[pandas] annotated tag v0.8.0b2 created (now 98bb53d)
Andreas Tille
tille at debian.org
Wed Dec 28 15:20:28 UTC 2016
This is an automated email from the git hooks/post-receive script.
tille pushed a change to annotated tag v0.8.0b2
in repository pandas.
at 98bb53d (tag)
tagging fde270b20861fbf36d3063d504fb299d0b58695b (commit)
replaces v0.8.0b1
tagged by Wes McKinney
on Tue Jun 12 12:41:46 2012 -0400
- Log -----------------------------------------------------------------
Version 0.8.0 beta 2
Adam Klein (5):
BUG: plotting breaks if inferred_freq has multiple #1357
DOC: removes temporary csv after examples
ENH: add parameter to nancorr to suppress div by zero
ENH: fix nancorr further
ENH: continuing #1349, optimize DataFrame.cov for missing data
Chang She (43):
ENH: better error msg for fillna() with invalid method
DOC: cleaned up parser doc string to stop sphinx from complaining
BUG: fixed doc bug that caused latex build to fail
BUG: PeriodIndex.map tries to get super(DatetimIndex, self)
BUG: set_xlim for time series plots #1339
ENH: KdePlot with DataFrame #1342. TST: frame kde and kde with logy #1341
TST: additional tests for parsers and minor code cleanup
BUG: parser with multiple date col and multiple index col #1344
DOC: new parser functionality #1347
BLD: turning isreleased to False
DOC: string args to how in resample #1355
ENH: add dialect keyword to parsers to expose quoting specifications #1363
BUG: sparse reduction bug #1375
ENH: alias density to kde in plot methods #1343
BUG: raise exception on NaT in DatetimeIndex.astype('O') #1340
push exception down to asobject
BUG: replacing isinstance(value, (int, long)) with com.is_integer
BUG: replacing isinstance with com.is_integer in frequencies.py
BUG: typo in frequencies.py
DOC: links for some of whatsnew features
BUG: typo in test_multilevel.py
DOC: links from whatsnew
BUG: fixed Series.replace with later replacements clobbering earlier ones. Still need DataFrame fix
DOC: additional linking from whatsnew and added examples for various features
BUG: DataFrame.replace later values clobber earlier ones
DOC: some more examples and links from whatsnew
ENH: passing kwds to rolling func
BUG: Period.strftime no return
TST: test Period.strftime
BUG: buffer access out of bounds with roll_generic on empty
ENH: bad line number in error msg for unclosed quotes in parser
BUG: finder place year markers on Q2 and Feb instead of Jan
BUG: DatetimeIndex.to_datetime loses freq and tz
BUG: to_csv duplicate index values #1373
BUG: isin method buggy on SparseSeries #1412
ENH: set_xlim using str for irregular index #1427
ENH: check the periods argument is a number #1438
ENH: DataFrame any all #1416
TST: mixed case
ENH: make any/all conform to sum/mean interface. Fixed bug in copy keyword in BlockManager.get_numeric_data #1416
BUG: fix compat errors for yahoo data reader
BUG: import BytesIO for py3compat
BUG: bytes_to_str for read_csv
Kieran O'Mahony (8):
BLD: ujson remove dependence on npymath lib
BLD: mingw32 fix for ujson
ujson remove troublesome uint32_t def affecting mingw build
ujson check for malloc success
BUG: ujson don't compare pointers outside their allocated blocks
BUG: ujson memory alignment fixes
BLD: ujson fix compiler warnings
BUG: ujson fix sniffing of dtype when decoding
Lorenzo Bolla (1):
Close cursor before committing the connection.
Marc Abramowitz (5):
Add support for tox (http://tox.testrun.org/)
Add .travis.yml for Travis CI (http://travis-ci.org/)
Add dependency on pytz
Reenable py31 and py32 in .travis.yml
Make tox work across versions of Python from 2.5 to 3.2
Paddy Mullen (2):
fixes #826, adds a a test that asserts yahoo is minimally functional as a datasource
added pause and retry_count for yahoo
Vytautas Jancauskas (8):
Added andrews_curves to plotting.py
Added tests for andrews curves
Added documentation
Added iris data to documentation too
Fixed for samples with odd number of attributes and label handling when they are not strings
Added lag plot function
Added lag plot test
Added lag plot documentation
Wes McKinney (97):
BUG: respect logy argument in KdePlot, close #1341
DOC: release notes and what's new
DOC: release notes, what's new
DOC: what's new
DOC: another pass at release notes
ENH: Cython nancorr speeds up DataFrame.corr with method='pearson' by > 100x
DOC: release notes to close #1349
ENH: handle Series input in DataFrame.corrwith, close #1367
ENH: implement cut function for turning continuous variable into categorical variable, close #415
ENH: shell of quantile cut function qcut, #1378
BUG: cast other datetime64 units to nanoseconds for storage in DataFrame BlockManager, close #1379
DOC: release notes
DOC: release notes per #1325
BUG: honor column selection in GroupBy.transform on DataFrame, close #1365
ENH: exclude nuisance columns in GroupBy.transform, close #1364
ENH: support functions passed as strings in groupby.transform just like aggregate, close #1362
BLD: can't assign to PyArray_BYTES on all platforms #1380
TST: fix NumPy 1.7 test failures per #1361, skip tests depending on scipy.stats
BUG: use style parameter in tsplots, close #1338
ENH: enable df[bool_df] += value for compatible shapes, close #1366
BLD: move iris.data to iris.csv
