Messages by Thread
-
[PR] docs: fix a/an articles in spill manager and SHJ comments [datafusion]
aarushkandukoori (via GitHub) via github
-
Re: [PR] feat: support native Iceberg Bloom writes on older runtimes [datafusion-comet]
NikitaMatskevich (via GitHub) via github
-
[PR] perf: inline the first value in grouped integer DISTINCT counts [datafusion]
QinXi-ai (via GitHub) via github
-
[PR] test: exercise folded UnsafeArrayData literals [datafusion-comet]
narutomanchester (via GitHub) via github
-
[I] Be More Explicit About Concurrency in Execution Plans [datafusion]
tschwarzinger (via GitHub) via github
-
[PR] feat: hide two ballista clients behind common interface [datafusion-ballista]
milenkovicm (via GitHub) via github
-
[PR] chore(deps): bump runs-on/action from 2.6.0 to 2.7.0 [datafusion]
dependabot[bot] (via GitHub) via github
-
[PR] chore(deps): bump the codeql-actions group across 1 directory with 2 updates [datafusion]
dependabot[bot] (via GitHub) via github
-
[PR] chore(deps): bump pydata-sphinx-theme from 0.23.0 to 0.24.0 in the all-uv-deps group [datafusion]
dependabot[bot] (via GitHub) via github
-
[PR] chore(deps): bump taiki-e/install-action from 2.88.1 to 2.90.2 in /.github/actions/setup-docs-builder [datafusion]
dependabot[bot] (via GitHub) via github
-
[PR] chore(deps): bump taiki-e/install-action from 2.88.1 to 2.90.2 [datafusion]
dependabot[bot] (via GitHub) via github
-
[PR] chore(deps): bump the all-other-cargo-deps group with 4 updates [datafusion]
dependabot[bot] (via GitHub) via github
-
[PR] fix: return error on offset overflow in ScalarValue::to_array_of_size [datafusion]
ETS2K7 (via GitHub) via github
-
Re: [PR] refactor: move EnforceDistribution per-child flags into ChildPolicy [datafusion]
Aaryan-Dadu (via GitHub) via github
-
[PR] fix(parquet): validate float bounds using NaN counts [datafusion]
rank-p (via GitHub) via github
-
[PR] fix: pass the session's batch_size to the Iceberg table scan [datafusion-iceberg]
nikkhiil-hadoop (via GitHub) via github
-
[PR] fix: Spark xxhash64 ignores NaN sign/payload, including list elements [datafusion]
sohailshk (via GitHub) via github
-
Re: [I] Spark `xxhash64` hashes the raw bits of a NaN instead of the canonical NaN [datafusion]
sohailshk (via GitHub) via github
-
Re: [I] Native columnar-to-row conversion is much slower than the JVM implementation for small batches [datafusion-comet]
andygrove (via GitHub) via github
-
[PR] fix: evaluate a NULL-guarded argument for every row when only shared parts of it can fail [datafusion-comet]
dwsmith1983 (via GitHub) via github
-
[PR] fix: keep a literal out of scan equivalence classes for complex_expr = literal filters [datafusion]
cetra3 (via GitHub) via github
-
[PR] fix: let a NULL-guarded argument skip a shared subexpression that cannot fail [datafusion-comet]
dwsmith1983 (via GitHub) via github
-
Re: [PR] fix: keep Iceberg residual pushdown within Spark semantics [datafusion-comet]
rich7420 (via GitHub) via github
-
[I] Restore native row-group pruning for FLOAT and DOUBLE range filters in Iceberg scans [datafusion-comet]
rich7420 (via GitHub) via github
-
[I] REGRESSION: SanityCheckPlan error for GROUP BY ... HAVING on a CASE key with parquet filter pushdown [datafusion]
cetra3 (via GitHub) via github
-
[PR] fix: return 1 from regr_r2 for a horizontal line, not NULL [datafusion]
MaxFreedomPollard (via GitHub) via github
-
Re: [I] Native approx_top_k functions: audit the frequent-items sketch against Java and implement [datafusion-comet]
coderfender (via GitHub) via github
-
[PR] fix: reuse evaluated hash join build keys [datafusion]
Phoenix500526 (via GitHub) via github
-
[I] LEAD/LAG with offset 0 and IGNORE NULLS returns the default for a NULL row, unlike the form without IGNORE NULLS [datafusion]
Sunninsky (via GitHub) via github
-
[I] Speculate string functions such as upper in NullShortCircuit by adding them to is_cheap [datafusion-comet]
tlfua (via GitHub) via github
-
[PR] chore(deps): bump uuid from 1.29.0 to 1.29.1 in /python [datafusion-ballista]
dependabot[bot] (via GitHub) via github
-
[PR] chore(deps): bump uuid from 1.29.0 to 1.29.1 [datafusion-ballista]
dependabot[bot] (via GitHub) via github
-
Re: [PR] feat: let custom ExecutionPlan nodes restrict themselves to a task's partitions [datafusion-ballista]
NoahKusaba (via GitHub) via github
-
Re: [I] Plain `SELECT` of a Parquet float column returns wrong values: NaN rows are replaced by the statistics constant when min == max [datafusion]
compiler-segfault (via GitHub) via github
-
[PR] fix: do not treat floating point columns with equal min/max statistics as constant [datafusion]
compiler-segfault (via GitHub) via github
-
[I] TPC-H Q02: DataFusion vs Polars comparison [datafusion]
