Messages by Thread
-
[I] sort should not be removed when order by a + b where a asc and b asc as they can overflow and wrap [datafusion]
via GitHub
-
Re: [PR] fix(substrait): roundtrip scalar subqueries in projections [datafusion]
via GitHub
-
Re: [PR] fix(substrait): preserve window function semantics [datafusion]
via GitHub
-
Re: [PR] fix(substrait): preserve grouping set output order [datafusion]
via GitHub
-
Re: [PR] fix: use build_join_schema for CrossJoinExec output schema metadata [datafusion]
via GitHub
-
[PR] fix: don't infer join predicates for null-aware joins in push_down_filter [datafusion]
via GitHub
-
[PR] build(deps): bump serde_json from 1.0.150 to 1.0.151 [datafusion-python]
via GitHub
-
[PR] build(deps): bump datafusion from 54.0.0 to 54.1.0 [datafusion-python]
via GitHub
-
[PR] build(deps): bump datafusion-ffi from 54.0.0 to 54.1.0 [datafusion-python]
via GitHub
-
[PR] build(deps): bump datafusion-common from 54.0.0 to 54.1.0 [datafusion-python]
via GitHub
-
[I] push_down_filter infers a null-rejecting predicate onto the subquery side of a null-aware NOT IN join, producing wrong results [datafusion]
via GitHub
-
[PR] build(deps): bump datafusion-proto from 54.0.0 to 54.1.0 [datafusion-python]
via GitHub
-
[PR] build(deps): bump tokio from 1.52.3 to 1.53.1 [datafusion-python]
via GitHub
-
[PR] build(deps): bump datafusion-spark from 54.0.0 to 54.1.0 [datafusion-python]
via GitHub
-
[PR] build(deps): bump async-trait from 0.1.89 to 0.1.91 [datafusion-python]
via GitHub
-
[PR] build(deps): bump datafusion-catalog from 54.0.0 to 54.1.0 [datafusion-python]
via GitHub
-
[PR] chore: adjust `size` accounting for `min_max` [datafusion]
via GitHub
-
[PR] fix(spark): correct pmod overflow, ANSI zero divisor and negative zero handling [datafusion]
via GitHub
-
[I] [Bug] mod returns NaN instead of raising for a zero floating point divisor in ANSI mode, and ignores -0.0 divisors [datafusion]
via GitHub
-
[I] [Bug] pmod rejects string arguments that Spark implicitly casts to a numeric type [datafusion]
via GitHub
-
[I] [Bug] pmod and mod report Arrow's divide-by-zero message instead of Spark's REMAINDER_BY_ZERO text [datafusion]
via GitHub
-
[I] [Bug] pmod returns a wider decimal type than Spark [datafusion]
via GitHub
-
[PR] feat: add native make_interval support [datafusion-comet]
via GitHub
-
[PR] perf: reuse Arrow IPC compression context across shuffle blocks [datafusion-comet]
via GitHub
-
Re: [PR] Cli entrypoint pr2 [datafusion]
via GitHub
-
[PR] docs: add an agent skill for auditing datafusion-spark expressions [datafusion]
via GitHub
-
[PR] fix: make next_day match Spark 4.2.0 semantics [datafusion]
via GitHub
-
[I] [Bug] next_day returns NULL for far-future start dates where Spark returns a value [datafusion]
via GitHub
-
[I] [Bug] next_day cross-version divergence: ANSI default flip and error class change between Spark 3.5 and 4.x [datafusion]
via GitHub
-
[I] [Bug] next_day rejects STRING and TIMESTAMP start dates that Spark accepts [datafusion]
via GitHub
-
[PR] chore(deps-dev): bump shell-quote from 1.8.3 to 1.10.0 in /datafusion/wasmtest/datafusion-wasm-app [datafusion-sandbox]
via GitHub
-
Re: [PR] chore(deps-dev): bump shell-quote from 1.8.3 to 1.8.4 in /datafusion/wasmtest/datafusion-wasm-app [datafusion-sandbox]
via GitHub
-
[I] Support Spark-version-specific expectations in datafusion-spark sqllogictest files [datafusion]
via GitHub
-
