From 91afd548fa1e33ea1fc73b49f9766921547a2e2b Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 12:45:44 -0400
Subject: [PATCH 01/42] BUG: minimal out of bounds checking on future/past
 timestamps, gotcha docs, pre-1900 repr

---
 doc/source/gotchas.rst                  | 24 +++++++++++++++
 pandas/src/datetime.pyx                 | 40 ++++++++++++++++++++++---
 pandas/tseries/tests/test_timeseries.py | 15 +++++++++-
 3 files changed, 74 insertions(+), 5 deletions(-)

diff --git a/doc/source/gotchas.rst b/doc/source/gotchas.rst
index 57d6471a8fe7b..f4c0eae4cfca0 100644
--- a/doc/source/gotchas.rst
+++ b/doc/source/gotchas.rst
@@ -217,3 +217,27 @@ passed in the index, thus finding the integers ``0`` and ``1``. While it would
 be possible to insert some logic to check whether a passed sequence is all
 contained in the index, that logic would exact a very high cost in large data
 sets.
+
+Timestamp limitations
+---------------------
+
+Minimum and maximum timestamps
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+Since pandas represents timestamps in nanosecond resolution, the timespan that
+can be represented using a 64-bit integer is limited to approximately 584 years:
+
+.. ipython:: python
+
+   begin = Timestamp(-9223285636854775809L)
+   begin
+   end = Timestamp(np.iinfo(np.int64).max)
+   end
+
+If you need to represent time series data outside the nanosecond timespan, use
+PeriodIndex:
+
+.. ipython:: python
+
+   span = period_range('1215-01-01', '1381-01-01', freq='D')
+   span
diff --git a/pandas/src/datetime.pyx b/pandas/src/datetime.pyx
index ec50275ccc4a6..38adf46285e0d 100644
--- a/pandas/src/datetime.pyx
+++ b/pandas/src/datetime.pyx
@@ -65,6 +65,7 @@ def ints_to_pydatetime(ndarray[int64_t] arr, tz=None):
     return result
 
 
+
 # Python front end to C extension type _Timestamp
 # This serves as the box for datetime64
 class Timestamp(_Timestamp):
@@ -101,10 +102,27 @@ class Timestamp(_Timestamp):
         return ts_base
 
     def __repr__(self):
-        result = self.strftime('<Timestamp: %Y-%m-%d %H:%M:%S%z')
-        if self.tzinfo:
-            result += self.strftime(' %%Z, tz=%s' % self.tzinfo.zone)
-        return result + '>'
+        result = '%d-%.2d-%.2d %.2d:%.2d:%.2d' % (self.year, self.month,
+                                                  self.day, self.hour,
+                                                  self.minute, self.second)
+
+        if self.nanosecond != 0:
+            nanos = self.nanosecond + 1000 * self.microsecond
+            result += '.%.9d' % nanos
+        elif self.microsecond != 0:
+            result += '.%.6d' % self.microsecond
+
+        try:
+            result += self.strftime('%z')
+            if self.tzinfo:
+                result += self.strftime(' %%Z, tz=%s' % self.tzinfo.zone)
+        except ValueError:
+            year2000 = self.replace(year=2000)
+            result += year2000.strftime('%z')
+            if self.tzinfo:
+                result += year2000.strftime(' %%Z, tz=%s' % self.tzinfo.zone)
+
+        return '<Timestamp: %s>' % result
 
     @property
     def tz(self):
@@ -507,6 +525,7 @@ cpdef convert_to_tsobject(object ts, object tz=None):
             obj.tzinfo = ts.tzinfo
             if obj.tzinfo is not None:
                 obj.value -= _delta_to_nanoseconds(obj.tzinfo._utcoffset)
+        _check_dts_bounds(obj.value, &obj.dts)
         return obj
     elif PyDate_Check(ts):
         obj.value  = _date_to_datetime64(ts, &obj.dts)
@@ -514,6 +533,9 @@ cpdef convert_to_tsobject(object ts, object tz=None):
         raise ValueError("Could not construct Timestamp from argument %s" %
                          type(ts))
 
+    if obj.value != NPY_NAT:
+        _check_dts_bounds(obj.value, &obj.dts)
+
     if tz is not None:
         if tz is pytz.utc:
             obj.tzinfo = tz
@@ -530,6 +552,16 @@ cpdef convert_to_tsobject(object ts, object tz=None):
 
     return obj
 
+cdef int64_t _NS_LOWER_BOUND = -9223285636854775809LL
+cdef int64_t _NS_UPPER_BOUND = -9223372036854775807LL
+
+cdef inline _check_dts_bounds(int64_t value, pandas_datetimestruct *dts):
+    cdef pandas_datetimestruct dts2
+    if dts.year <= 1677 or dts.year >= 2262:
+        pandas_datetime_to_datetimestruct(value, PANDAS_FR_ns, &dts2)
+        if dts2.year != dts.year:
+            raise ValueError('Out of bounds timestamp in year: %s' % dts.year)
+
 # elif isinstance(ts, _Timestamp):
 #     tmp = ts
 #     obj.value = (<_Timestamp> ts).value
diff --git a/pandas/tseries/tests/test_timeseries.py b/pandas/tseries/tests/test_timeseries.py
index 06fc71c2cc65f..9c6542d4c2edc 100644
--- a/pandas/tseries/tests/test_timeseries.py
+++ b/pandas/tseries/tests/test_timeseries.py
@@ -898,6 +898,20 @@ def test_timestamp_fields(self):
         self.assertEqual(idx.freq, Timestamp(idx[0], idx.freq).freq)
         self.assertEqual(idx.freqstr, Timestamp(idx[0], idx.freq).freqstr)
 
+    def test_timestamp_date_out_of_range(self):
+        self.assertRaises(ValueError, Timestamp, '1676-01-01')
+        self.assertRaises(ValueError, Timestamp, '2263-01-01')
+
+    def test_timestamp_repr(self):
+        # pre-1900
+        stamp = Timestamp('1850-01-01', tz='US/Eastern')
+        repr(stamp)
+
+        iso8601 = '1850-01-01 01:23:45.012345'
+        stamp = Timestamp(iso8601, tz='US/Eastern')
+        result = repr(stamp)
+        self.assert_(iso8601 in result)
+
     def test_datetimeindex_integers_shift(self):
         rng = date_range('1/1/2000', periods=20)
 
@@ -918,7 +932,6 @@ def test_astype_object(self):
 
         self.assert_(np.array_equal(casted, exp_values))
 
-
     def test_catch_infinite_loop(self):
         offset = datetools.DateOffset(minute=5)
         # blow up, don't loop forever

From 5fbffe4d1cd8b422119eb2b86c8a3f58e940f18c Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Thu, 21 Jun 2012 12:36:35 -0400
Subject: [PATCH 02/42] BUG: sub-second plotting was not working

---
 pandas/tools/plotting.py              | 19 ++++++++++++-------
 pandas/tseries/frequencies.py         | 24 ++++++++++++++++++++++--
 pandas/tseries/plotting.py            |  1 -
 pandas/tseries/tests/test_plotting.py |  9 +++++++++
 4 files changed, 43 insertions(+), 10 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index adf5abdebe0df..fa351207f0059 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -10,7 +10,7 @@
 from pandas.core.series import Series
 from pandas.tseries.index import DatetimeIndex
 from pandas.tseries.period import PeriodIndex
-from pandas.tseries.frequencies import get_period_alias
+from pandas.tseries.frequencies import get_period_alias, get_base_alias
 from pandas.tseries.offsets import DateOffset
 import pandas.tseries.tools as datetools
 
@@ -590,14 +590,19 @@ def __init__(self, data, **kwargs):
     def has_ts_index(self):
         from pandas.core.frame import DataFrame
         if isinstance(self.data, (Series, DataFrame)):
-            if isinstance(self.data.index, (DatetimeIndex, PeriodIndex)):
-                has_freq = (hasattr(self.data.index, 'freq') and
-                            self.data.index.freq is not None)
-                has_inferred = (hasattr(self.data.index, 'inferred_freq') and
-                                self.data.index.inferred_freq is not None)
-                return has_freq or has_inferred
+            freq = (getattr(self.data.index, 'freq', None)
+                    or getattr(self.data.index, 'inferred_freq', None))
+            return (freq is not None) and  self._has_dynamic_index_freq(freq)
         return False
 
