Merge commit 'v0.9.0-166-g7156920' into debian

* commit 'v0.9.0-166-g7156920': (159 commits)
  DOC: spacing fix
  BUG: catch any/all case that fails in NumPy > 1.6 with stricter casting rules
  DOC: fixed up release notes a little more
  TST: misc coverage and cleanup
  Additional test coverage for v0.9.1
  DOC: add note about converters now blocking type coersion in file parsers
  BUG: partial slicing bugs for PeriodIndex
  BUG: account for different fp exponent formatting in some pythons
  BUG: parse floats outside of PyFloat_FromString for python 2.5
  BUG: do not coerce types in parser if converters specified pandas-dev#2184
  TST: more misc test coverage
  TST: take out IPython check for timeseries plotting tests
  TST: fix panel/test_panel file modes. don't skip plotting tests if ipython imported
  TST: rogue foo
  TST: disable test_console_encode post unicode refactor
  TST: misc test coverage
  DOC: remove note on not overriding mpl registry
  BUG: put mpl unit registry back in for set_xlim
  BUG: ensure axes are Index objects in _arrays_to_mgr
  DOC: examples for release notes
  ...

Author: yarikoptic

CONTRIBUTING.md

@@ -0,0 +1 @@
+Please see [Developers](http://pandas.pydata.org/developers.html) page on the project website.

RELEASE.rst

@@ -22,6 +22,78 @@ Where to get it
 * Binary installers on PyPI: http://pypi.python.org/pypi/pandas
 * Documentation: http://pandas.pydata.org
 
+pandas 0.9.1
+============
+
+**Release date:** NOT YET RELEASED
+
+**New features**
+
+  - Can specify multiple sort orders in DataFrame/Series.sort/sort_index (#928)
+  - New `top` and `bottom` options for handling NAs in rank (#1508, #2159)
+  - Add `where` and `mask` functions to DataFrame (#2109, #2151)
+  - Add `at_time` and `between_time` functions to DataFrame (#2149)
+
+**API Changes**
+
+  - Upsampling period index "spans" intervals. Example: annual periods
+    upsampled to monthly will span all months in each year
+  - Period.end_time will yield timestamp at last nanosecond in the interval
+    (#2124, #2125, #1764)
+  - File parsers no longer coerce to float or bool for columns that have custom
+    converters specified (#2184)
+
+**Improvements to existing features**
+
+  - Time rule inference for week-of-month (e.g. WOM-2FRI) rules (#2140)
+  - Improve performance of datetime + business day offset with large number of
+    offset periods
+  - Improve HTML display of DataFrame objects with hierarchical columns
+  - Enable referencing of Excel columns by their column names (#1936)
+  - DataFrame.dot can accept ndarrays (#2042)
+  - Support negative periods in Panel.shift (#2164)
+  - Make .drop(...) work with non-unique indexes (#2101)
+  - Improve performance of Series/DataFrame.diff (re: #2087)
+  - Support unary ~ (__invert__) in DataFrame (#2110)
+
+**Bug fixes**
+
+  - Fix some duplicate-column DataFrame constructor issues (#2079)
+  - Fix bar plot color cycle issues (#2082)
+  - Fix off-center grid for stacked bar plots (#2157)
+  - Fix plotting bug if inferred frequency is offset with N > 1 (#2126)
+  - Implement comparisons on date offsets with fixed delta (#2078)
+  - Handle inf/-inf correctly in read_* parser functions (#2041)
+  - Fix matplotlib unicode interaction bug
+  - Make WLS r-squared match statsmodels 0.5.0 fixed value
+  - Fix zero-trimming DataFrame formatting bug
+  - Correctly compute/box datetime64 min/max values from Series.min/max (#2083)
+  - Fix unstacking edge case with unrepresented groups (#2100)
+  - Fix Series.str failures when using pipe pattern '|' (#2119)
+  - Fix pretty-printing of dict entries in Series, DataFrame (#2144)
+  - Cast other datetime64 values to nanoseconds in DataFrame ctor (#2095)
+  - Alias Timestamp.astimezone to tz_convert, so will yield Timestamp (#2060)
+  - Fix timedelta64 formatting from Series (#2165, #2146)
+  - Handle None values gracefully in dict passed to Panel constructor (#2075)
+  - Box datetime64 values as Timestamp objects in Series/DataFrame.iget (#2148)
+  - Fix Timestamp indexing bug in DatetimeIndex.insert (#2155)
+  - Use index name(s) (if any) in DataFrame.to_records (#2161)
+  - Don't lose index names in Panel.to_frame/DataFrame.to_panel (#2163)
+  - Work around length-0 boolean indexing NumPy bug (#2096)
+  - Fix partial integer indexing bug in DataFrame.xs (#2107)
+  - Fix variety of cut/qcut string-bin formatting bugs (#1978, #1979)
+  - Raise Exception when xs view not possible of MultiIndex'd DataFrame (#2117)
+  - Fix groupby(...).first() issue with datetime64 (#2133)
+  - Better floating point error robustness in some rolling_* functions (#2114)
+  - Fix ewma NA handling in the middle of Series (#2128)
+  - Fix numerical precision issues in diff with integer data (#2087)
+  - Fix bug in MultiIndex.__getitem__ with NA values (#2008)
+  - Fix DataFrame.from_records dict-arg bug when passing columns (#2179)
+  - Fix Series and DataFrame.diff for integer dtypes (#2087, #2174)
+  - Fix bug when taking intersection of DatetimeIndex with empty index (#2129)
+  - Pass through timezone information when calling DataFrame.align (#2127)
+
+
 pandas 0.9.0
 ============
 

