github

Messages by Date

2026/10/01 Re: [PR] feat: add native Delta Lake scan contrib module (page/row-group pruning) [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] feat: Add option to assign Parquet row groups to file ranges by midpoint [datafusion] via GitHub
2026/10/01 [PR] fix: defer Parquet conversion errors until a row group is decoded, as Spark does [datafusion-comet] via GitHub
2026/10/01 Re: [D] DISCUSSION: San Francisco DataFusion Meetup November 2026 [datafusion] via GitHub
2026/10/01 Re: [D] DISCUSSION: San Francisco DataFusion Meetup November 2026 [datafusion] via GitHub
2026/10/01 Re: [PR] perf(pwmj):settle never-matching streamed rows against the buffered extreme… [datafusion] via GitHub
2026/10/01 [I] SQL unparser drops the time zone from timestamp literals [datafusion] via GitHub
2026/10/01 Re: [PR] perf(pwmj):settle never-matching streamed rows against the buffered extreme… [datafusion] via GitHub
2026/10/01 Re: [I] A native final aggregate that has spilled can fail the task during its replay [datafusion-comet] via GitHub
2026/10/01 Re: [I] array_distinct and array_union diverge from Spark on -0.0 for Spark versions without SPARK-54918 [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: Add option to assign Parquet row groups to file ranges by midpoint [datafusion] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 [PR] fix(proto): preserve EmptyRelation schema across logical plan round trip [datafusion] via GitHub
2026/10/01 [I] Queries with multiple `SUM(expr + literal)` aggregates fail or change result type [datafusion] via GitHub
2026/10/01 [I] Spark `negative` returns wrong results for decimal columns [datafusion] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 [I] Comparing a timestamp cast to a time zone against a literal returns wrong results [datafusion] via GitHub
2026/10/01 Re: [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: reuse S3 credentials until shortly before their reported expiry [datafusion-comet] via GitHub
2026/10/01 Re: [I] INSERT into an unpartitioned table fails when fanout is disabled [datafusion-iceberg] via GitHub
2026/10/01 Re: [I] INSERT into a partitioned table with fanout disabled always fails: the optimizer removes the partition sort [datafusion-iceberg] via GitHub
2026/10/01 Re: [I] INSERT into an unpartitioned table fails when fanout is disabled [datafusion-iceberg] via GitHub
2026/10/01 [PR] feat: Add option to assign Parquet row groups to file ranges by midpoint [datafusion] via GitHub
2026/10/01 Re: [PR] fix: Fix make_array null input handling [datafusion] via GitHub
2026/10/01 [PR] docs: add community showcase volumes 4 and 5 [datafusion] via GitHub
2026/10/01 Re: [PR] fix(scheduler): read a swapped broadcast partitioned on the probe side [datafusion-ballista] via GitHub
2026/10/01 [I] `convert_filters_to_predicate` negates a weakened AND, producing a predicate that drops matching rows [datafusion-iceberg] via GitHub
2026/10/01 [I] LIKE prefix pushdown ignores backslash escapes and drops matching rows [datafusion-iceberg] via GitHub
2026/10/01 Re: [I] Native Iceberg write merges -0.0 and 0.0 rows into one float/double identity partition [datafusion-comet] via GitHub
2026/10/01 [I] Option to assign Parquet row groups to file ranges by midpoint, as Spark does [datafusion] via GitHub
2026/10/01 [I] `SELECT *` fails after a column is added until the next write to the table [datafusion-iceberg] via GitHub
2026/10/01 [I] INSERT into an unpartitioned table fails when fanout is disabled [datafusion-iceberg] via GitHub
2026/10/01 [I] Partitioned INSERT computes partition values from the wrong columns when the input plan ends in a projection [datafusion-iceberg] via GitHub
2026/10/01 [I] Clustered (fanout disabled) writes fail on sorted input because batches are split in HashMap order [datafusion-iceberg] via GitHub
2026/10/01 [I] INSERT into a partitioned table with fanout disabled always fails: the optimizer removes the partition sort [datafusion-iceberg] via GitHub
2026/10/01 [I] Metadata tables ignore the scan projection, so projected and aggregate queries return wrong columns or fail [datafusion-iceberg] via GitHub
2026/10/01 [I] Filter pushdown strips casts on columns, which drops matching rows or fails valid queries [datafusion-iceberg] via GitHub
2026/10/01 [I] `CREATE TABLE` / `DROP TABLE` hang on a current-thread tokio runtime and panic outside a runtime [datafusion-iceberg] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [I] datafusion-proto: EmptyRelation loses its schema on a logical plan round trip [datafusion] via GitHub
2026/10/01 Re: [PR] feat: Column statistics support v1 [datafusion-ballista] via GitHub
2026/10/01 [I] datafusion-proto: EmptyRelation loses its schema on a logical plan round trip [datafusion] via GitHub
2026/10/01 Re: [I] Grouped `count(DISTINCT x)` on numbers takes a slower path [datafusion] via GitHub