+    def _has_dynamic_index_freq(self, freq):
+        if isinstance(freq, DateOffset):
+            freq = freq.rule_code
+        else:
+            freq = get_base_alias(freq)
+        freq = get_period_alias(freq)
+        return freq is not None
+
     def _make_plot(self):
         # this is slightly deceptive
         if self.use_index and self.has_ts_index:
diff --git a/pandas/tseries/frequencies.py b/pandas/tseries/frequencies.py
index 8b87f8ccbf820..433e88e2d7ecb 100644
--- a/pandas/tseries/frequencies.py
+++ b/pandas/tseries/frequencies.py
@@ -246,7 +246,16 @@ def _get_freq_str(base, mult=1):
     'BA' : 'A',
     'AS' : 'A',
     'BAS' : 'A',
-    'MS' : 'M'
+    'MS' : 'M',
+    'D' : 'D',
+    'B' : 'B',
+    'T' : 'T',
+    'S' : 'S',
+    'H' : 'H',
+    'Q' : 'Q',
+    'A' : 'A',
+    'W' : 'W',
+    'M' : 'M'
 }
 
 need_suffix = ['QS', 'BQ', 'BQS', 'AS', 'BA', 'BAS']
@@ -257,9 +266,20 @@ def _get_freq_str(base, mult=1):
         _offset_to_period_map['%s-%s' % (prefix, m)] = \
             _offset_to_period_map[prefix]
 
+months = ['JAN', 'FEB', 'MAR', 'APR', 'MAY', 'JUN', 'JUL', 'AUG', 'SEP',
+          'OCT', 'NOV', 'DEC']
+for prefix in ['A', 'Q']:
+    for m in months:
+        alias = '%s-%s' % (prefix, m)
+        _offset_to_period_map[alias] = alias
+
+_days = ['MON', 'TUE', 'WED', 'THU', 'FRI', 'SAT', 'SUN']
+for d in _days:
+    _offset_to_period_map['W-%s' % d] = 'W-%s' % d
+
 def get_period_alias(offset_str):
     """ alias to closest period strings BQ->Q etc"""
-    return _offset_to_period_map.get(offset_str, offset_str)
+    return _offset_to_period_map.get(offset_str, None)
 
 _rule_aliases = {
     # Legacy rules that will continue to map to their original values
diff --git a/pandas/tseries/plotting.py b/pandas/tseries/plotting.py
index 8ee78f4df652e..79823e8c12dca 100644
--- a/pandas/tseries/plotting.py
+++ b/pandas/tseries/plotting.py
@@ -136,7 +136,6 @@ def _get_default_annual_spacing(nyears):
         (min_spacing, maj_spacing) = (factor * 20, factor * 100)
     return (min_spacing, maj_spacing)
 
-
 def period_break(dates, period):
     """
     Returns the indices where the given period changes.
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index 84dadf03bf9eb..81f972debe6c2 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -68,6 +68,14 @@ def test_tsplot(self):
         ax = ts.plot(style='k')
         self.assert_((0., 0., 0.) == ax.get_lines()[0].get_color())
 
+    @slow
+    def test_high_freq(self):
+        freaks = ['ms', 'us']
+        for freq in freaks:
+            rng = date_range('1/1/2012', periods=100000, freq=freq)
+            ser = Series(np.random.randn(len(rng)), rng)
+            _check_plot_works(ser.plot)
+
     def test_get_datevalue(self):
         from pandas.tseries.plotting import get_datevalue
         self.assert_(get_datevalue(None, 'D') is None)
@@ -268,6 +276,7 @@ def test_finder_monthly(self):
     @slow
     def test_finder_annual(self):
         import matplotlib.pyplot as plt
+        plt.close('all')
         xp = [1987, 1988, 1990, 1990, 1995, 2020, 2070, 2170]
         for i, nyears in enumerate([5, 10, 19, 49, 99, 199, 599, 1001]):
             rng = period_range('1987', periods=nyears, freq='A')

From 32d73012285d1228665b05f01059bd82aea04ceb Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 13:02:51 -0400
Subject: [PATCH 03/42] BUG: nanosecond out-of-bounds checking in array
 conversions. close #1475

---
 pandas/src/datetime.pyx                 | 11 ++++++++++-
 pandas/tseries/tests/test_timeseries.py |  4 ++++
 2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/pandas/src/datetime.pyx b/pandas/src/datetime.pyx
index 38adf46285e0d..17f35a73501aa 100644
--- a/pandas/src/datetime.pyx
+++ b/pandas/src/datetime.pyx
@@ -560,7 +560,11 @@ cdef inline _check_dts_bounds(int64_t value, pandas_datetimestruct *dts):
     if dts.year <= 1677 or dts.year >= 2262:
         pandas_datetime_to_datetimestruct(value, PANDAS_FR_ns, &dts2)
         if dts2.year != dts.year:
-            raise ValueError('Out of bounds timestamp in year: %s' % dts.year)
+            fmt = '%d-%.2d-%.2d %.2d:%.2d:%.2d' % (dts.year, dts.month,
+                                                   dts.day, dts.hour,
+                                                   dts.min, dts.sec)
+
+            raise ValueError('Out of bounds nanosecond timestamp: %s' % fmt)
 
 # elif isinstance(ts, _Timestamp):
 #     tmp = ts
@@ -645,8 +649,10 @@ def array_to_datetime(ndarray[object] values, raise_=False, dayfirst=False):
                 iresult[i] = iNaT
             elif PyDateTime_Check(val):
                 iresult[i] = _pydatetime_to_dts(val, &dts)
+                _check_dts_bounds(iresult[i], &dts)
             elif PyDate_Check(val):
                 iresult[i] = _date_to_datetime64(val, &dts)
+                _check_dts_bounds(iresult[i], &dts)
             elif util.is_datetime64_object(val):
                 iresult[i] = _get_datetime64_nanos(val)
             elif util.is_integer_object(val):
@@ -659,6 +665,9 @@ def array_to_datetime(ndarray[object] values, raise_=False, dayfirst=False):
                     result[i] = parse(val, dayfirst=dayfirst)
                 except Exception:
                     raise TypeError
+                pandas_datetime_to_datetimestruct(iresult[i], PANDAS_FR_ns,
+                                                  &dts)
+                _check_dts_bounds(iresult[i], &dts)
         return result
     except TypeError:
         oresult = np.empty(n, dtype=object)
diff --git a/pandas/tseries/tests/test_timeseries.py b/pandas/tseries/tests/test_timeseries.py
index 9c6542d4c2edc..f04f0a5b8259c 100644
--- a/pandas/tseries/tests/test_timeseries.py
+++ b/pandas/tseries/tests/test_timeseries.py
@@ -902,6 +902,10 @@ def test_timestamp_date_out_of_range(self):
         self.assertRaises(ValueError, Timestamp, '1676-01-01')
         self.assertRaises(ValueError, Timestamp, '2263-01-01')
 
+        # 1475
+        self.assertRaises(ValueError, DatetimeIndex, ['1400-01-01'])
+        self.assertRaises(ValueError, DatetimeIndex, [datetime(1400, 1, 1)])
+
     def test_timestamp_repr(self):
         # pre-1900
         stamp = Timestamp('1850-01-01', tz='US/Eastern')

From 1372d0f09f8b5b625b20587d3608298a1fd44b98 Mon Sep 17 00:00:00 2001
From: Wouter Overmeire <lodagro@gmail.com>
Date: Thu, 21 Jun 2012 20:36:37 +0200
Subject: [PATCH 04/42] ENH: Add sharex and sharey to DataFrame.hist()

---
 pandas/tests/test_graphics.py | 3 +++
 pandas/tools/plotting.py      | 8 ++++++--
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/pandas/tests/test_graphics.py b/pandas/tests/test_graphics.py
index 65fc503f2eb7f..57bc17c5c33b6 100644
--- a/pandas/tests/test_graphics.py
+++ b/pandas/tests/test_graphics.py
@@ -223,6 +223,9 @@ def test_hist(self):
         df = DataFrame(np.random.randn(100, 6))
         _check_plot_works(df.hist)
 
+        #make sure sharex, sharey is handled
+        _check_plot_works(df.hist, sharex=True, sharey=True)
+
         #make sure kwargs are handled
         ser = df[0]
         xf, yf = 20, 20
diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index adf5abdebe0df..a669b15596fe6 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -1095,7 +1095,8 @@ def plot_group(group, ax):
 
 
 def hist_frame(data, grid=True, xlabelsize=None, xrot=None,
-               ylabelsize=None, yrot=None, ax=None, **kwds):
+               ylabelsize=None, yrot=None, ax=None,
+               sharex=False, sharey=False, **kwds):
     """
     Draw Histogram the DataFrame's series using matplotlib / pylab.
 
@@ -1112,6 +1113,8 @@ def hist_frame(data, grid=True, xlabelsize=None, xrot=None,
     yrot : float, default None
         rotation of y axis labels
     ax : matplotlib axes object, default None
+    sharex : bool, if True, the X axis will be shared amongst all subplots.
+    sharey : bool, if True, the Y axis will be shared amongst all subplots.
     kwds : other plotting keyword arguments
         To be passed to hist function
     """
@@ -1123,7 +1126,8 @@ def hist_frame(data, grid=True, xlabelsize=None, xrot=None,
             rows += 1
         else:
             cols += 1
-    _, axes = _subplots(nrows=rows, ncols=cols, ax=ax, squeeze=False)
+    _, axes = _subplots(nrows=rows, ncols=cols, ax=ax, squeeze=False,
+                        sharex=sharex, sharey=sharey)
 
     for i, col in enumerate(com._try_sort(data.columns)):
         ax = axes[i / cols][i % cols]

From 3815815639753b82afbd192cd1f01ab3f7a735bd Mon Sep 17 00:00:00 2001
From: Wouter Overmeire <lodagro@gmail.com>
Date: Thu, 21 Jun 2012 20:38:36 +0200
Subject: [PATCH 05/42] DOC: Fix typo.

