LimitedBatchCoalescer [datafusion] via GitHub
  • Re: [PR] Add reservation-backed memory accounting to LimitedBatchCoalescer [datafusion] via GitHub
  • Re: [PR] Add reservation-backed memory accounting to LimitedBatchCoalescer [datafusion] via GitHub
  • Re: [PR] Add reservation-backed memory accounting to LimitedBatchCoalescer [datafusion] via GitHub
  • Re: [PR] Add reservation-backed memory accounting to LimitedBatchCoalescer [datafusion] via GitHub
• [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
  • Re: [PR] WIP: DO NOT REVIEW: Benchmark enabling parquet filter pushdown by default with one-shot row group I/O [datafusion] via GitHub
• [PR] perf(scheduler): merge a task's sorted partitions before the shuffle write [datafusion-ballista] via GitHub
• [I] Unify integer decimal sign and width assembly with numeric formatting [datafusion] via GitHub
  • Re: [I] Unify integer decimal sign and width assembly with numeric formatting [datafusion] via GitHub
  • Re: [I] Unify integer decimal sign and width assembly with numeric formatting [datafusion] via GitHub
• [PR] fix: reject zero parallel parquet row group writers [datafusion] via GitHub
  • Re: [PR] fix: reject zero parallel parquet row group writers [datafusion] via GitHub
• Re: [I] Expr simplifier doesn't simplify exprs that are same if you swap lhs with rhs regardless of cycles [datafusion] via GitHub
  • Re: [I] Expr simplifier doesn't simplify exprs that are same if you swap lhs with rhs regardless of cycles [datafusion] via GitHub
• Re: [I] Bug: ScalarSubqueryToJoin fails with "No field named __always_true" when CTE inside correlated scalar subquery [datafusion] via GitHub
• [PR] perf(core): push a consumer's row limit into RangeShuffleReaderExec [datafusion-ballista] via GitHub
  • Re: [PR] perf(core): push a consumer's row limit into RangeShuffleReaderExec [datafusion-ballista] via GitHub
• Re: [PR] fix: avoid double-counting build_time in HashJoinExec elapsed_compute [datafusion] via GitHub
  • Re: [PR] fix: avoid double-counting build_time in HashJoinExec elapsed_compute [datafusion] via GitHub
  • Re: [PR] fix: avoid double-counting build_time in HashJoinExec elapsed_compute [datafusion] via GitHub
  • Re: [PR] fix: avoid double-counting build_time in HashJoinExec elapsed_compute [datafusion] via GitHub
• [PR] Add per-aggregate accumulator phase metrics [datafusion] via GitHub
  • Re: [PR] Add per-aggregate accumulator phase metrics [datafusion] via GitHub
  • Re: [PR] Add per-aggregate accumulator phase metrics [datafusion] via GitHub
• [PR] chore(deps): bump bytesize from 2.4.2 to 2.7.0 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump bytesize from 2.4.2 to 2.7.0 [datafusion-ballista] via GitHub
• [PR] chore(deps): bump rustls from 0.23.42 to 0.23.43 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump rustls from 0.23.42 to 0.23.43 [datafusion-ballista] via GitHub
• [PR] chore(deps): bump clap from 4.6.4 to 4.6.6 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump clap from 4.6.4 to 4.6.6 [datafusion-ballista] via GitHub
• [PR] chore(ci): bump astral-sh/setup-uv from 9.0.0 to 10.0.1 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump astral-sh/setup-uv from 9.0.0 to 10.0.1 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump astral-sh/setup-uv from 9.0.0 to 10.0.1 [datafusion-ballista] via GitHub
• [PR] chore(deps): bump lru from 0.18.1 to 0.18.2 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump lru from 0.18.1 to 0.18.2 [datafusion-ballista] via GitHub
• [PR] chore(ci): bump taiki-e/install-action from 2.85.11 to 2.85.13 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump taiki-e/install-action from 2.85.11 to 2.85.13 [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump taiki-e/install-action from 2.85.11 to 2.85.13 [datafusion-ballista] via GitHub
• [PR] chore(deps): bump http from 1.4.2 to 1.5.0 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump http from 1.4.2 to 1.5.0 [datafusion-ballista] via GitHub
• [PR] chore(deps): bump futures from 0.3.33 to 0.3.34 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump futures from 0.3.33 to 0.3.34 [datafusion-ballista] via GitHub
• [PR] chore(ci): bump the github-codeql group with 2 updates [datafusion-ballista] via GitHub
  • Re: [PR] chore(ci): bump the github-codeql group across 1 directory with 2 updates [datafusion-ballista] via GitHub
• [PR] chore(deps): bump aws-config from 1.10.0 to 1.10.1 [datafusion-ballista] via GitHub
  • Re: [PR] chore(deps): bump aws-config from 1.10.0 to 1.10.1 [datafusion-ballista] via GitHub
• Re: [I] Include `HashJoinExec` in the metrics user-guide [datafusion] via GitHub
• Re: [I] PostgreSQL compatibility: `SIMILAR TO` should treat `%` as a wildcard [datafusion] via GitHub
• [PR] fix: Streaming aggregation produces wrong result with grouping sets [datafusion] via GitHub
  • Re: [PR] fix: Streaming aggregation produces wrong result with grouping sets [datafusion] via GitHub
