[pandas] annotated tag v0.4.3 created (now 3ec60f3)

[Andreas Tille]

This is an automated email from the git hooks/post-receive script.

tille pushed a change to annotated tag v0.4.3
in repository pandas.

        at  3ec60f3   (tag)
   tagging  a8b14798fe18050e486845b26370d4899f46a8b3 (commit)
  replaces  v0.4.2
 tagged by  Wes McKinney
        on  Sun Oct 9 17:57:38 2011 -0400

- Log -----------------------------------------------------------------
Version 0.4.3

Thomas Kluyver (13):
      BUG: Use stdlib csv module to write csv, so commas in data are escaped
      Pandas installs on Python 3
      Pandas can now be imported successfully in Python 3.
      Fixes so more tests pass on Python 3.
      Fix up more tests for Python 3.
      More changes to support Python 3 better.
      Hack so test works on Python 3.
      Use .iterkv() instead of .iteritems(), which gets incorrectly converted to .items() by 2to3.
      Sort out division for in sparse code.
      All but one test now passing on Python 3.
      Add isnull and notnull methods to Series.
      Add trove classifiers for Python 3 compatibility.
      Fix bug writing Series to CSV in Python 3.

Wes McKinney (35):
      BUG: intersection fails on monotonic Index.intersection(Int64Index) or vice-versa
      RLS: bump minor version to 0.4.3
      TST: test coverage in pandas.core
      TST: skip ExcelFile tests if no xlrd
      BUG: MultiIndex.sortlevel did not preserve names GH #202
      BUG: fixed a series of bugs resulting from incorrect appending of MultiIndex
      RLS: update release notes
      BUG: isnull/notnull fail on other objects like datetime
      ENH: unit testing (failing) for Series.name integration
      BUG: raise more helpful exception when trying to write empty object to HDFStore, address GH #204
      ENH: beginning name integration in Index, Series
      ENH: next stage of name attribute integration in Series. more tests needed
      ENH: proper handling of name in stack, to_csv bugfixes with name/names
      ENH: for Series.name integration and tests
      ENH: droplevel name/names fix, pivot correctly sets names now (nice)
      ENH: Series.name integration, working on sparse compat and adding unit
      RLS: update release notes
      Merge PR #200 bringing Python 3 support
      ENH: SparseSeries binary op speed enhancement in the block case, address GH #205
      BUG: change python access to buffer access
      RLS: update release notes
      RLS: update release notes
      ENH: print 25%-75% quartiles instead of 10%-90% deciles in describe. Address
      BLD: only import setuptools in py3k as cython command doesn't currently work
      TST: skip test on sparse. release notes
      BUG: work around unicode bug in py3.1 on win32
      TST: is_lexsorted test failure on py3.1/win32
      BUG: explicit CSV lineterminator='\n' to work around 3.1/3.2 csv module bug on Win32
      TST: use to_string instead to avoid futurewarning
      BUG: version_info is a tuple in python <= 2.5
      TST: skip scikits.statsmodels-depending tests if not installed
      TST: test failure on windows with Series.to_csv
      BUG: to_csv test failure on windows python3
      RLS: release notes
      Version 0.4.3

-----------------------------------------------------------------------

No new revisions were added by this update.

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/pandas.git