---
 pandas/tools/plotting.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index a669b15596fe6..7ca432e89ecd5 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -1316,7 +1316,7 @@ def _subplots(nrows=1, ncols=1, sharex=False, sharey=False, squeeze=True,
     sharex : bool
       If True, the X axis will be shared amongst all subplots.
 
-    sharex : bool
+    sharey : bool
       If True, the Y axis will be shared amongst all subplots.
 
     squeeze : bool

From 214c24b44dba241a4fce0fc988cedfde135bde05 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 15:52:40 -0400
Subject: [PATCH 06/42] DOC: finish updating release notes

---
 RELEASE.rst                    | 15 ++++++++++++++-
 doc/source/whatsnew/v0.8.0.txt |  6 ++++--
 2 files changed, 18 insertions(+), 3 deletions(-)

diff --git a/RELEASE.rst b/RELEASE.rst
index 7fe8b017cf71b..d187b211e3f2f 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -86,7 +86,9 @@ pandas 0.8.0
   - Add lag plot (#1440)
   - Add autocorrelation_plot (#1425)
   - Add support for tox and Travis CI (#1382)
-  - Add support for ordered factors and use in GroupBy (#292)
+  - Add support for Categorical use in GroupBy (#292)
+  - Add ``any`` and ``all`` methods to DataFrame (#1416)
+  - Add ``secondary_y`` option to Series.plot
 
 **Improvements to existing features**
 
@@ -124,9 +126,16 @@ pandas 0.8.0
   - Add ``convert_dtype`` option to Series.apply to be able to leave data as
     dtype=object (#1414)
   - Can specify all index level names in concat (#1419)
+  - Add ``dialect`` keyword to parsers for quoting conventions (#1363)
+  - Enable DataFrame[bool_DataFrame] += value (#1366)
+  - Add ``retries`` argument to ``get_data_yahoo`` to try to prevent Yahoo! API
+    404s (#826)
+  - Improve performance of reshaping by using O(N) categorical sorting
 
 **API Changes**
 
+  - Rename Factor to Categorical and add improvements. Numerous Categorical bug
+    fixes
   - Frequency name overhaul, WEEKDAY/EOM and rules with @
     deprecated. get_legacy_offset_name backwards compatibility function added
   - Raise ValueError in DataFrame.__nonzero__, so "if df" no longer works
@@ -190,6 +199,10 @@ pandas 0.8.0
   - Fix outer/inner DataFrame.join with non-unique indexes (#1421)
   - Fix MultiIndex groupby bugs with empty lower levels (#1401)
   - Calling fillna with a Series will have same behavior as with dict (#1486)
+  - SparseSeries reduction bug (#1375)
+  - Fix unicode serialization issue in HDFStore (#1361)
+  - Pass keywords to pyplot.boxplot in DataFrame.boxplot (#1493)
+  - Bug fixes in MonthBegin (#1483)
 