2026/10/01 Re: [I] General framework to decorrelate the subqueries [datafusion] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: reuse S3 credentials until shortly before their reported expiry [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: Column statistics support v1 [datafusion-ballista] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] chore: Hide test-only internal public APIs [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [I] Grouped `count(DISTINCT x)` on numbers takes a slower path [datafusion] via GitHub
2026/10/01 Re: [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 [I] Grouped `count(DISTINCT x)` on numbers takes a slower path [datafusion] via GitHub
2026/10/01 Re: [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: support col reference for percentiles [datafusion] via GitHub
2026/10/01 [I] Support native existence sort-merge joins (blocked on DataFusion mark-SMJ output buffering) [datafusion-comet] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 [PR] fix: prefer Unsupported when combining nested cast support [datafusion-comet] via GitHub
2026/10/01 Re: [I] Correlated `EXISTS` subquery with `OFFSET` returns wrong results [datafusion] via GitHub
2026/10/01 Re: [PR] perf: cut schema check and subquery rescan overhead in optimizer [datafusion] via GitHub
2026/10/01 Re: [PR] perf: skip functional dependency work when inputs carry none [datafusion] via GitHub
2026/10/01 Re: [PR] chore(deps): bump datafusion-sqllogictest from 55.0.0 to 55.1.0 [datafusion-iceberg] via GitHub
2026/10/01 [PR] docs: check for behavior changes against the latest release in the review skill [datafusion-comet] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [PR] refactor: Make AggregateExec state modeling and public updates safe (part2: topk) [datafusion] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [I] Native existence join enumerates every duplicate build match (N*M candidates for M markers) [datafusion-comet] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [PR] chore: Hide test-only internal public APIs [datafusion] via GitHub
2026/10/01 Re: [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [PR] chore: Hide test-only internal public APIs [datafusion] via GitHub
2026/10/01 Re: [PR] docs: promote Spark 4.2 from experimental to supported [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 [PR] chore: Upgrade to Rust 1.99, MSRV to 1.95 [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [I] Avoid redundant sorts for scalar subquery expressions [datafusion] via GitHub
2026/10/01 Re: [PR] feat: close upstream coverage gaps for DataFusion 55.1.0 [datafusion-python] via GitHub
2026/10/01 [I] Native Parquet scan assigns row groups to splits differently from Spark [datafusion-comet] via GitHub
2026/10/01 [PR] chore(deps): bump pyjwt from 2.13.0 to 2.15.0 in /python [datafusion-ballista] via GitHub
2026/10/01 Re: [PR] IN LIST: treat signed zeros as equal [datafusion] via GitHub
2026/10/01 Re: [PR] feat: blocked agg with dynamic flat and blocked [datafusion] via GitHub
2026/10/01 Re: [I] Audit the PRs in 1.1.0 for regressions since 1.0.0 [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: add fully ordered group values implementation for primitive [datafusion] via GitHub
2026/10/01 Re: [PR] feat: reuse S3 credentials until shortly before their reported expiry [datafusion-comet] via GitHub
2026/10/01 [I] Record whether a window frame was set explicitly, so `ExprFunctionExt` can build on an existing window function [datafusion] via GitHub
2026/10/01 Re: [PR] fix: account for every partition in local TopK and limit statistics [datafusion] via GitHub
2026/10/01 Re: [PR] perf: emit initial filter output sooner for dynamic pruning [datafusion] via GitHub
2026/10/01 Re: [PR] test: enable SPARK-57298 collect_set tests [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: keep the recovery projection when leaf pushdown would drop a computed same-name column [datafusion] via GitHub
2026/10/01 Re: [PR] feat: add builders for `CreateExternalCatalog` and `DropCatalog` [datafusion] via GitHub
2026/10/01 Re: [I] Wrong results: leaf expression pushdown removes a computed column that has the same name as its input column [datafusion] via GitHub
2026/10/01 Re: [PR] fix: Fix make_array null input handling [datafusion] via GitHub
2026/10/01 Re: [PR] feat: add native Delta Lake scan contrib module (page/row-group pruning) [datafusion-comet] via GitHub
2026/10/01 Re: [PR] refactor: Make AggregateExec state modeling and public updates safe (part2: topk) [datafusion] via GitHub
2026/10/01 Re: [PR] feat: add parquet_file_metadata and parquet_page_index to datafusion-cli [datafusion] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] refactor: Make AggregateExec state modeling and public updates safe (part2: topk) [datafusion] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] IN LIST: treat signed zeros as equal [datafusion] via GitHub
