All pull requests

Pull requests

• fix: close Iceberg cleanup ownership gap during JVM handoff

  area:Icebergarea:writerbug
  #6582·unikdahal opened Oct 3, 2026Contributor·

  ·
  1

• fix: [branch-1.0] zero sliced boolean offsets at every level before exporting to the JVM (#6339)

  area:expressionsarea:memoryarea:udfbug
  #6579·andygrove opened Oct 3, 2026Member·

  ·

• fix: write columnar structs to Arrow schema width

  bug
  #6578·unikdahal opened Oct 3, 2026Contributor·

  ·
  1

• fix: draw the cached plan below CometInMemoryTableScan in the SQL tab and event log

  area:scanbugrun-all-spark-profiles
  #6577·comphead opened Oct 3, 2026Contributor·

  ·
  1

  2

• fix: draw the cached plan below CometInMemoryTableScan in EXPLAIN

  area:scanbugrun-all-spark-profiles
  #6574·comphead opened Oct 3, 2026Contributor·

  ·
  1

• perf: write Spark columns to Arrow in bulk in CometSparkToColumnarExec

  area:ffiarea:scanenhancementperformance
  #6566·andygrove opened Oct 3, 2026Member·

  ·
  3

• feat: run the operators above a typed Dataset operation natively

  area:shuffleenhancementrun-benchmark-checkrun-spark-4.1-tests
  #6564·andygrove opened Oct 3, 2026Member·

  ·
  1

  5

• feat: normalize floats in native array_distinct and array_union

  area:expressionsenhancement
  #6563·andygrove opened Oct 3, 2026Member·

  ·
  1

• fix: convert native ReadAncientDatetime errors to SparkUpgradeException

  area:ffibug
  #6559·dwsmith1983 opened Oct 3, 2026Contributor·

  ·
  1

  1

• fix: stop the leak of Arrow memory from struct outputs of the codegen dispatcher

  area:expressionsbug
  #6552·peterxcli opened Oct 2, 2026Member·

  ·

• feat: support native wide-decimal hashing

  area:expressionsarea:shuffleenhancement
  #6548·rich7420 opened Oct 2, 2026Contributor·

  ·
  1

• fix: let a spilled final aggregate read its spill files back past its memory share

  area:aggregationarea:memorybugregression
  #6544·andygrove opened Oct 2, 2026Member·

  ·
  1

• fix: keep Spark's cache format when Kryo would reject Comet's or Comet disables itself

  bug
  #6537·andygrove opened Oct 2, 2026Member·

  ·
  2

• refactor: move the remaining JNI-free entry point logic into core functions

  enhancement
  #6534·viirya opened Oct 2, 2026Member·

  ·
  1

• Fix array_except type mismatch issue #3646

  area:expressions
  #6533·contentforge-press opened Oct 2, 2026·4 of 6 tasks·

  ·
  1

  6

• fix: read the Iceberg write gate's storage scheme the same way the native factory does

  area:Icebergarea:scanarea:writerbug
  #6502·0lai0 opened Oct 1, 2026Contributor·

  ·
  1

  4

• fix: prefer Unsupported when combining nested cast support

  area:expressionsbug
  #6501·liupoyi-1031 opened Oct 1, 2026Contributor·

  ·
  1

  2

• feat: Experimental local execution mode that runs a whole query as one DataFusion graph

  enhancementrun-spark-4.1-tests
  #6500·viirya opened Oct 1, 2026Member·

  ·
  1

  13

• fix: compact oversized nested list_extract buffers

  area:expressionsbug
  #6498·rich7420 opened Oct 1, 2026Contributor·

  ·
  1

  6

• perf: reduce Parquet runtime filter schema guard overhead

  area:joinsarea:scanenhancementperformancerun-all-spark-profiles
  #6496·pingzh opened Oct 1, 2026Contributor·

  ·
  1

  3

• test: share collation coverage across Spark 4.x

  enhancementrun-all-spark-profilestest
  #6495·sunchao opened Oct 1, 2026Member·

  ·
  1

• fix: Route collated array element comparisons through the codegen dispatcher

  area:expressionsbug
  #6471·viirya opened Sep 30, 2026Member·

  ·
  1

• docs: add release notes for 1.1.0 with known regressions and workarounds

  documentation
  #6469·andygrove opened Sep 30, 2026Member·

  ·

• fix: coalesce shuffle partitions under Comet unions the way Spark does

  bug
  #6459·andygrove opened Sep 30, 2026Member·

  ·
  1

  9

• fix: match Spark's float ordering in min, max, greatest and least

  area:aggregationarea:expressionsbugrun-spark-4.1-tests
  #6457·andygrove opened Sep 30, 2026Member·

  ·
  13