 pandas 0.7.3
 ============
diff --git a/doc/source/whatsnew/v0.8.0.txt b/doc/source/whatsnew/v0.8.0.txt
index 5ea0be0ddd971..cbf094752d4a9 100644
--- a/doc/source/whatsnew/v0.8.0.txt
+++ b/doc/source/whatsnew/v0.8.0.txt
@@ -111,8 +111,10 @@ index duplication in many-to-many joins)
 Other new features
 ~~~~~~~~~~~~~~~~~~
 
-- New :ref:`cut <reshaping.tile.cut>` function (like R's cut function) for
-  computing a categorical variable from a continuous variable by binning values
+- New :ref:`cut <reshaping.tile.cut>` and ``qcut`` functions (like R's cut
+  function) for computing a categorical variable from a continuous variable by
+  binning values either into value-based (``cut``) or quantile-based (``qcut``)
+  bins
 - Add :ref:`limit <missing_data.fillna.limit>` argument to fillna/reindex
 - More flexible multiple function application in GroupBy, and can pass list
   (name, function) tuples to get result in particular order with given names

From c0151bdcd91282080277974939f18f6efd222d43 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Thu, 21 Jun 2012 16:45:39 -0400
Subject: [PATCH 07/42] DOC: example for plotting on secondary y-axis

---
 doc/source/visualization.rst | 15 +++++++++++++++
 1 file changed, 15 insertions(+)

diff --git a/doc/source/visualization.rst b/doc/source/visualization.rst
index 34bd803516468..dcf0f3b87343f 100644
--- a/doc/source/visualization.rst
+++ b/doc/source/visualization.rst
@@ -91,6 +91,20 @@ You may pass ``logy`` to get a log-scale Y axis.
    @savefig series_plot_logy.png width=4.5in
    ts.plot(logy=True)
 
+Plotting on a Secondary Y-axis
+~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
+
+To plot data on a secondary y-axis, use the ``secondary_y`` keyword:
+
+.. ipython:: python
+
+   plt.figure()
+
+   df.A.plot()
+
+   @savefig series_plot_secondary_y.png width=4.5in
+   df.B.plot(secondary_y=True, style='g')
+
 
 Targeting different subplots
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -107,6 +121,7 @@ You can pass an ``ax`` argument to ``Series.plot`` to plot on a particular axis:
    @savefig series_plot_multi.png width=4.5in
    df['D'].plot(ax=axes[1,1]); axes[1,1].set_title('D')
 
+
 Other plotting features
 -----------------------
 

From 39687071b8627cf9fb86be39bf6bdcf5bd0c9a90 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 18:26:07 -0400
Subject: [PATCH 08/42] RLS: 0.8.0 what's new docs

---
 RELEASE.rst                          |  2 +-
 doc/source/{whatsnew => }/v0.4.x.txt |  0
 doc/source/{whatsnew => }/v0.5.0.txt |  0
 doc/source/{whatsnew => }/v0.6.0.txt |  0
 doc/source/{whatsnew => }/v0.6.1.txt |  0
 doc/source/{whatsnew => }/v0.7.0.txt |  0
 doc/source/{whatsnew => }/v0.7.1.txt |  0
 doc/source/{whatsnew => }/v0.7.2.txt |  0
 doc/source/{whatsnew => }/v0.7.3.txt |  0
 doc/source/{whatsnew => }/v0.8.0.txt | 60 +++++++++++++++++++++++-----
 doc/source/visualization.rst         |  1 +
 doc/source/whatsnew.rst              | 18 ++++-----
 pandas/tseries/frequencies.py        | 41 +++++++++----------
 13 files changed, 79 insertions(+), 43 deletions(-)
 rename doc/source/{whatsnew => }/v0.4.x.txt (100%)
 rename doc/source/{whatsnew => }/v0.5.0.txt (100%)
 rename doc/source/{whatsnew => }/v0.6.0.txt (100%)
 rename doc/source/{whatsnew => }/v0.6.1.txt (100%)
 rename doc/source/{whatsnew => }/v0.7.0.txt (100%)
 rename doc/source/{whatsnew => }/v0.7.1.txt (100%)
 rename doc/source/{whatsnew => }/v0.7.2.txt (100%)
 rename doc/source/{whatsnew => }/v0.7.3.txt (100%)
 rename doc/source/{whatsnew => }/v0.8.0.txt (86%)

diff --git a/RELEASE.rst b/RELEASE.rst
index d187b211e3f2f..120f7fca03ce6 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -43,7 +43,7 @@ pandas 0.8.0
     conversion method (#1018)
   - Implement robust frequency inference function and `inferred_freq` attribute
     on DatetimeIndex (#391)
-  - New ``tz_convert`` methods in Series / DataFrame
+  - New ``tz_convert`` and ``tz_localize`` methods in Series / DataFrame
   - Convert DatetimeIndexes to UTC if time zones are different in join/setops
     (#864)
   - Add limit argument for forward/backward filling to reindex, fillna,
diff --git a/doc/source/whatsnew/v0.4.x.txt b/doc/source/v0.4.x.txt
similarity index 100%
rename from doc/source/whatsnew/v0.4.x.txt
rename to doc/source/v0.4.x.txt
diff --git a/doc/source/whatsnew/v0.5.0.txt b/doc/source/v0.5.0.txt
similarity index 100%
rename from doc/source/whatsnew/v0.5.0.txt
rename to doc/source/v0.5.0.txt
diff --git a/doc/source/whatsnew/v0.6.0.txt b/doc/source/v0.6.0.txt
similarity index 100%
rename from doc/source/whatsnew/v0.6.0.txt
rename to doc/source/v0.6.0.txt
diff --git a/doc/source/whatsnew/v0.6.1.txt b/doc/source/v0.6.1.txt
similarity index 100%
rename from doc/source/whatsnew/v0.6.1.txt
rename to doc/source/v0.6.1.txt
diff --git a/doc/source/whatsnew/v0.7.0.txt b/doc/source/v0.7.0.txt
similarity index 100%
rename from doc/source/whatsnew/v0.7.0.txt
rename to doc/source/v0.7.0.txt
diff --git a/doc/source/whatsnew/v0.7.1.txt b/doc/source/v0.7.1.txt
similarity index 100%
rename from doc/source/whatsnew/v0.7.1.txt
rename to doc/source/v0.7.1.txt
diff --git a/doc/source/whatsnew/v0.7.2.txt b/doc/source/v0.7.2.txt
similarity index 100%
rename from doc/source/whatsnew/v0.7.2.txt
rename to doc/source/v0.7.2.txt
diff --git a/doc/source/whatsnew/v0.7.3.txt b/doc/source/v0.7.3.txt
similarity index 100%
rename from doc/source/whatsnew/v0.7.3.txt
rename to doc/source/v0.7.3.txt
diff --git a/doc/source/whatsnew/v0.8.0.txt b/doc/source/v0.8.0.txt
similarity index 86%
rename from doc/source/whatsnew/v0.8.0.txt
rename to doc/source/v0.8.0.txt
index cbf094752d4a9..c4049c208029e 100644
--- a/doc/source/whatsnew/v0.8.0.txt
+++ b/doc/source/v0.8.0.txt
@@ -67,15 +67,16 @@ Time series changes and improvements
   PeriodIndex and DatetimeIndex
 - New Timestamp data type subclasses `datetime.datetime`, providing the same
   interface while enabling working with nanosecond-resolution data. Also
-  provides **easy time zone conversions**
-- Enhanced support for **time zones**. Add `tz_convert` methods to TimeSeries
-  and DataFrame. All timestamps are stored as UTC; Timestamps from
-  DatetimeIndex objects with time zone set will be localized to localtime. Time
-  zone conversions are therefore essentially free. User needs to know very
-  little about pytz library now; only time zone names as as strings are
-  required. Timestamps are equal if and only if their UTC timestamps
-  match. Operations between time series with different time zones will result
-  in a UTC-indexed time series
+  provides :ref:`easy time zone conversions <timeseries.timezone>`.
+- Enhanced support for :ref:`time zones <timeseries.timezone>`. Add
+  `tz_convert` and ``tz_lcoalize`` methods to TimeSeries and DataFrame. All
+  timestamps are stored as UTC; Timestamps from DatetimeIndex objects with time
+  zone set will be localized to localtime. Time zone conversions are therefore
+  essentially free. User needs to know very little about pytz library now; only
+  time zone names as as strings are required. Time zone-aware timestamps are
+  equal if and only if their UTC timestamps match. Operations between time
+  zone-aware time series with different time zones will result in a UTC-indexed
+  time series.
 - Time series **string indexing conveniences** / shortcuts: slice years, year
   and month, and index values with strings
 - Enhanced time series **plotting**; adaptation of scikits.timeseries
@@ -115,6 +116,7 @@ Other new features
   function) for computing a categorical variable from a continuous variable by
   binning values either into value-based (``cut``) or quantile-based (``qcut``)
   bins
+- Rename ``Factor`` to ``Categorical`` and add a number of usability features
 - Add :ref:`limit <missing_data.fillna.limit>` argument to fillna/reindex
 - More flexible multiple function application in GroupBy, and can pass list
   (name, function) tuples to get result in particular order with given names
@@ -135,8 +137,8 @@ Other new features
   memory usage than Python's dict
 - Add first, last, min, max, and prod optimized GroupBy functions
 - New :ref:`ordered_merge <merging.ordered_merge>` function
-- Add flexible :ref:`comparison <basics.binop>` instance methods eq, ne, lt, gt, etc. to DataFrame,
-  Series
+- Add flexible :ref:`comparison <basics.binop>` instance methods eq, ne, lt,
+  gt, etc. to DataFrame, Series
 - Improve :ref:`scatter_matrix <visualization.scatter_matrix>` plotting
   function and add histogram or kernel density estimates to diagonal
 - Add :ref:`'kde' <visualization.kde>` plot option for density plots
@@ -148,6 +150,42 @@ Other new features
 - Can select multiple columns from GroupBy
 - Add :ref:`update <merging.combine_first.update>` methods to Series/DataFrame
   for updating values in place
+- Add ``any`` and ``all method to DataFrame
+
+New plotting methods
+~~~~~~~~~~~~~~~~~~~~
+
+.. ipython:: python
+   :suppress:
+
+   import pandas as pd
+   fx = pd.load('data/fx_prices')
+   import matplotlib.pyplot as plt
+
+``Series.plot`` now supports a ``secondary_y`` option:
+
+.. ipython:: python
+
+   plt.figure()
+
+   fx['FR'].plot(style='g')
+
+   @savefig whatsnew_secondary_y.png width=4.5in
+   fx['IT'].plot(style='k--', secondary_y=True)
+
+Vytautas Jancauskas, the 2012 GSOC participant, has added many new plot
+types. For example, ``'kde'`` is a new option:
+
+.. ipython:: python
+
+   s = Series(np.concatenate((np.random.randn(1000),
+                              np.random.randn(1000) * 0.5 + 3)))
+   plt.figure()
+   s.hist(normed=True, alpha=0.2)
+   @savefig whatsnew_kde.png width=4.5in
+   s.plot(kind='kde')
+
+See :ref:`the plotting page <visualization.other>` for much more.
 
 Other API changes
 ~~~~~~~~~~~~~~~~~
diff --git a/doc/source/visualization.rst b/doc/source/visualization.rst
index dcf0f3b87343f..eb15d7f77252d 100644
--- a/doc/source/visualization.rst
+++ b/doc/source/visualization.rst
@@ -121,6 +121,7 @@ You can pass an ``ax`` argument to ``Series.plot`` to plot on a particular axis:
    @savefig series_plot_multi.png width=4.5in
    df['D'].plot(ax=axes[1,1]); axes[1,1].set_title('D')
 
+.. _visualization.other:
 
 Other plotting features
 -----------------------
diff --git a/doc/source/whatsnew.rst b/doc/source/whatsnew.rst
index b930bdbbde1b1..e60baa37820c9 100644
--- a/doc/source/whatsnew.rst
+++ b/doc/source/whatsnew.rst
@@ -16,21 +16,21 @@ What's New
 
 These are new features and improvements of note in each release.
 
-.. include:: whatsnew/v0.8.0.txt
+.. include:: v0.8.0.txt
 
-.. include:: whatsnew/v0.7.3.txt
+.. include:: v0.7.3.txt
 
-.. include:: whatsnew/v0.7.2.txt
+.. include:: v0.7.2.txt
 
-.. include:: whatsnew/v0.7.1.txt
+.. include:: v0.7.1.txt
 
-.. include:: whatsnew/v0.7.0.txt
+.. include:: v0.7.0.txt
 
-.. include:: whatsnew/v0.6.1.txt
+.. include:: v0.6.1.txt
 
-.. include:: whatsnew/v0.6.0.txt
+.. include:: v0.6.0.txt
 
-.. include:: whatsnew/v0.5.0.txt
+.. include:: v0.5.0.txt
 
-.. include:: whatsnew/v0.4.x.txt
+.. include:: v0.4.x.txt
 
diff --git a/pandas/tseries/frequencies.py b/pandas/tseries/frequencies.py
index 433e88e2d7ecb..98f9fbe3c5b6c 100644
--- a/pandas/tseries/frequencies.py
+++ b/pandas/tseries/frequencies.py
@@ -259,23 +259,20 @@ def _get_freq_str(base, mult=1):
 }
 
 need_suffix = ['QS', 'BQ', 'BQS', 'AS', 'BA', 'BAS']
-months = ['JAN', 'FEB', 'MAR', 'APR', 'MAY', 'JUN', 'JUL', 'AUG', 'SEP',
-          'OCT', 'NOV', 'DEC']
-for prefix in need_suffix:
-    for m in months:
-        _offset_to_period_map['%s-%s' % (prefix, m)] = \
-            _offset_to_period_map[prefix]
-
-months = ['JAN', 'FEB', 'MAR', 'APR', 'MAY', 'JUN', 'JUL', 'AUG', 'SEP',
-          'OCT', 'NOV', 'DEC']
-for prefix in ['A', 'Q']:
-    for m in months:
-        alias = '%s-%s' % (prefix, m)
-        _offset_to_period_map[alias] = alias
+_months = ['JAN', 'FEB', 'MAR', 'APR', 'MAY', 'JUN', 'JUL', 'AUG', 'SEP',
+           'OCT', 'NOV', 'DEC']
+for __prefix in need_suffix:
+    for _m in _months:
+        _offset_to_period_map['%s-%s' % (__prefix, _m)] = \
+            _offset_to_period_map[__prefix]
+for __prefix in ['A', 'Q']:
+    for _m in _months:
+        _alias = '%s-%s' % (__prefix, _m)
+        _offset_to_period_map[_alias] = _alias
 