2026/10/01 Re: [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 Re: [PR] chore: move native write configs under spark.comet.write [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] refactor: Use cast preimages for cast predicate rewrites [datafusion] via GitHub
2026/10/01 Re: [PR] fix: Fix make_array null input handling [datafusion] via GitHub
2026/10/01 Re: [PR] perf: add fully ordered group values implementation for primitive [datafusion] via GitHub
2026/10/01 Re: [I] `sum(DISTINCT x)` with `GROUP BY` uses twice the memory or more, unless the query also has `count(*)` [datafusion] via GitHub
2026/10/01 Re: [I] Specialized GroupValues Implementation for Ordered Data [datafusion] via GitHub
2026/10/01 Re: [PR] feat: add native support for MergeRowsExec (row-level MERGE INTO) [datafusion-comet] via GitHub
2026/10/01 Re: [I] Comet 1.1.0 Release (September) [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: wire native existence join support [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: support zero-field struct group keys in aggregate emit [datafusion] via GitHub
2026/10/01 [I] `sum(DISTINCT x)` with `GROUP BY` uses twice the memory or more, unless the query also has `count(*)` [datafusion] via GitHub
2026/10/01 [PR] fix: [branch-1.1] fall back to Spark for _metadata.file_block_start and file_block_length (#6510) [datafusion-comet] via GitHub
2026/10/01 Re: [PR] IN LIST: treat signed zeros as equal [datafusion] via GitHub
2026/10/01 [PR] fix: fall back to Spark for _metadata.file_block_start and file_block_length [datafusion-comet] via GitHub
2026/10/01 Re: [PR] docs: request changes for correctness and performance findings in the review skill [datafusion-comet] via GitHub
2026/10/01 [PR] docs: add AGENTS.md for coding agents [datafusion-iceberg] via GitHub
2026/10/01 [I] why python version older than the rust version [datafusion-python] via GitHub
2026/10/01 [PR] feat: reuse S3 credentials until shortly before their reported expiry [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: Validate `execution.time_zone` at `SET` time [datafusion] via GitHub
2026/10/01 [I] Quoted AST values and tokens can serialize to non-equivalent SQL [datafusion-sqlparser-rs] via GitHub
2026/10/01 Re: [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 [I] Add an AGENTS.md like other DataFusion repositories [datafusion-iceberg] via GitHub
2026/10/01 [I] S3 credential SPI: honor expirationEpochMillis consistently on the Parquet and Iceberg paths [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: gate array distinct and union signed-zero semantics by Spark version [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: reject scalar subqueries inside a codegen-dispatch kernel [datafusion-comet] via GitHub
2026/10/01 [PR] chore(deps): bump urllib3 from 2.7.0 to 2.8.0 in /python [datafusion-ballista] via GitHub
2026/10/01 Re: [PR] fix: Avoid pushing down sort under limits [datafusion] via GitHub
2026/10/01 Re: [PR] feat: route map lookups through codegen dispatcher [datafusion-comet] via GitHub
2026/10/01 [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 [I] Add release infrastructure [datafusion-iceberg] via GitHub
2026/10/01 [PR] fix: build with Rust 1.99, which deprecates the legacy f64 constants and fetch_update [datafusion-comet] via GitHub
2026/10/01 Re: [I] Audit the PRs in 1.1.0 for regressions since 1.0.0 [datafusion-comet] via GitHub
2026/10/01 [I] Native Parquet scan returns wrong _metadata.file_block_start when Spark splits a file [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: optimize JVM columnar-to-row conversion [datafusion-comet] via GitHub
2026/10/01 [I] Native Parquet scan fails on an empty file whose nested field type differs from the read schema [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: stop redoing per-chunk setup on the hash join probe path [datafusion] via GitHub
2026/10/01 Re: [PR] feat: per-location credentials for native Iceberg reads and writes [datafusion-comet] via GitHub
2026/10/01 Re: [PR] chore(deps): bump pyjwt from 2.13.0 to 2.15.0 in /python [datafusion-ballista] via GitHub
2026/10/01 [PR] chore: add release infrastructure and rename crate back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 Re: [PR] perf(pwmj):settle never-matching streamed rows against the buffered extreme… [datafusion] via GitHub
2026/10/01 Re: [PR] fix: match Spark's float ordering for sort and window keys nested in arrays and structs [datafusion-comet] via GitHub
2026/10/01 Re: [I] Crate name `datafusion-iceberg` is taken on crates.io; switch back to `iceberg-datafusion` [datafusion-iceberg] via GitHub
2026/10/01 Re: [PR] fix: report a cached relation's decoded size to the planner, not its compressed size [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: custom Rust UDFs via arrow-ffi [experimental] [datafusion-comet] via GitHub
2026/10/01 Re: [PR] refactor: remove unused cast compatibility flag [datafusion-comet] via GitHub
