All pull requests

Pull requests

• [WIP][PYTHON] Improve type annotation for vector operations and factories

  #59586·bearomorphism opened Oct 18, 2026·

  ·

• [SPARK-60327][PYTHON] ML and MLlib Vector cleanups

  #59585·bearomorphism opened Oct 18, 2026·

  ·

• [SPARK-60326][SQL] Pass the ORC SearchArgument to the reader directly

  #59584·SEPURI-SAI-KRISHNA opened Oct 18, 2026Contributor·

  ·

• [SPARK-60324][SQL] Normalize -0.0 and NaN in floating-point dynamic partition values

  #59583·SEPURI-SAI-KRISHNA opened Oct 18, 2026Contributor·

  ·

• [SPARK-60325][SQL] Re-enable Arrow columnar input to Python UDFs, holding only vectors that are not reused

  #59582·viirya opened Oct 18, 2026Member·

  ·

• [SPARK-60306][SQL] Release consumed columnar batches in non-codegen ColumnarToRow and Arrow Python UDFs

  #59581·viirya opened Oct 18, 2026Member·

  ·
  3

• [SPARK-60188][PYTHON] Warn on unknown transpiler names in pyTranspilers

  #59580·guan404ming opened Oct 17, 2026Member·

  ·

• [SPARK-57838][SQL] Harden overflow and range handling for nanosecond timestamps

  #59579·stevomitric opened Oct 17, 2026Contributor·

  ·

• [SPARK-57826][SQL] Reject nanosecond timestamps in percentile_approx and histogram_numeric

  #59578·stevomitric opened Oct 17, 2026Contributor·

  ·

• [SPARK-57836][SQL] Support nanosecond precision in to_timestamp, to_timestamp_ntz, to_timestamp_ltz and try_to_timestamp

  #59577·stevomitric opened Oct 17, 2026Contributor·

  ·

• [WIP][SQL][DOCS] Document the migration path from spark.sql.legacy.parquet.nanosAsLong to nanosecond timestamp types.

  #59576·stevomitric opened Oct 17, 2026Contributor·

  ·

• [WIP][SQL] Do not implicitly narrow nanosecond-precision timestamps

  #59575·stevomitric opened Oct 17, 2026Contributor·

  ·

• [SPARK-60322][SQL] Fix Parquet and ORC filter pushdown dropping rows for comparisons with a zero

  #59574·SEPURI-SAI-KRISHNA opened Oct 17, 2026Contributor·

  ·

• [SPARK-60315][SQL] Fix inconsistent filter behavior with whole-stage codegen on and off when a conjunct is NULL

  #59573·MaxGekk opened Oct 17, 2026Member·

  ·
  2

• [SPARK-60320][SQL] Fix wrong values from the vectorized DELTA_BYTE_ARRAY reader for uncompressed pages

  #59572·SEPURI-SAI-KRISHNA opened Oct 17, 2026Contributor·

  ·

• [SPARK-60314][SQL] Make Literal.sql preserve the precision of TIME(7)-TIME(9) literals

  #59571·dongjoon-hyun opened Oct 17, 2026Member·

  ·
  7

• [SPARK-60309][SQL] Fix mode to count binary values by content instead of by reference

  #59567·dongjoon-hyun opened Oct 17, 2026Member·

  ·

• [SPARK-60321][SQL] Avoid intermediate copies of byte-array-backed strings in Parquet writes

  #59566·saths008 opened Oct 17, 2026·

  ·
  1

• [SPARK-60305][PYTHON] Keep the Python runner's input buffer alive when an Arrow write is interrupted

  #59565·viirya opened Oct 17, 2026Member·

  ·
  20

• [SPARK-60308][SQL][WEBUI] Fix exponential re-renders on stage/task checkbox clicks in SQL plan detailed mode

  #59564·dongjoon-hyun opened Oct 17, 2026Member·

  ·

• [SPARK-XXXXX][CONNECT][PYTHON] Use FieldDescriptor.is_repeated when truncating plans for logging

  #59563·RamonZhou opened Oct 16, 2026Contributor·

  ·

• [SPARK-60301] Use the session time zone when printing variant values as strings

  #59561·chenhao-db opened Oct 16, 2026Contributor·

  ·
  1

• [SPARK-39142][SPARK-42235][PYTHON] Fix type checker errors for pandas_udf without functionType

  #59558·hzmsh opened Oct 16, 2026·

  ·
  2

• [SPARK-52878][SQL] Rewrite LogicalRDD attribute references by exprId to fix checkpoint with case-insensitive references

  #59557·mbyne opened Oct 16, 2026·

  ·

• [SPARK-60293][SQL] Resolve Parquet pushed filters and vectorized read types by field id

  #59552·andygrove opened Oct 16, 2026Member·

  ·