 _days = ['MON', 'TUE', 'WED', 'THU', 'FRI', 'SAT', 'SUN']
-for d in _days:
-    _offset_to_period_map['W-%s' % d] = 'W-%s' % d
+for _d in _days:
+    _offset_to_period_map['W-%s' % _d] = 'W-%s' % _d
 
 def get_period_alias(offset_str):
     """ alias to closest period strings BQ->Q etc"""
@@ -322,11 +319,11 @@ def get_period_alias(offset_str):
     'us': 'U'
 }
 
-for i, weekday in enumerate(['MON', 'TUE', 'WED', 'THU', 'FRI']):
-    for iweek in xrange(4):
-        name = 'WOM-%d%s' % (iweek + 1, weekday)
-        _offset_map[name] = offsets.WeekOfMonth(week=iweek, weekday=i)
-        _rule_aliases[name.replace('-', '@')] = name
+for _i, _weekday in enumerate(['MON', 'TUE', 'WED', 'THU', 'FRI']):
+    for _iweek in xrange(4):
+        _name = 'WOM-%d%s' % (_iweek + 1, _weekday)
+        _offset_map[_name] = offsets.WeekOfMonth(week=_iweek, weekday=_i)
+        _rule_aliases[_name.replace('-', '@')] = _name
 
 _legacy_reverse_map = dict((v, k) for k, v in _rule_aliases.iteritems())
 
@@ -562,8 +559,8 @@ def get_standard_freq(freq):
 }
 
 _reverse_period_code_map = {}
-for k, v in _period_code_map.iteritems():
-    _reverse_period_code_map[v] = k
+for _k, _v in _period_code_map.iteritems():
+    _reverse_period_code_map[_v] = _k
 
 # Additional aliases
 _period_code_map.update({

From a6f553424b1aa8c433d6727d1622594ffe68ed26 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 18:57:18 -0400
Subject: [PATCH 09/42] BUG: DataFrame constructor regression with dict of
 tuples close #1491

---
 pandas/core/frame.py       | 2 +-
 pandas/tests/test_frame.py | 8 ++++++++
 2 files changed, 9 insertions(+), 1 deletion(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index b071bca8cf5de..139f0bfb94c20 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4604,7 +4604,7 @@ def extract_index(data):
             elif isinstance(v, dict):
                 have_dicts = True
                 indexes.append(v.keys())
-            elif isinstance(v, (list, np.ndarray)):
+            elif isinstance(v, (list, tuple, np.ndarray)):
                 have_raw_arrays = True
                 raw_lengths.append(len(v))
 
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 6c3cc89f18f1e..58b7a8c847735 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -1579,6 +1579,14 @@ def test_constructor_dict_dont_upcast(self):
         dm = DataFrame([[1,2],['a','b']], index=[1,2], columns=[1,2])
         self.assert_(isinstance(dm[1][1], int))
 
+    def test_constructor_dict_of_tuples(self):
+        # GH #1491
+        data = {'a': (1, 2, 3), 'b': (4, 5, 6)}
+
+        result = DataFrame(data)
+        expected = DataFrame(dict((k, list(v)) for k, v in data.iteritems()))
+        assert_frame_equal(result, expected)
+
     def test_constructor_ndarray(self):
         mat = np.zeros((2, 3), dtype=float)
 

From 2c511445b27ca61dfb2a4e9ca36901c36cb6a891 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 19:02:52 -0400
Subject: [PATCH 10/42] BUG: don't accidentally upcast values in
 DataFrame.itertuples, close #1505

---
 pandas/core/frame.py       | 8 ++++++--
 pandas/tests/test_frame.py | 6 ++++++
 2 files changed, 12 insertions(+), 2 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 139f0bfb94c20..e84d51bbe92ab 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -615,12 +615,16 @@ def iterrows(self):
             s.name = k
             yield k, s
 
-    def itertuples(self):
+    def itertuples(self, index=True):
         """
         Iterate over rows of DataFrame as tuples, with index value
         as first element of the tuple
         """
-        return izip(self.index, *self.values.T)
+        arrays = []
+        if index:
+            arrays.append(self.index)
+        arrays.extend(self[k] for k in self.columns)
+        return izip(*arrays)
 
     iterkv = iteritems
     if py3compat.PY3:  # pragma: no cover
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 58b7a8c847735..1906c3936292d 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -2544,6 +2544,12 @@ def test_itertuples(self):
             expected = self.frame.ix[i,:].reset_index(drop=True)
             assert_series_equal(s, expected)
 
+        df = DataFrame({'floats': np.random.randn(5),
+                        'ints': range(5)}, columns=['floats', 'ints'])
+
+        for tup in df.itertuples(index=False):
+            self.assert_(isinstance(tup[1], np.integer))
+
     def test_len(self):
         self.assertEqual(len(self.frame), len(self.frame.index))
 

From 87d6da19b68c459573a15c911f1a827705767fe5 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Thu, 21 Jun 2012 19:11:05 -0400
Subject: [PATCH 11/42] BUG: revert to circle marker for markers not in
 matplotlib < 1.1. close #1484

---
 pandas/tools/plotting.py | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 9c9165e61ff9a..6aae74571182f 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -50,6 +50,8 @@ def scatter_matrix(frame, alpha=0.5, figsize=None, ax=None, grid=False,
 
     mask = com.notnull(df)
 
+    marker = _get_marker_compat(marker)
+
     for i, a in zip(range(n), df.columns):
         for j, b in zip(range(n), df.columns):
             if i == j:
@@ -130,6 +132,12 @@ def _gcf():
     import matplotlib.pyplot as plt
     return plt.gcf()
 
+def _get_marker_compat(marker):
+    import matplotlib.lines as mlines
+    if marker not in mlines.lineMarkers:
+        return 'o'
+    return marker
+
 def andrews_curves(data, class_column, ax=None, samples=200):
     """
     Parameters:

From 9d2d4bfb9bcb7a3003ab423aa17e922e6b0cf29d Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Fri, 22 Jun 2012 10:29:09 -0400
Subject: [PATCH 12/42] BUG: clean NAs in DataFrame.boxplot, close #1506

---
 pandas/tools/plotting.py | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 6aae74571182f..485d568f91353 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -7,7 +7,7 @@
 from pandas.util.decorators import cache_readonly
 import pandas.core.common as com
 from pandas.core.index import Index, MultiIndex
-from pandas.core.series import Series
+from pandas.core.series import Series, remove_na
 from pandas.tseries.index import DatetimeIndex
 from pandas.tseries.period import PeriodIndex
 from pandas.tseries.frequencies import get_period_alias, get_base_alias
@@ -1007,7 +1007,7 @@ def boxplot(data, column=None, by=None, ax=None, fontsize=None,
     def plot_group(grouped, ax):
         keys, values = zip(*grouped)
         keys = [_stringify(x) for x in keys]
-        ax.boxplot(values, **kwds)
+        ax.boxplot(remove_na(values), **kwds)
         if kwds.get('vert', 1):
             ax.set_xticklabels(keys, rotation=rot, fontsize=fontsize)
         else:
@@ -1043,7 +1043,8 @@ def plot_group(grouped, ax):
 
         # Return boxplot dict in single plot case
 
-        bp = ax.boxplot(list(data[cols].values.T), **kwds)
+        clean_values = [remove_na(x) for x in data[cols].values.T]
+        bp = ax.boxplot(clean_values, **kwds)
         if kwds.get('vert', 1):
             ax.set_xticklabels(keys, rotation=rot, fontsize=fontsize)
         else:

From 44a8295c8f708b40a9b1279cf5d741edacf9d58e Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Fri, 22 Jun 2012 15:16:00 -0400
Subject: [PATCH 13/42] BUG: fix NumPy 1.7 cast-to-bool failure and buglet

---
 doc/source/timeseries.rst | 2 +-
 pandas/core/frame.py      | 6 +++---
 pandas/core/index.py      | 2 +-
 3 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/doc/source/timeseries.rst b/doc/source/timeseries.rst
index c616a07670e02..ac62bae5071cf 100644
--- a/doc/source/timeseries.rst
+++ b/doc/source/timeseries.rst
@@ -297,7 +297,7 @@ We could have done the same thing with ``DateOffset``:
 
 .. ipython:: python
 
-   from pandas.core.datetools import *
+   from pandas.tseries.offsets import *
    d + DateOffset(months=4, days=5)
 
 The key features of a ``DateOffset`` object are:
diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index e84d51bbe92ab..4d41c02ad4200 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4226,9 +4226,9 @@ def _reduce(self, op, axis=0, skipna=True, numeric_only=None,
         if result.dtype == np.object_:
             try:
                 if filter_type is None or filter_type == 'numeric':
-                    result = result.astype('f8')
-                elif filter_type == 'bool':
-                    result = result.astype('b')
+                    result = result.astype(np.float64)
+                elif filter_type == 'bool' and notnull(result).all():
+                    result = result.astype(np.bool_)
                 else:
                     raise ValueError('Invalid dtype %s ' % str(filter_type))
 
diff --git a/pandas/core/index.py b/pandas/core/index.py
index db0bd1bd7147e..14edc86cc600c 100644
--- a/pandas/core/index.py
+++ b/pandas/core/index.py
@@ -405,7 +405,7 @@ def asof(self, label):
     def asof_locs(self, where, mask):
         """
         where : array of timestamps
-        mask : array of booleans where data is NA
+        mask : array of booleans where data is not NA
 