2026/10/01 Re: [I] Examples of using `TreeNode` APIs to walk and manipulate LogicalPlans [datafusion] via GitHub
2026/10/01 Re: [PR] feat: Use DataFusion date_trunc for compatible trunc and date_trunc paths [datafusion-comet] via GitHub
2026/10/01 Re: [PR] Make StatisticsContext cache entries node-lifetime safe [datafusion] via GitHub
2026/10/01 Re: [PR] fix: run reverse of a binary value through the codegen dispatcher [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: keep Spark's cache scan for a relation whose cached plan records observed metrics [datafusion-comet] via GitHub
2026/10/01 Re: [PR] Allow the `HLLAccumulator` used by `approx_distinct()` to consume state produced by `HllGroupsAccumulator` [datafusion] via GitHub
2026/10/01 Re: [PR] fix: Fix make_array null input handling [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] fix: deduplicate StringView/BinaryView buffer refs in CollectLeft Has... [datafusion] via GitHub
2026/10/01 Re: [PR] feat: cost-based join order enumeration [datafusion] via GitHub
2026/10/01 [I] Native Iceberg scan fails on files written before a nested field was added, and since 1.1.0 null checks and explode hit it too [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: read the Iceberg write gate's storage scheme the same way the native factory does [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 [PR] refactor: move the self-contained operators from core into a new operators crate [datafusion-comet] via GitHub
2026/10/01 Re: [PR] refactor: move the remaining PhysicalExpr impls from core into spark-expr [datafusion-comet] via GitHub
2026/10/01 Re: [PR] feat: Use NativeType in get_example_type, information schema [datafusion] via GitHub
2026/10/01 Re: [PR] Add SECURITY.md for DataFusion [datafusion] via GitHub
2026/10/01 Re: [PR] fix: arrays_zip with duplicate field names falls back to Spark [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: rescale decimal results from dispatched functions to the declared type [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: mirror compound ASOF key predicates to right [datafusion] via GitHub
2026/10/01 Re: [PR] fix: match Iceberg's rounding for pre-1970 timestamps in native years/months/days/hours [datafusion-comet] via GitHub
2026/10/01 Re: [PR] build: bump semanticdb to 4.13.10 so scalafix runs on Spark 4.1 and 4.2 [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: stop redoing per-chunk setup on the hash join probe path [datafusion] via GitHub
2026/10/01 Re: [PR] fix: match Iceberg's rounding for pre-1970 timestamps in native years/months/days/hours [datafusion-comet] via GitHub
2026/10/01 Re: [PR] test: close five mutation-testing gaps in leaf expression pushdown guards [datafusion] via GitHub
2026/10/01 Re: [PR] fix: fall back to Spark for regr_* aggregates until their merge matches Spark [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: mirror compound ASOF key predicates to right [datafusion] via GitHub
2026/10/01 Re: [I] Physical optimizer re-runs rules on plans they have already settled [datafusion] via GitHub
2026/10/01 Re: [PR] fix: keep a subquery correlated when its filter is below a window function [datafusion] via GitHub
2026/10/01 Re: [PR] feat: add builders for `CreateExternalCatalog` and `DropCatalog` [datafusion] via GitHub
2026/10/01 Re: [PR] perf: Optimize `array_agg(... IGNORE NULLS)` [datafusion] via GitHub
2026/10/01 Re: [PR] IN LIST: treat signed zeros as equal [datafusion] via GitHub
2026/10/01 [PR] fix: read the Iceberg write gate's storage scheme the same way the native factory does [datafusion-comet] via GitHub
2026/10/01 Re: [PR] fix: match Iceberg's rounding for pre-1970 timestamps in native years/months/days/hours [datafusion-comet] via GitHub
2026/10/01 Re: [PR] perf: mirror compound ASOF key predicates to right [datafusion] via GitHub
2026/10/01 Re: [PR] perf: Optimize `array_agg(... IGNORE NULLS)` [datafusion] via GitHub
2026/10/01 Re: [PR] fix: show results in the nine silent top-level examples [datafusion-python] via GitHub
2026/10/01 Re: [PR] chore: Hide test-only internal public APIs [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] perf: add fully ordered group values implementation for primitive [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] feat: run physical requirement enforcement first, as a PhysicalAnalyzer phase [datafusion] via GitHub
2026/10/01 Re: [PR] fix: coalesce shuffle partitions under Comet unions the way Spark does [datafusion-comet] via GitHub
2026/10/01 Re: [I] Split physical optimizer rules into enforcement and optimization phases, with a convergence loop for the optimizers [datafusion] via GitHub
2026/10/01 Re: [I] `sum(DISTINCT x)` with `GROUP BY` uses twice the memory or more, unless the query also has `count(*)` [datafusion] via GitHub
2026/10/01 Re: [PR] perf: stop redoing per-chunk setup on the hash join probe path [datafusion] via GitHub