  • Re: [PR] fix: Streaming aggregation produces wrong result with grouping sets [datafusion] via GitHub
  • Re: [PR] fix: Streaming aggregation produces wrong result with grouping sets [datafusion] via GitHub
  • Re: [PR] fix: Streaming aggregation produces wrong result with grouping sets [datafusion] via GitHub
• [I] Streaming aggregation produces wrong result with grouping sets [datafusion] via GitHub
• [PR] Perf/19938 grouped flat array [datafusion] via GitHub
  • Re: [PR] Perf/19938 grouped flat array [datafusion] via GitHub
• [PR] docs: Add docstring examples to the unparser module [datafusion-python] via GitHub
• [I] Missing docstring examples in the `unparser` module [datafusion-python] via GitHub
• [PR] feat: decimal support for percentile_cont and median [datafusion] via GitHub
  • Re: [PR] feat: decimal support for percentile_cont and median [datafusion] via GitHub
  • Re: [PR] feat: decimal support for percentile_cont and median [datafusion] via GitHub
  • Re: [PR] feat: decimal support for percentile_cont and median [datafusion] via GitHub
  • Re: [PR] feat: decimal support for percentile_cont and median [datafusion] via GitHub
• [PR] Postgres: Accept generated columns with no mode keyword or VIRTUAL [datafusion-sqlparser-rs] via GitHub
• Re: [PR] feat(substrait): serialize correlated subqueries with OuterReference field references [datafusion] via GitHub
  • Re: [PR] feat(substrait): serialize correlated subqueries with OuterReference field references [datafusion] via GitHub
  • Re: [PR] feat(substrait): serialize correlated subqueries with OuterReference field references [datafusion] via GitHub
• [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
  • Re: [I] Q10 SF1000 regression after #2315: SortPreservingMergeExec exhausts fair memory pool under AQE default-on [datafusion-ballista] via GitHub
• [I] Report native child-operator spill metrics in Spark task metrics for unified shuffle plans [datafusion-comet] via GitHub
• Re: [PR] build(deps): bump astral-sh/setup-uv from 8.2.0 to 9.0.0 [datafusion-python] via GitHub
  • Re: [PR] build(deps): bump astral-sh/setup-uv from 8.2.0 to 9.0.0 [datafusion-python] via GitHub
• [PR] build(deps): bump astral-sh/setup-uv from 8.2.0 to 10.0.0 [datafusion-python] via GitHub
• [PR] build(deps): bump github/codeql-action/analyze from 4.36.2 to 4.37.7 [datafusion-python] via GitHub
• [PR] build(deps): bump github/codeql-action/init from 4.36.2 to 4.37.7 [datafusion-python] via GitHub
• [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
  • Re: [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
  • Re: [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
  • Re: [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
  • Re: [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
  • Re: [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
  • Re: [PR] perf: cache dictionary arc pointer [datafusion] via GitHub
• [I] Allow a stage to start on partially complete inputs instead of waiting for the whole upstream stage [datafusion-ballista] via GitHub
• [I] Buffer shuffle output in memory and spill to disk only when a configured budget is exceeded [datafusion-ballista] via GitHub
• [I] Shuffle affinity: prototype a task distribution policy that places consumer tasks where their input already lives [datafusion-ballista] via GitHub
• [PR] Apply same sparse routing for range partition dynamic filter as hash partition and restructure `build_filter` [datafusion] via GitHub
  • Re: [PR] Apply same sparse routing for range partition dynamic filter as hash partition and restructure `build_filter` [datafusion] via GitHub
  • Re: [PR] Apply same sparse routing for range partition dynamic filter as hash partition and restructure `build_filter` [datafusion] via GitHub
  • Re: [PR] Apply same sparse routing for range partition dynamic filter as hash partition and restructure `build_filter` [datafusion] via GitHub
  • Re: [PR] Apply same sparse routing for range partition dynamic filter as hash partition and restructure `build_filter` [datafusion] via GitHub
• [PR] introduce new dictionary benchmarks [datafusion] via GitHub
  • Re: [PR] introduce new dictionary benchmarks [datafusion] via GitHub
  • Re: [PR] introduce new dictionary benchmarks [datafusion] via GitHub
  • Re: [PR] introduce new dictionary benchmarks [datafusion] via GitHub
  • Re: [PR] introduce new dictionary benchmarks [datafusion] via GitHub
  • Re: [PR] introduce new dictionary benchmarks [datafusion] via GitHub
  • Re: [PR] introduce new dictionary benchmarks [datafusion] via GitHub
• [PR] test: add explode operator microbenchmark [datafusion-comet] via GitHub
  • Re: [PR] test: add explode operator microbenchmark [datafusion-comet] via GitHub
• [PR] perf(core): keep small sort-shuffle output in memory instead of on disk [datafusion-ballista] via GitHub
  • Re: [PR] perf(core): keep small sort-shuffle output in memory instead of on disk [datafusion-ballista] via GitHub

• Earlier messages
• Later messages