         """
         locs = self.values[mask].searchsorted(where.values, side='right')

From 836ecab757b3b952e6599c8dcc6b6c761b27db7a Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Fri, 22 Jun 2012 15:26:56 -0400
Subject: [PATCH 14/42] BUG: preserve index names in MultiIndex.drop, close
 #1513

---
 pandas/core/index.py            |  3 ++-
 pandas/tests/test_multilevel.py | 10 ++++++++++
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/pandas/core/index.py b/pandas/core/index.py
index 14edc86cc600c..35096651b90ef 100644
--- a/pandas/core/index.py
+++ b/pandas/core/index.py
@@ -2295,7 +2295,8 @@ def delete(self, loc):
         new_index : MultiIndex
         """
         new_labels = [np.delete(lab, loc) for lab in self.labels]
-        return MultiIndex(levels=self.levels, labels=new_labels)
+        return MultiIndex(levels=self.levels, labels=new_labels,
+                          names=self.names)
 
     get_major_bounds = slice_locs
 
diff --git a/pandas/tests/test_multilevel.py b/pandas/tests/test_multilevel.py
index e6698ff52afaf..1189df2ac3ce6 100644
--- a/pandas/tests/test_multilevel.py
+++ b/pandas/tests/test_multilevel.py
@@ -1355,6 +1355,16 @@ def test_drop_level(self):
         expected = self.frame.ix[[0, 2, 3, 6, 7, 9]].T
         assert_frame_equal(result, expected)
 
+    def test_drop_preserve_names(self):
+        index = MultiIndex.from_arrays([[0, 0, 0, 1, 1, 1],
+                                        [1, 2, 3, 1, 2, 3]],
+                                       names=['one', 'two'])
+
+        df = DataFrame(np.random.randn(6, 3), index=index)
+
+        result = df.drop([(0, 2)])
+        self.assert_(result.index.names == ['one', 'two'])
+
     def test_unicode_repr_issues(self):
         levels = [Index([u'a/\u03c3', u'b/\u03c3',u'c/\u03c3']),
                   Index([0, 1])]

From 5f5df2aa34f027d3fc83abb19cf7840ababd4557 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Fri, 22 Jun 2012 15:34:42 -0400
Subject: [PATCH 15/42] RLS: 0.8.0 release candidate 1

---
 RELEASE.rst | 3 ++-
 setup.py    | 4 ++--
 2 files changed, 4 insertions(+), 3 deletions(-)

diff --git a/RELEASE.rst b/RELEASE.rst
index 120f7fca03ce6..bfb2ca515372a 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -25,7 +25,7 @@ Where to get it
 pandas 0.8.0
 ============
 
-**Release date:** NOT YET RELEASED
+**Release date:** 6/22/2012
 
 **New features**
 
@@ -203,6 +203,7 @@ pandas 0.8.0
   - Fix unicode serialization issue in HDFStore (#1361)
   - Pass keywords to pyplot.boxplot in DataFrame.boxplot (#1493)
   - Bug fixes in MonthBegin (#1483)
+  - Preserve MultiIndex names in drop (#1513)
 
 pandas 0.7.3
 ============
diff --git a/setup.py b/setup.py
index ba80a464060dc..615d3a0b4ff80 100755
--- a/setup.py
+++ b/setup.py
@@ -174,9 +174,9 @@
 MAJOR = 0
 MINOR = 8
 MICRO = 0
-ISRELEASED = False
+ISRELEASED = True
 VERSION = '%d.%d.%d' % (MAJOR, MINOR, MICRO)
-QUALIFIER = 'b2'
+QUALIFIER = 'rc1'
 
 FULLVERSION = VERSION
 if not ISRELEASED:

From 5d90f7d0e7e9bb53e55fdb7b70f972aa502004ed Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Fri, 22 Jun 2012 15:44:31 -0400
Subject: [PATCH 16/42] BUG: fix failure in grouped boxplot causing docs bug.
 push rc number to 2

---
 pandas/tests/test_graphics.py | 3 +++
 pandas/tools/plotting.py      | 3 ++-
 setup.py                      | 2 +-
 3 files changed, 6 insertions(+), 2 deletions(-)

diff --git a/pandas/tests/test_graphics.py b/pandas/tests/test_graphics.py
index 57bc17c5c33b6..2c386076b69b7 100644
--- a/pandas/tests/test_graphics.py
+++ b/pandas/tests/test_graphics.py
@@ -194,6 +194,9 @@ def test_boxplot(self):
         _check_plot_works(df.boxplot, notch=1)
         _check_plot_works(df.boxplot, by='indic', notch=1)
 
+        df = DataFrame(np.random.rand(10,2), columns=['Col1', 'Col2'] )
+        df['X'] = Series(['A','A','A','A','A','B','B','B','B','B'])
+        _check_plot_works(df.boxplot, by='X')
 
     @slow
     def test_kde(self):
diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 485d568f91353..97bea5cb7f7ec 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -1007,7 +1007,8 @@ def boxplot(data, column=None, by=None, ax=None, fontsize=None,
     def plot_group(grouped, ax):
         keys, values = zip(*grouped)
         keys = [_stringify(x) for x in keys]
-        ax.boxplot(remove_na(values), **kwds)
+        values = [remove_na(v) for v in values]
+        ax.boxplot(values, **kwds)
         if kwds.get('vert', 1):
             ax.set_xticklabels(keys, rotation=rot, fontsize=fontsize)
         else:
diff --git a/setup.py b/setup.py
index 615d3a0b4ff80..2c62f7051bfce 100755
--- a/setup.py
+++ b/setup.py
@@ -176,7 +176,7 @@
 MICRO = 0
 ISRELEASED = True
 VERSION = '%d.%d.%d' % (MAJOR, MINOR, MICRO)
-QUALIFIER = 'rc1'
+QUALIFIER = 'rc2'
 
 FULLVERSION = VERSION
 if not ISRELEASED:

From 896140435e9cfbd8e893668c92b2e5d9b9389276 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 20:47:55 -0400
Subject: [PATCH 17/42] ENH: freqstr with offset #1184

---
 pandas/tseries/offsets.py            | 54 ++++++++++++++++++++++++++--
 pandas/tseries/tests/test_offsets.py |  7 ++++
 2 files changed, 59 insertions(+), 2 deletions(-)

diff --git a/pandas/tseries/offsets.py b/pandas/tseries/offsets.py
index b03d8febbfbc2..8f62c3304cb53 100644
--- a/pandas/tseries/offsets.py
+++ b/pandas/tseries/offsets.py
@@ -221,10 +221,11 @@ def freqstr(self):
             return repr(self)
 
         if self.n != 1:
-            return '%d%s' % (self.n, code)
+            fstr = '%d%s' % (self.n, code)
         else:
-            return code
+            fstr = code
 
+        return fstr
 
 class BusinessDay(CacheableOffset, DateOffset):
     """
@@ -261,6 +262,55 @@ def __repr__(self):
         out += '>'
         return out
 
+    @property
+    def freqstr(self):
+        try:
+            code = self.rule_code
+        except NotImplementedError:
+            return repr(self)
+
+        if self.n != 1:
+            fstr = '%d%s' % (self.n, code)
+        else:
+            fstr = code
+
+        if self.offset:
+            fstr += self._offset_str()
+
+        return fstr
+
+    def _offset_str(self):
+
+        def get_str(td):
+            off_str = ''
+            if td.days > 0:
+                off_str += str(td.days) + 'D'
+            if td.seconds > 0:
+                s = td.seconds
+                hrs = int(s / 3600)
+                if hrs != 0:
+                    off_str += str(hrs) + 'H'
+                    s -= hrs * 3600
+                mts = int(s / 60)
+                if mts != 0:
+                    off_str += str(mts) + 'Min'
+                    s -= mts * 60
+                if s != 0:
+                    off_str += str(s) + 's'
+            if td.microseconds > 0:
+                off_str += str(td.microseconds) + 'us'
+            return off_str
+
+        if isinstance(self.offset, timedelta):
+            tot_sec = self.offset.total_seconds()
+            if tot_sec > 0:
+                off_str = '+' + get_str(self.offset)
+            if tot_sec < 0:
+                off_str = '-' + get_str(-self.offset)
+            return off_str
+        else:
+            return '+' + repr(self.offset)
+
     def isAnchored(self):
         return (self.n == 1)
 
diff --git a/pandas/tseries/tests/test_offsets.py b/pandas/tseries/tests/test_offsets.py
index 05f638afaebed..1311b9ca451a2 100644
--- a/pandas/tseries/tests/test_offsets.py
+++ b/pandas/tseries/tests/test_offsets.py
@@ -1389,6 +1389,13 @@ def test_dateoffset_misc():
 
     assert(not offsets.DateOffset(months=2) == 2)
 