comphead (via GitHub) via github
-
[PR] feat: allow `CaseExpr` to reuse its first required operand as ELSE [datafusion]
Weijun-H (via GitHub) via github
-
[I] Let physical-plan producers reuse a CASE's first-WHEN operand as its ELSE value [datafusion]
Weijun-H (via GitHub) via github
-
[I] TPC-H Q01: DataFusion vs Polars comparison [datafusion]
comphead (via GitHub) via github
-
[PR] fix: send a skipped shared subexpression to Spark where Spark's sharing depends on the plan [datafusion-comet]
dwsmith1983 (via GitHub) via github
-
Re: [I] Broadcast/hash join fallback reasons are lost from the AQE-final plan [datafusion-comet]
dwsmith1983 (via GitHub) via github
-
Re: [PR] chore(deps-dev): bump baseline-browser-mapping from 2.9.19 to 2.11.21 in /datafusion/wasmtest/datafusion-wasm-app [datafusion-sandbox]
github-actions[bot] (via GitHub) via github
-
Re: [PR] fix: preserve fetched merges during requirement enforcement [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] feat: add opt-in strict predicate short-circuit evaluation [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] feat: support floating-point ASOF equality keys [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] feat: Prune ListingTable file groups for range-key filters [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] Datasource: Add file_metadata() to DataSink for per-file write metadata [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] dev: add local CodeQL analysis for GitHub Actions [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] benchmarks: count simulated object store GET requests per query [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] fix: align Spark collect aggregate element nullability [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] refactor: Rewrite Hash Join's regular (inner/left/right/full) join path [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] feat(physical-plan): Make `HashTableLookupExpr` serializable [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] Aggregate byte types internally with i64 offsets [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] Prototype: physical filter API (PhysicalFilter / FilterConjunct) with optional conjuncts [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [PR] feat: add native GroupsAccumulator for any_value [datafusion]
github-actions[bot] (via GitHub) via github
-
Re: [I] Native string-to-timestamp cast rejects zone suffixes Spark accepts (PST, CET, Zulu, Etc/GMT+8, +08:00:00, no space) [datafusion-comet]
andygrove (via GitHub) via github
-
Re: [PR] fix: resolve SHORT_IDS and region zone suffixes in native string-to-timestamp cast [datafusion-comet]
andygrove (via GitHub) via github
-
[PR] Hive: Replace `dialect_of!` checks in `CREATE TABLE` and `INSERT` with dialect methods [datafusion-sqlparser-rs]
andygrove (via GitHub) via github
-
[PR] Add dialect hooks for `CREATE FUNCTION` and `USE` [datafusion-sqlparser-rs]
andygrove (via GitHub) via github
-
Re: [PR] fix: plan a range RepartitionExec as a shuffle stage instead of dropping it [datafusion-ballista]
andygrove (via GitHub) via github
-
[I] Consolidate the Scala "cannot fail" checks: arrays.cannotThrow and QueryPlanSerde.cannotFail [datafusion-comet]
tlfua (via GitHub) via github
-
Re: [I] Performance regression when a stage of WholeStageCodegen-fusible operators runs as separate native Comet operators [datafusion-comet]
comphead (via GitHub) via github
-
[PR] perf: filter sort-merge inner join pairs before materializing them [datafusion]
comphead (via GitHub) via github
-
[PR] PostgreSQL: Parse subscript expressions in slice bounds [datafusion-sqlparser-rs]
bendu (via GitHub) via github
-
[PR] fix: report input and SQL metrics while a native shuffle write runs [datafusion-comet]
comphead (via GitHub) via github
-
[I] lpad/rpad do not use withNullShortCircuit, so ANSI errors are raised on rows Spark skips [datafusion-comet]
tlfua (via GitHub) via github
-
[I] A native scan feeding a native shuffle write reports no metrics until the task ends [datafusion-comet]
comphead (via GitHub) via github
-
[PR] fix(python): use a non-default codec in the shuffle compression setter test [datafusion-ballista]
andygrove (via GitHub) via github
-
Re: [I] `ExecutorMetatada` too verbose [datafusion-ballista]
andygrove (via GitHub) via github
-
[I] Native string to decimal cast fails on a string with more than 38 digits: NULL or `CAST_INVALID_INPUT` where Spark returns a value [datafusion-comet]
snmvaughan (via GitHub) via github
-
[I] `ScalarValue::to_array_of_size` panics when a `Utf8` or `Binary` scalar repeated `size` times passes `i32::MAX` bytes [datafusion]
mbutrovich (via GitHub) via github
-