BUG: infer unicode separate from string for index serialization in HDFStore re: #1361
TST: catch malformed scipy.stats
ENH: use kvec.h instead of PyList for accumulating uniques in hashtable.pyx, return arrays from pandas.unique, close #1353
TST: 32-bit integer checking issues
BUG: fix multiindex partial slicing bug in .ix close #1352
BUG: hacks to support duplicate DataFrame columns in BlockManager, irow/icol with duplicates, ix[...] with dups too. close #1374
TST: check tzinfos in boxed values, close #1345
TST: alter test to include DST transition, remove asobject_tz_box #1345
DOC: Timestamp scalar time zone handling and conversions, close #1356
ENH: handle mixed multiple function spec in SeriesGroupBy.aggregate, close #1359
ENH: add to_pydatetime methods on Timestamp, DatetimeIndex, close #1372
DOC: 0.7.3 -> 0.8.0 migration guide in what's new, close #1272
BLD: fix mingw32 build
DOC: start migration guide for skts users #1273
BLD: break period code into separate extension for sanity
BLD: add plib.pyx
BUG: port skts_strftime to Cython, hopefully fix mingw32 issues #1360
TST: float formatting differences in 2.6/2.5
DOC: release notes
DOC: barebones but reasonably complete scikits.timeseries -> pandas migration guide. close #1273
BUG: cast other datetime64 units to nanos from Index constructor, properly handle Index.astype(datetime64-unit), close #1384
BUG: Cython time zone localization, numerous bug fixes and Timestamp behavior improvements. close #1389
ENH: tighter Timestamp repr, close #1391
ENH: implement qcut for quantile cuts, fix 32-bit build close #1378
ENH: add value_counts as top level function in API, close #1392
DOC: couple fixes
API: don't check for duplicate by default in Series.append/DataFrame.append, close #1394
BLD: fix type declaration in period.h, close #1398
BLD: fix plib build, depend on ultrajson.h
BUG: fix broken UTC localization, add tz_localize method to Timestamp, close #1397
BUG: raise exception in comparisons between tz-naive and tz-aware Timestamp objects, close #1404
ENH: split out time zone localize logic from conversion, add tz_localize methods to Series, DataFrame. adjust unit tests. close #1403
REF: refactor Factor class
ENH: refactoring to support ordered factors, cut/qcut return factors. #292
ENH: support for ordered factors in groupby, close #292
DOC: release notes
BUG: cast Series to array in quantile to avoid segfault in Factor from qcut
ENH: boundscheck take_1d
ENH: attach name to Factor
BUG: handle smallest value in qcut
TST: fix broken cut test
ENH: use index name as xlabel/ylabel in plots, close #1415
BUG: concatenate Series, don't create DataFrame in GroupBy.apply when indexes not all the same
BUG: fix asfreq bug on PeriodIndex
BUG: preserve name in PeriodIndex.to_timestamp
BUG: fix a number of negative ordinal bugs, check for out-of-range quarters
BUG: fix Series .ix bug and test failure described in #1396
BUG: fix numpy 1.6 erroneous type-casting bug causing NumPy 1.7 issues, close #1396
ENH: implement comparison methods on Factor, close #1405
BUG: pass on limit argument in upsampling, close #1424
BUG: fix DatetimeIndex.groupby bug close #1430
TST: change assertEquals to assertAlmostEqual
BUG: override DatetimeIndex.tolist so it returns a list of Timestamp objects, close #1437
BUG: assignment of rows via ix in mixed-dtype DataFrame. close #1432
BUG: properly detect tz-aware/tz-naive joins and raise exceptions. close #1426
BUG: ensure arithmetic ops between equal, but different time zone DatetimeIndex always yield UTC, close #1422
ENH: better error message with non-unique columns in DataFrame.from_records, close #1298
ENH: commit autocorrelation_plot by hand
ENH: commit autocorrelation_plot stuff by hand #1425
ENH: add convert_dtype option to Series.apply to be able to leave result as dtype=object, close #1414
BUG: reset index mapping after each function call in SeriesGrouper, close #1423
REF: remove offset names from pandas namespace
BUG: implement multiple DataFrame.join / merge on non-unique indexes by multiple merges, close #1421
DOC: release notes
BUG: fix MultiIndex bugs described in #1401
ENH: allow renaming of index levels when concatenating, close #1419
ENH: clearer out of bounds error message in cut/qcut, close #1409
ENH: implement passed quantile array to qcut and document that plus factors, close #1407
BUG: always raise exception when concat keys aren't found in passed levels, close #1406
DOC: refresh time zone docs close #1447
TST: eliminate copies in datetime64 serialization; don't copy data in DatetimeIndex, close #1320
REF: factor out ujson extension into pandasjson for now
TST: oops, delete stray line
TST: test coverage
ENH: convert datetime.datetime ourselves, 15x speedup
RLS: 0.8.0 beta 2
Wouter Overmeire (2):
DOC: fix typos
DOC: Indicate that to_csv and from_csv methods can handle string file path and file-like objects for io.
thuske (1):
ujson fix compile error using clang
timmie (2):
DOC: install updated, faq updated conf.py with added sphinx extensions
add pandas/docs/build/html/index.html to gitignore
-----------------------------------------------------------------------
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
More information about the debian-science-commits
mailing list