+def test_freq_offsets():
+    off = BDay(1, offset=timedelta(0, 1800))
+    assert(off.freqstr == 'B+30Min')
+
+    off = BDay(1, offset=timedelta(0, -1800))
+    assert(off.freqstr == 'B-30Min')
+
 if __name__ == '__main__':
     import nose
     nose.runmodule(argv=[__file__,'-vvs','-x','--pdb', '--pdb-failure'],

From d90a8290ffbdb3e18b44a7b7db56b6e6408704ba Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 19:10:13 -0400
Subject: [PATCH 18/42] ENH: option to disable empty string conversion to NaN
 in file readers

---
 pandas/io/parsers.py            |  4 ++--
 pandas/io/tests/test_parsers.py | 26 ++++++++++++++++++++++++++
 pandas/src/inference.pyx        | 15 ++++++++++-----
 3 files changed, 38 insertions(+), 7 deletions(-)

diff --git a/pandas/io/parsers.py b/pandas/io/parsers.py
index 8261b889b4500..3b2fedd65b45f 100644
--- a/pandas/io/parsers.py
+++ b/pandas/io/parsers.py
@@ -1029,9 +1029,9 @@ def _convert_types(values, na_values):
         return values, na_count
 
     try:
-        result = lib.maybe_convert_numeric(values, na_values)
+        result = lib.maybe_convert_numeric(values, na_values, False)
     except Exception:
-        na_count = lib.sanitize_objects(values, na_values)
+        na_count = lib.sanitize_objects(values, na_values, False)
         result = values
 
     if result.dtype == np.object_:
diff --git a/pandas/io/tests/test_parsers.py b/pandas/io/tests/test_parsers.py
index 7a75e687a4a8a..1f14df528af8f 100644
--- a/pandas/io/tests/test_parsers.py
+++ b/pandas/io/tests/test_parsers.py
@@ -52,6 +52,32 @@ def setUp(self):
         self.csv2 = os.path.join(self.dirpath, 'test2.csv')
         self.xls1 = os.path.join(self.dirpath, 'test.xls')
 
+    def test_empty_string(self):
+        data = """\
+One,Two,Three
+a,1,one
+b,2,two
+,3,three
+d,4,nan
+e,5,five
+nan,6,
+g,7,seven
+"""
+        df = read_csv(StringIO(data))
+        xp = DataFrame({'One' : ['a', 'b', np.nan, 'd', 'e', np.nan, 'g'],
+                        'Two' : [1,2,3,4,5,6,7],
+                        'Three' : ['one', 'two', 'three', np.nan, 'five',
+                                   np.nan, 'seven']})
+        assert_frame_equal(xp.reindex(columns=df.columns), df)
+
+        df = read_csv(StringIO(data), na_values={'One': [], 'Three': []})
+        xp = DataFrame({'One' : ['a', 'b', '', 'd', 'e', 'nan', 'g'],
+                        'Two' : [1,2,3,4,5,6,7],
+                        'Three' : ['one', 'two', 'three', 'nan', 'five',
+                                   '', 'seven']})
+        assert_frame_equal(xp.reindex(columns=df.columns), df)
+
+
     def test_read_csv(self):
         pass
 
diff --git a/pandas/src/inference.pyx b/pandas/src/inference.pyx
index 4057d240fceb6..451b998b0481b 100644
--- a/pandas/src/inference.pyx
+++ b/pandas/src/inference.pyx
@@ -241,7 +241,8 @@ def is_date_array(ndarray[object] values):
     return True
 
 
-def maybe_convert_numeric(ndarray[object] values, set na_values):
+def maybe_convert_numeric(ndarray[object] values, set na_values,
+                          convert_empty=True):
     '''
     Type inference function-- convert strings to numeric (potentially) and
     convert to proper dtype array
@@ -275,8 +276,11 @@ def maybe_convert_numeric(ndarray[object] values, set na_values):
             floats[i] = complexes[i] = nan
             seen_float = 1
         elif len(val) == 0:
-            floats[i] = complexes[i] = nan
-            seen_float = 1
+            if convert_empty:
+                floats[i] = complexes[i] = nan
+                seen_float = 1
+            else:
+                raise ValueError('Empty string encountered')
         elif util.is_complex_object(val):
             complexes[i] = val
             seen_complex = 1
@@ -573,7 +577,8 @@ def try_parse_datetime_components(ndarray[object] years, ndarray[object] months,
 
     return result
 
-def sanitize_objects(ndarray[object] values, set na_values):
+def sanitize_objects(ndarray[object] values, set na_values,
+                     convert_empty=True):
     cdef:
         Py_ssize_t i, n
         object val, onan
@@ -585,7 +590,7 @@ def sanitize_objects(ndarray[object] values, set na_values):
 
     for i from 0 <= i < n:
         val = values[i]
-        if val == '' or val in na_values:
+        if (convert_empty and val == '') or (val in na_values):
             values[i] = onan
             na_count += 1
         elif val in memo:

From bf84631de8b37f53162695a7272075d9234a6d39 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 16:47:54 -0400
Subject: [PATCH 19/42] BUG: HDFStore converting all to datetime64 #1520

---
 pandas/io/pytables.py            |  4 ++--
 pandas/io/tests/test_pytables.py | 12 ++++++++++++
 2 files changed, 14 insertions(+), 2 deletions(-)

diff --git a/pandas/io/pytables.py b/pandas/io/pytables.py
index 52af5eff72336..4ec176556eaa8 100644
--- a/pandas/io/pytables.py
+++ b/pandas/io/pytables.py
@@ -716,7 +716,7 @@ def _write_array(self, group, key, value):
             vlarr.append(value)
         elif value.dtype.type == np.datetime64:
             self.handle.createArray(group, key, value.view('i8'))
-            group._v_attrs.value_type = 'datetime64'
+            getattr(group, key)._v_attrs.value_type = 'datetime64'
         else:
             self.handle.createArray(group, key, value)
 
@@ -958,7 +958,7 @@ def _read_array(group, key):
     if isinstance(node, tables.VLArray):
         return data[0]
     else:
-        dtype = getattr(group._v_attrs, 'value_type', None)
+        dtype = getattr(node._v_attrs, 'value_type', None)
         if dtype == 'datetime64':
             return np.array(data, dtype='M8[ns]')
         return data
diff --git a/pandas/io/tests/test_pytables.py b/pandas/io/tests/test_pytables.py
index 8576b3d69ad99..cddd8e53fe55b 100644
--- a/pandas/io/tests/test_pytables.py
+++ b/pandas/io/tests/test_pytables.py
@@ -660,6 +660,18 @@ def test_unicode_index(self):
         s = Series(np.random.randn(len(unicode_values)), unicode_values)
         self._check_roundtrip(s, tm.assert_series_equal)
 
+    def test_store_datetime_mixed(self):
+        df = DataFrame({'a': [1,2,3], 'b': [1.,2.,3.], 'c': ['a', 'b', 'c']})
+        ts = tm.makeTimeSeries()
+        df['d'] = ts.index[:3]
+        store = HDFStore('data.h5')
+        store['df'] = df
+        store.close()
+        store = HDFStore('data.h5')
+        test = store['df']
+        assert_series_equal(df.dtypes, test.dtypes)
+        store.close()
+        os.remove('data.h5')
 
 def curpath():
     pth, _ = os.path.split(os.path.abspath(__file__))

From e2091854057028403c3f004787643ce1ed3ec5a3 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 17:18:14 -0400
Subject: [PATCH 20/42] TST: use _check_roundtrip

---
 pandas/io/tests/test_pytables.py | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)

diff --git a/pandas/io/tests/test_pytables.py b/pandas/io/tests/test_pytables.py
index cddd8e53fe55b..87a6329ef9368 100644
--- a/pandas/io/tests/test_pytables.py
+++ b/pandas/io/tests/test_pytables.py
@@ -664,14 +664,7 @@ def test_store_datetime_mixed(self):
         df = DataFrame({'a': [1,2,3], 'b': [1.,2.,3.], 'c': ['a', 'b', 'c']})
         ts = tm.makeTimeSeries()
         df['d'] = ts.index[:3]
-        store = HDFStore('data.h5')
-        store['df'] = df
-        store.close()
-        store = HDFStore('data.h5')
-        test = store['df']
-        assert_series_equal(df.dtypes, test.dtypes)
-        store.close()
-        os.remove('data.h5')
+        self._check_roundtrip(df, tm.assert_frame_equal)
 
 def curpath():
     pth, _ = os.path.split(os.path.abspath(__file__))

From 3027871161409f84ecac029231cb444dd6a42c69 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Fri, 22 Jun 2012 20:54:34 -0400
Subject: [PATCH 21/42] ENH: retain Series names when constructing DataFrame
 from list of Series #1494