bench/bench_merge.py

@@ -47,9 +47,9 @@ def get_test_data(ngroups=100, n=N):
 
 
 join_methods = ['inner', 'outer', 'left', 'right']
-results = DataFrame(index=join_methods, columns=[False])
+results = DataFrame(index=join_methods, columns=[False, True])
 niter = 10
-for sort in [False]:
+for sort in [False, True]:
     for join_method in join_methods:
         f = lambda: merge(left, right, how=join_method, sort=sort)
         gc.disable()
@@ -59,8 +59,8 @@ def get_test_data(ngroups=100, n=N):
         elapsed = (time.time() - start) / niter
         gc.enable()
         results[sort][join_method] = elapsed
-results.columns = ['pandas']
-# results.columns = ['dont_sort', 'sort']
+# results.columns = ['pandas']
+results.columns = ['dont_sort', 'sort']
 
 
 # R results
@@ -73,20 +73,21 @@ def get_test_data(ngroups=100, n=N):
 right      0.3102 0.0536     0.0376
 """), sep='\s+')
 
-all_results = results.join(r_results)
+presults = results[['dont_sort']].rename(columns={'dont_sort': 'pandas'})
+all_results = presults.join(r_results)
 
 all_results = all_results.div(all_results['pandas'], axis=0)
 
 all_results = all_results.ix[:, ['pandas', 'data.table', 'plyr', 'base::merge']]
 
 sort_results = DataFrame.from_items([('pandas', results['sort']),
-                                     ('R', r_results['sort'])])
+                                     ('R', r_results['base::merge'])])
 sort_results['Ratio'] = sort_results['R'] / sort_results['pandas']
 
 
 nosort_results = DataFrame.from_items([('pandas', results['dont_sort']),
-                                       ('R', r_results['dont_sort'])])
-nosort_results['Ratio'] = sort_results['R'] / sort_results['pandas']
+                                       ('R', r_results['base::merge'])])
+nosort_results['Ratio'] = nosort_results['R'] / nosort_results['pandas']
 
 # many to many
 
@@ -99,6 +100,6 @@ def get_test_data(ngroups=100, n=N):
 right      0.6425 0.0522     0.0428
 """), sep='\s+')
 
-all_results = results.join(r_results)
+all_results = presults.join(r_results)
 all_results = all_results.div(all_results['pandas'], axis=0)
 all_results = all_results.ix[:, ['pandas', 'data.table', 'plyr', 'base::merge']]

bench/bench_merge_sqlite.py

@@ -74,8 +74,8 @@
         conn.commit()
 
         sql_results[sort][join_method] = elapsed
-sql_results.columns = ['sqlite3'] # ['dont_sort', 'sort']
-sql_results.index = ['inner', 'outer', 'left']
+        sql_results.columns = ['sqlite3'] # ['dont_sort', 'sort']
+        sql_results.index = ['inner', 'outer', 'left']
 
         sql = """select *
         from left

doc/data/test.xls

@@ -110,15 +110,11 @@ Series input is of primary interest. Using these functions, you can use to
 either match on the *index* or *columns* via the **axis** keyword:
 