[PR] feat(parquet): multi-column lexicographic stats reorder for TopK sort pushdown [datafusion]
via GitHub
-
[PR] fix: coalesce AQE shuffle partitions per alignment group [datafusion-ballista]
via GitHub
-
[I] DataFusion 54.1.0 has a breaking API change [datafusion]
via GitHub
-
Re: [PR] docs: add 54.0.0 changelog [datafusion-ballista]
via GitHub
-
[PR] perf: parse next_day's day-of-week without allocating, and once per batch [datafusion]
via GitHub
-
[PR] perf: remove per-row String allocation from Spark url_encode [datafusion]
via GitHub
-
[I] Preserve bitmap-backed `RowSelection` in `ParquetAccessPlan` [datafusion]
via GitHub
-
[PR] bench: add Criterion benchmarks for Spark bin, soundex, quote, and url_encode [datafusion]
via GitHub
-
[PR] feat: support Spark encode expression via codegen dispatch [datafusion-comet]
via GitHub
-
[PR] perf: remove per-row String allocation from Spark soundex and quote [datafusion]
via GitHub
-
[PR] perf: avoid per-row String allocation in Spark bin and char [datafusion]
via GitHub
-
[PR] fix(sql): preserve source qualifiers in CTAS with explicit schema [datafusion]
via GitHub
-
[I] CTAS with an explicit schema fails for qualified columns with the same name [datafusion]
via GitHub
-
[I] Numeric SQL literals with underscores are not correctly parsed [datafusion]
via GitHub
-
Re: [D] DISCUSSION: DataFusion Meetup in Denver, CO USA [datafusion]
via GitHub
-
[PR] fix: prevent libhdfs thread destructor use-after-free [datafusion-comet]
via GitHub
-
[I] [C++][Engine] Substrait consumer silently drops modern FetchRel and AggregateRel fields [datafusion]
via GitHub
-
[PR] chore(deps-dev): bump fast-uri from 3.1.0 to 3.1.4 in /datafusion/wasmtest/datafusion-wasm-app [datafusion-sandbox]
via GitHub
-
Re: [PR] chore(deps-dev): bump fast-uri from 3.1.0 to 3.1.2 in /datafusion/wasmtest/datafusion-wasm-app [datafusion-sandbox]
via GitHub
-
[PR] chore(deps): bump setuptools from 82.0.0 to 83.0.0 [datafusion-sandbox]
via GitHub
-
[PR] build(deps-dev): bump setuptools from 82.0.1 to 83.0.0 [datafusion-python]
via GitHub
-
Re: [I] Shuffle-fetch failures lose their type, so the map-stage resubmit never fires and the job fails [datafusion-ballista]
via GitHub
-
Re: [PR] feat(substrait): serialize correlated subqueries with OuterReference field references [datafusion]
via GitHub
-
[PR] feat: native randstr implementation compatible with Spark [datafusion-comet]
via GitHub
-
[PR] fix: bump DataFusion to 54.1.0 and adapt to RecursiveQuery API change [datafusion-python]
via GitHub
-
[I] Build fails against DataFusion 54.1.0 due to new `schema` field on `RecursiveQuery` [datafusion-python]
via GitHub
-
[PR] feat: native uuid() implementation compatible with Spark [datafusion-comet]
via GitHub
-
[PR] kll: add native Rust KLL quantile sketch [datafusion-ballista]
via GitHub
-
Re: [PR] chore(deps): track DataFusion git main and port to its APIs [datafusion-ballista]
via GitHub
-
[PR] chore: use taplo for Cargo.toml formatting [datafusion-ballista]
via GitHub
-
Re: [I] Garbled Ballista banner in TUI [datafusion-ballista]
via GitHub
-
[PR] feat: native pow implementation compatible with Spark [datafusion-comet]
via GitHub
-
Re: [I] Proto: migrate AsyncFuncExec [datafusion]
via GitHub
-
[PR] feat: route pow, translate, and to_csv through codegen dispatch by default [datafusion-comet]
via GitHub
-
[PR] test: add IN list slt coverage for temporal, Decimal128 and Interval types [datafusion]
via GitHub
-