[I] `split_part` with position 0 errors on a NULL string only when the position is a literal [datafusion]
mbutrovich (via GitHub) via github
-
Re: [PR] feat: native Parquet writes to S3 [datafusion-comet]
comphead (via GitHub) via github
-
Re: [PR] chore: run CI on contributor forks [datafusion-comet]
comphead (via GitHub) via github
-
[PR] fix: leave wide integer truncate to Spark [datafusion-comet]
ErikBPF (via GitHub) via github
-
[PR] Support `TABLESAMPLE RESERVOIR` [datafusion-sqlparser-rs]
abdolence (via GitHub) via github
-
[PR] Reduce allocations when parsing identifier expressions [datafusion-sqlparser-rs]
andygrove (via GitHub) via github
-
[PR] fix: accept second-precision zone offsets in native timestamp casts [datafusion-comet]
ErikBPF (via GitHub) via github
-
[PR] fix: validate column-specific parquet encoding [datafusion]
vinycalves (via GitHub) via github
-
Re: [I] Iceberg `truncate` on integers with widths above 2^30: native kernel and native writer compute different partition values [datafusion-comet]
erikbogado-nstech (via GitHub) via github
-
Re: [I] Move the version checks for Spark patch-release fixes into SparkFixes [datafusion-comet]
snmvaughan (via GitHub) via github
-
[PR] fix: raise NUMERIC_OUT_OF_SUPPORTED_RANGE for a string with more than 38 integral digits cast to decimal [datafusion-comet]
snmvaughan (via GitHub) via github
-
[PR] refactor: move the patch-release version checks into SparkFixes [datafusion-comet]
snmvaughan (via GitHub) via github
-
[I] Native string to decimal cast raises `CAST_INVALID_INPUT` instead of `NUMERIC_VALUE_OUT_OF_RANGE` under ANSI when the scale adjustment is more than 38 [datafusion-comet]
snmvaughan (via GitHub) via github
-
Re: [I] The config values aren't checked [datafusion]
vinycalves (via GitHub) via github
-
[PR] fix: unparse NULL nested-type literals as NULL [datafusion]
vinycalves (via GitHub) via github
-
Re: [I] Unparser turns NULL list literals into `[]` and fails on NULL struct/map literals [datafusion]
vinycalves (via GitHub) via github
-
Re: [I] Native string to timestamp cast returns NULL for zone offsets with a non-zero seconds part [datafusion-comet]
erikbogado-nstech (via GitHub) via github
-
[PR] fix: make Spark encode follow Spark 4 by default [datafusion]
Pastoray (via GitHub) via github
-
[PR] Optimize single-column dictionary grouping [datafusion]
Rich-T-kid (via GitHub) via github
-
[I] Support `TIME(p)` SQL type [datafusion]
sdf-jkl (via GitHub) via github
-
[PR] test: rewrite escaped-regex error checks as plain text in 3 sqllogictest files (#27239) [datafusion]
ETS2K7 (via GitHub) via github
-
Re: [I] Deferred page-index load in the Parquet opener bypasses FileMetadataCache, causing repeated uncached I/O when the skip heuristic doesn't fire [datafusion]
mbutrovich (via GitHub) via github
-
[PR] fix: support run-end encoded keys in piecewise merge existence joins [datafusion]
JonSnow1807 (via GitHub) via github
-
Re: [I] Piecewise merge join: semi/anti joins fail with an internal error on RunEndEncoded keys [datafusion]
JonSnow1807 (via GitHub) via github
-
Re: [I] Support range-key locality analysis for `date_bin` with a literal origin [datafusion]
alamb (via GitHub) via github
-
[PR] [DO NOT MERGE] native Iceberg reads of unshredded Variant columns [datafusion-comet]
mbutrovich (via GitHub) via github
-
Re: [I] Prune files by statistics in `ListingTable` before building file groups [datafusion]
jayzhan211 (via GitHub) via github
-
Re: [I] read_side_padding ignores the pad string for dictionary inputs in lpad and rpad [datafusion-comet]
kazuyukitanimura (via GitHub) via github
-
[I] Consider row-count skew when skipping aggregate repartitioning [datafusion]
benbellick (via GitHub) via github
-
Re: [I] Native Azure credential lookup still differs from Hadoop ABFS in a few configurations [datafusion-comet]
andygrove (via GitHub) via github
-
[PR] fix(spark): match Spark 4 encode charset behavior [datafusion]
Subodh-17 (via GitHub) via github
-
Re: [PR] Intermediate result blocked approach to aggregation memory management [datafusion]
github-actions[bot] (via GitHub) via github
-
[PR] fix: reject unbounded null-aware joins [datafusion]
Phoenix500526 (via GitHub) via github
-
Re: [I] Reserve computed hash join build keys before allocating the hash map [datafusion]
dwsmith1983 (via GitHub) via github
-
[I] Bug triage results: 2026-11-09 [datafusion-comet]
andygrove (via GitHub) via github
-
[PR] fix: Address overly aggressive parquet pruning of map cols [datafusion]
ianthetechie (via GitHub) via github
-
Re: [I] Filtered semi/anti sort-merge join is slow on small key groups and charges each group a whole batch [datafusion]
dwsmith1983 (via GitHub) via github