All pull requests

Pull requests

• fix: treat str.split delimiters as literal substrings

  #4124·Klu5ure opened Nov 2, 2026·6 of 8 tasks·

  ·
  4

• feat: Add Series.from_arrow and Series.__arrow_c_array__

  #4122·RamyAkin opened Oct 31, 2026·7 of 13 tasks·

  ·
  1

  3

• feat: add {Expr,Series}.approx_quantile

  #4121·yash-bitla opened Oct 31, 2026·6 of 13 tasks·

  ·
  1

  8

• fix: return empty strings from str.tail(0)

  #4120·Mike-Animal-Counseling opened Oct 31, 2026·8 of 13 tasks·

  ·
  1

• fix: match NaN join keys on DuckDB<2.0

  #4119·AbderrahimNeji opened Oct 29, 2026·6 of 13 tasks·

  ·
  3

• feat: accept an expression in Expr.dt.offset_by

  #4116·555aaditya opened Oct 28, 2026Contributor·5 of 13 tasks·

  ·
  1

  5

• feat: implement .get operation for Map datatype

  enhancementnested data
  #4115·danielgafni opened Oct 27, 2026Contributor·5 of 12 tasks·

  ·
  1

  45

• feat: allow negative indices in list.get

  #4108·555aaditya opened Oct 25, 2026Contributor·5 of 13 tasks·

  ·
  8

• feat: support str.replace with n=1 for DuckDB

  #4107·555aaditya opened Oct 25, 2026Contributor·4 of 13 tasks·

  ·
  4

• fix: support grouped null_count on pandas-like and PyArrow

  #4102·dsilvadeepal opened Oct 22, 2026·6 of 13 tasks·

  ·
  1

  1

• fix(pandas-like): return Python scalars from rows/iter_rows for nullable dtypes

  #4094·senanurcetin opened Oct 20, 2026·5 of 13 tasks·

  ·
  1

  2

• refactor: Simplify Selector set ops implementation

  internal
  #4088·FBruzzesi opened Oct 18, 2026Member·1 of 7 tasks·

  ·
  1

• fix: preserve dtype for skew / kurtosis on empty or all-null columns for pandas-like backends

  #4083·lucaluo925 opened Oct 17, 2026Contributor·4 of 12 tasks·

  ·
  1

  8

• fix: return null for modulo by zero on pandas-like backends (#4036)

  #4062·ioseba opened Oct 11, 2026Contributor·

  ·
  1

  1

• fix: return null from mean_horizontal for an all-null row on PyArrow, DuckDB, Ibis and Spark-like

  fix
  #4059·SausageMan99 opened Oct 11, 2026·5 of 13 tasks·

  ·
  1

  3

• fix: return integer count from Series.hist on empty or all-null p…

  #4047·RavSinghChandan opened Oct 8, 2026·

  ·
  1

  1

• test: pin Polars 2.0 zero-width frame semantics for with_columns; match them on polars<2

  eager-onlyfix
  #4046·rohanz opened Oct 8, 2026·5 of 12 tasks·

  ·
  12

• feat: Add maintain_order to DataFrame.group_by

  #4042·Stewyn opened Oct 7, 2026·6 of 13 tasks·

  ·
  1

  5

• Align fractional float casts to signed integers in the main API

  #4037·Lingikaushikreddy opened Oct 6, 2026·8 of 9 tasks·

  ·
  1

  1

• fix: return Int64 from null_count on SQL backends, and 0 on empty frames

  duckdbfixibisspark-likesql-like
  #4030·FBruzzesi opened Oct 5, 2026Member·1 of 7 tasks·

  ·
  1

  1

• fix: broadcast scalar receiver in fill_null when value is a column for pandas-like and pyarrow

  fixpandas-likepyarrow
  #4009·FBruzzesi opened Sep 29, 2026Member·1 of 7 tasks·

  ·
  4

• fix: compute %s (seconds since epoch) explicitly in PyArrow dt.to_string

  #4000·Vinayak19112003 opened Sep 27, 2026Contributor·5 of 12 tasks·

  ·
  1

• fix: align round to round half to even

  fix
  #3977·FBruzzesi opened Sep 23, 2026Member·2 of 9 tasks·

  ·
  1

• feat: Expose __version__ from stable namespaces

  enhancement
  #3974·maxzw opened Sep 21, 2026Contributor·3 of 13 tasks·

  ·
  1

  4

• feat: add on_columns argument to DataFrame.pivot

  enhancement
  #3971·FBruzzesi opened Sep 21, 2026Member·1 of 7 tasks·

  ·
  1