[PR] test: add CometExpressionCoverageSuite for runtime expression coverage audit [datafusion-comet]
via GitHub
-
[PR] fix: sliding window MIN returns wrong value for all-NULL windows [datafusion]
via GitHub
-
[PR] feat: implement pmod natively for all numeric types [datafusion-comet]
via GitHub
-
[PR] feat: support timestampadd and timestampdiff via codegen dispatch [datafusion-comet]
via GitHub
-
[PR] ci: allow the docs site deploy to be triggered manually [datafusion-ballista]
via GitHub
-
Re: [I] Proto: migrate HashJoinExec [datafusion]
via GitHub
-
[PR] docs: show implementation kind for each expression [datafusion-comet]
via GitHub
-
[PR] fix: eliminate group by constant empty input [datafusion]
via GitHub
-
[PR] chore(deps): bump quinn-proto from 0.11.14 to 0.11.16 [datafusion-ballista]
via GitHub
-
Re: [PR] feat: Optionally split the Iceberg V2 write operator into distinct writer and committer operations [datafusion-comet]
via GitHub
-
[PR] test: cover `array_agg(DISTINCT)` on dictionaries and bounded `retract_batch` memory [datafusion]
via GitHub
-
Re: [PR] feat: eliminate LEFT/RIGHT JOINs with redundant sides [datafusion]
via GitHub
-
Re: [PR] fix: Handle `input_file_name()` pushdown into `ParquetSource` with filter pushdown enabled [datafusion]
via GitHub
-
[PR] feat(core,scheduler): transport RuntimeStatsExec reports to scheduler; log merged quantile cuts per stage [datafusion-ballista]
via GitHub
-
Re: [PR] Fix UnnestExec misattributing column properties on duplicated column names [datafusion]
via GitHub
-
[I] Sliding window `min()` returns stale value when all non-`NULL` values leave the frame [datafusion]
via GitHub
-
Re: [I] Integrate collect_list/array_agg to Comet [datafusion-comet]
via GitHub
-
[PR] feat: account JVM UDF Arrow allocations in Spark task memory [datafusion-comet]
via GitHub
-
Re: [PR] feat: memoize for FFI_PartitionEvaluator [datafusion]
via GitHub
-
[I] Add Iceberg integration [datafusion-ballista]
via GitHub
-
[PR] refactor: group orphan `spark.comet.explain.*` configs under one prefix [datafusion-comet]
via GitHub
-
[PR] chore(deps-dev): bump webpack-dev-server from 5.2.6 to 6.0.0 in /datafusion/wasmtest/datafusion-wasm-app [datafusion]
via GitHub
-
Re: [I] ci: Remove ORC testing from Iceberg Java tests [datafusion-comet]
via GitHub
-
[PR] feat: support interval codegen dispatch [datafusion-comet]
via GitHub
-
[I] Sporadic SIGSEGV in macOS [scans] workflow in CI [datafusion-comet]
via GitHub
-
[PR] feature: Calculate non-distinct `sum` from column statistics when available [datafusion]
via GitHub
-
[PR] Fix/11748 agg emit flag [datafusion]
via GitHub
-
[PR] feat: bump default gRPC max message size 16 MiB -> 128 MiB [datafusion-ballista]
via GitHub
-
Re: [PR] Move `main_inner` into datafusion-cli lib for subsequent PR [datafusion]
via GitHub
-
[I] Chunk oversized shuffle IPC messages so they never exceed the gRPC frame limit [datafusion-ballista]
via GitHub
-
[PR] feat(pwmj): support LeftSemi/LeftAnti existence joins via classic scan [datafusion]
via GitHub
-
[I] Follow-up: docs and test-coverage hardening for native collect_list / array_agg (#4720) [datafusion-comet]
via GitHub
-
Re: [I] Support WindowGroupLimitExec (window-based limit pushdown) [datafusion-comet]
via GitHub
-
[PR] docs: refresh benchmarking.md with dedicated shuffle PVC results [datafusion-ballista]
via GitHub
-
[PR] feat(core): eagerly reclaim job data when the client finishes consuming results [datafusion-ballista]
via GitHub