 .. ipython:: python
-   :suppress:
 
    d = {'one' : Series(randn(3), index=['a', 'b', 'c']),
         'two' : Series(randn(4), index=['a', 'b', 'c', 'd']),
         'three' : Series(randn(3), index=['b', 'c', 'd'])}
    df = DataFrame(d)
-
-.. ipython:: python
-
    df
    row = df.ix[1]
    column = df['two']

doc/source/basics.rst

@@ -26,7 +26,7 @@ objects. To get started, import numpy and load pandas into your namespace:
    randn = np.random.randn
    from pandas import *
 
-Here is a basic tenet to keep in mind: **data alignment is intrinsic**. Link
+Here is a basic tenet to keep in mind: **data alignment is intrinsic**. The link
 between labels and data will not be broken unless done so explicitly by you.
 
 We'll give a brief intro to the data structures, then consider all of the broad

doc/source/dsintro.rst

@@ -91,7 +91,7 @@ the data structures:
 
 There is an analogous ``set_value`` method which has the additional capability
 of enlarging an object. This method *always* returns a reference to the object
-it modified, which in the fast of enlargement, will be a **new object**:
+it modified, which in the case of enlargement, will be a **new object**:
 
 .. ipython:: python
 

doc/source/indexing.rst

@@ -164,7 +164,7 @@ You can also use a list of columns to create a hierarchical index:
 
 The ``dialect`` keyword gives greater flexibility in specifying the file format.
 By default it uses the Excel dialect but you can specify either the dialect name
-or a :class:``python:csv.Dialect`` instance.
+or a :class:`python:csv.Dialect` instance.
 
 .. ipython:: python
    :suppress:
@@ -286,6 +286,13 @@ data columns:
                  index_col=0) #index is the nominal column
    df
 
+**Note**: When passing a dict as the `parse_dates` argument, the order of
+the columns prepended is not guaranteed, because `dict` objects do not impose
+an ordering on their keys. On Python 2.7+ you may use `collections.OrderedDict`
+instead of a regular `dict` if this matters to you. Because of this, when using a
+dict for 'parse_dates' in conjunction with the `index_col` argument, it's best to
+specify `index_col` as a column label rather then as an index on the resulting frame.
+
 Date Parsing Functions
 ~~~~~~~~~~~~~~~~~~~~~~
 Finally, the parser allows you can specify a custom ``date_parser`` function to
@@ -647,7 +654,7 @@ function takes a number of arguments. Only the first is required.
     (default), and `header` and `index` are True, then the index names are
     used. (A sequence should be given if the DataFrame uses MultiIndex).
   - ``mode`` : Python write mode, default 'w'
-  - ``sep`` : Field delimiter for the output file (default "'")
+  - ``sep`` : Field delimiter for the output file (default ",")
   - ``encoding``: a string representing the encoding to use if the contents are
     non-ascii, for python versions prior to 3
 

doc/source/io.rst

@@ -414,11 +414,6 @@ either the left or right tables, the values in the joined table will be
     ``outer``, ``FULL OUTER JOIN``, Use union of keys from both frames
     ``inner``, ``INNER JOIN``, Use intersection of keys from both frames
 
-Note that if using the index from either the left or right DataFrame (or both)
-using the ``left_index`` / ``right_index`` options, the join operation is no
-longer a many-to-many join by construction, as the index values are necessarily
-unique. There will be some examples of this below.
-
 .. _merging.join.index:
 
 Joining on index

doc/source/merging.rst

@@ -45,7 +45,7 @@ API changes
 
 - Creating a Series from another Series, passing an index, will cause reindexing
   to happen inside rather than treating the Series like an ndarray. Technically
-  improper usages like ``Series(df[col1], index=df[col2])11 that worked before
+  improper usages like ``Series(df[col1], index=df[col2])`` that worked before
   "by accident" (this was never intended) will lead to all NA Series in some
   cases. To be perfectly clear: