Whats new in 0.24.2 (March 12, 2019)


The 0.24.x series of releases will be the last to support Python 2. Future feature releases will support Python 3 only. See Dropping Python 2.7 for more.

These are the changes in pandas 0.24.2. See Release Notes for a full changelog including other versions of pandas.

Fixed regressions

  • Fixed regression in DataFrame.all() and DataFrame.any() where bool_only=True was ignored (GH25101)
  • Fixed issue in DataFrame construction with passing a mixed list of mixed types could segfault. (GH25075)
  • Fixed regression in DataFrame.apply() causing RecursionError when dict-like classes were passed as argument. (GH25196)
  • Fixed regression in DataFrame.replace() where regex=True was only replacing patterns matching the start of the string (GH25259)
  • Fixed regression in DataFrame.duplicated(), where empty dataframe was not returning a boolean dtyped Series. (GH25184)
  • Fixed regression in Series.min() and Series.max() where numeric_only=True was ignored when the Series contained Categorical data (GH25299)
  • Fixed regression in subtraction between Series objects with datetime64[ns] dtype incorrectly raising OverflowError when the Series on the right contains null values (GH25317)
  • Fixed regression in TimedeltaIndex where np.sum(index) incorrectly returned a zero-dimensional object instead of a scalar (GH25282)
  • Fixed regression in IntervalDtype construction where passing an incorrect string with ‘Interval’ as a prefix could result in a RecursionError. (GH25338)
  • Fixed regression in creating a period-dtype array from a read-only NumPy array of period objects. (GH25403)
  • Fixed regression in Categorical, where constructing it from a categorical Series and an explicit categories= that differed from that in the Series created an invalid object which could trigger segfaults. (GH25318)
  • Fixed regression in to_timedelta() losing precision when converting floating data to Timedelta data (GH25077).
  • Fixed pip installing from source into an environment without NumPy (GH25193)
  • Fixed regression in DataFrame.replace() where large strings of numbers would be coerced into int64, causing an OverflowError (GH25616)
  • Fixed regression in factorize() when passing a custom na_sentinel value with sort=True (GH25409).
  • Fixed regression in DataFrame.to_csv() writing duplicate line endings with gzip compress (GH25311)

Bug fixes


  • Better handling of terminal printing when the terminal dimensions are not known (GH25080)
  • Bug in reading a HDF5 table-format DataFrame created in Python 2, in Python 3 (GH24925)
  • Bug in reading a JSON with orient='table' generated by DataFrame.to_json() with index=False (GH25170)
  • Bug where float indexes could have misaligned values when printing (GH25061)





  • Bug in Series.is_unique() where single occurrences of NaN were not considered unique (GH25180)
  • Bug in merge() when merging an empty DataFrame with an Int64 column or a non-empty DataFrame with an Int64 column that is all NaN (GH25183)
  • Bug in IntervalTree where a RecursionError occurs upon construction due to an overflow when adding endpoints, which also causes IntervalIndex to crash during indexing operations (GH25485)
  • Bug in Series.size raising for some extension-array-backed Series, rather than returning the size (GH25580)
  • Bug in resampling raising for nullable integer-dtype columns (GH25580)


A total of 25 people contributed patches to this release. People with a “+” by their names contributed a patch for the first time.

  • Albert Villanova del Moral
  • Arno Veenstra +
  • chris-b1
  • Devin Petersohn +
  • EternalLearner42 +
  • Flavien Lambert +
  • gfyoung
  • Gioia Ballin
  • jbrockmendel
  • Jeff Reback
  • Jeremy Schendel
  • Johan von Forstner +
  • Joris Van den Bossche
  • Josh
  • Justin Zheng
  • Kendall Masse
  • Matthew Roeschke
  • Max Bolingbroke +
  • rbenes +
  • Sterling Paramore +
  • Tao He +
  • Thomas A Caswell
  • Tom Augspurger
  • Vibhu Agarwal +
  • William Ayd
  • Zach Angell
Scroll To Top