---
 pandas/core/frame.py       | 19 +++++++++++++++++++
 pandas/tests/test_frame.py | 22 ++++++++++++++++++++++
 2 files changed, 41 insertions(+)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 4d41c02ad4200..b51fab43c9e33 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -389,6 +389,9 @@ def __init__(self, data=None, index=None, columns=None, dtype=None,
                                          copy=copy)
         elif isinstance(data, list):
             if len(data) > 0:
+                if index is None and isinstance(data[0], Series):
+                    index = _get_names_from_index(data)
+
                 if isinstance(data[0], (list, tuple, dict, Series)):
                     conv_data, columns = _to_sdict(data, columns)
                     if isinstance(conv_data, dict):
@@ -4758,6 +4761,22 @@ def _convert_object_array(content, columns, coerce_float=False):
                  for c, vals in zip(columns, content))
     return sdict, columns
 
+def _get_names_from_index(data):
+    index = range(len(data))
+    has_some_name = any([s.name is not None for s in data])
+    if not has_some_name:
+        return index
+
+    count = 0
+    for i, s in enumerate(data):
+        n = s.name
+        if n is not None:
+            index[i] = n
+        else:
+            index[i] = 'Unnamed %d' % count
+            count += 1
+
+    return index
 
 def _homogenize(data, index, columns, dtype=None):
     from pandas.core.series import _sanitize_array
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 1906c3936292d..15d465e8df966 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -1871,6 +1871,28 @@ def test_constructor_list_of_dicts(self):
         assert_frame_equal(result, expected)
 
     def test_constructor_list_of_series(self):
+        data = [{'a': 1.5, 'b': 3.0, 'c':4.0},
+                {'a': 1.5, 'b': 3.0, 'c':6.0}]
+        sdict = dict(zip(['x', 'y'], data))
+        idx = Index(['a', 'b', 'c'])
+
+        # all named
+        data2 = [Series([1.5, 3, 4], idx, dtype='O', name='x'),
+                 Series([1.5, 3, 6], idx, name='y')]
+        result = DataFrame(data2)
+        expected = DataFrame.from_dict(sdict, orient='index')
+        assert_frame_equal(result, expected)
+
+        # some unnamed
+        data2 = [Series([1.5, 3, 4], idx, dtype='O', name='x'),
+                 Series([1.5, 3, 6], idx)]
+        result = DataFrame(data2)
+
+        sdict = dict(zip(['x', 'Unnamed 0'], data))
+        expected = DataFrame.from_dict(sdict, orient='index')
+        assert_frame_equal(result.sort_index(), expected)
+
+        # none named
         data = [{'a': 1.5, 'b': 3, 'c':4, 'd':6},
                 {'a': 1.5, 'b': 3, 'd':6},
                 {'a': 1.5, 'd':6},

From c903094bd5ed3a6ce8509e001f5fc69b500ed220 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 10:13:59 -0400
Subject: [PATCH 22/42] DOC: release notes

---
 RELEASE.rst | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/RELEASE.rst b/RELEASE.rst
index bfb2ca515372a..79b980694860a 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -25,7 +25,7 @@ Where to get it
 pandas 0.8.0
 ============
 
-**Release date:** 6/22/2012
+**Release date:** 6/26/2012
 
 **New features**
 
@@ -131,6 +131,7 @@ pandas 0.8.0
   - Add ``retries`` argument to ``get_data_yahoo`` to try to prevent Yahoo! API
     404s (#826)
   - Improve performance of reshaping by using O(N) categorical sorting
+  - Series names will be used for index of DataFrame if no index passed (#1494)
 
 **API Changes**
 

From 66690f276619874061a10b0844f4df93292c2d18 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 00:02:03 -0400
Subject: [PATCH 23/42] BUG/TST: plot irregular and reg freq on same subplot

---
 pandas/tools/plotting.py              | 13 +++++++++---
 pandas/tseries/plotting.py            | 30 ++++++++++++++++++---------
 pandas/tseries/tests/test_plotting.py | 28 ++++++++++++++++++++++++-
 3 files changed, 57 insertions(+), 14 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 97bea5cb7f7ec..8cf584155d1c8 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -596,10 +596,13 @@ def __init__(self, data, **kwargs):
 
     @property
     def has_ts_index(self):
+        # TODO refactor this whole regular/irregular kludge
         from pandas.core.frame import DataFrame
         if isinstance(self.data, (Series, DataFrame)):
+            ax, _ = self._get_ax_and_style(0)
             freq = (getattr(self.data.index, 'freq', None)
-                    or getattr(self.data.index, 'inferred_freq', None))
+                    or getattr(self.data.index, 'inferred_freq', None)
+                    or getattr(ax, 'freq', None))
             return (freq is not None) and  self._has_dynamic_index_freq(freq)
         return False
 
@@ -649,8 +652,12 @@ def _maybe_convert_index(self, data):
 
             freq = get_period_alias(freq)
 
-            if freq is None and hasattr(data.index, 'inferred_freq'):
-                freq = data.index.inferred_freq
+            if freq is None:
+                freq = getattr(data.index, 'inferred_freq', None)
+
+            if freq is None:
+                ax, _ = self._get_ax_and_style(0)
+                freq = getattr(ax, 'freq', None)
 
             if isinstance(freq, DateOffset):
                 freq = freq.rule_code
diff --git a/pandas/tseries/plotting.py b/pandas/tseries/plotting.py
index 79823e8c12dca..7f4b4e5130c66 100644
--- a/pandas/tseries/plotting.py
+++ b/pandas/tseries/plotting.py
@@ -44,8 +44,17 @@ def tsplot(series, plotf, **kwargs):
     """
     # Used inferred freq is possible, need a test case for inferred
     freq = getattr(series.index, 'freq', None)
-    if freq is None and hasattr(series.index, 'inferred_freq'):
-        freq = series.index.inferred_freq
+    if freq is None:
+        freq = getattr(series.index, 'inferred_freq', None)
+
+    if 'ax' in kwargs:
+        ax = kwargs.pop('ax')
+    else:
+        import matplotlib.pyplot as plt
+        ax = plt.gca()
+
+    if freq is None:
+        freq = getattr(ax, 'freq', None)
 
     if isinstance(freq, DateOffset):
         freq = freq.rule_code
@@ -62,12 +71,6 @@ def tsplot(series, plotf, **kwargs):
 
     style = kwargs.pop('style', None)
 
-    if 'ax' in kwargs:
-        ax = kwargs.pop('ax')
-    else:
-        import matplotlib.pyplot as plt
-        ax = plt.gca()
-
     # Specialized ts plotting attributes for Axes
     ax.freq = freq
     xaxis = ax.get_xaxis()
@@ -93,12 +96,19 @@ def tsplot(series, plotf, **kwargs):
 
     format_dateaxis(ax, ax.freq)
 
-    left = series.index[0] #get_datevalue(series.index[0], freq)
-    right = series.index[-1] #get_datevalue(series.index[-1], freq)
+    left, right = _get_xlim(ax.get_lines())
     ax.set_xlim(left, right)
 
     return ax
 
+def _get_xlim(lines):
+    left, right = np.inf, -np.inf
+    for l in lines:
+        x = l.get_xdata()
+        left = min(x[0].ordinal, left)
+        right = max(x[-1].ordinal, right)
+    return left, right
+
 def get_datevalue(date, freq):
     if isinstance(date, Period):
         return date.asfreq(freq).ordinal
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index 81f972debe6c2..9331d79efe15a 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -234,8 +234,10 @@ def test_finder_daily(self):
 
     @slow
     def test_finder_quarterly(self):
+        import matplotlib.pyplot as plt
         xp = Period('1988Q1').ordinal
         yrs = [3.5, 11]
+        plt.close('all')
         for n in yrs:
             rng = period_range('1987Q2', periods=int(n * 4), freq='Q')
             ser = Series(np.random.randn(len(rng)), rng)
@@ -250,8 +252,10 @@ def test_finder_quarterly(self):
 
     @slow
     def test_finder_monthly(self):
+        import matplotlib.pyplot as plt
         xp = Period('1988-1').ordinal
         yrs = [1.15, 2.5, 4, 11]
+        plt.close('all')
         for n in yrs:
             rng = period_range('1987Q2', periods=int(n * 12), freq='M')
             ser = Series(np.random.randn(len(rng)), rng)
@@ -263,8 +267,12 @@ def test_finder_monthly(self):
             ax.set_xlim(vmin + 0.9, vmax)
             rs = xaxis.get_majorticklocs()[0]
             self.assertEqual(xp, rs)
+            plt.close('all')
 
-
+    @slow
+    def test_finder_monthly_long(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
         rng = period_range('1988Q1', periods=24*12, freq='M')
         ser = Series(np.random.randn(len(rng)), rng)
         ax = ser.plot()
@@ -433,6 +441,24 @@ def test_secondary_frame(self):
         self.assert_(axes[1].get_yaxis().get_ticks_position() == 'default')
         self.assert_(axes[2].get_yaxis().get_ticks_position() == 'right')
 
+    @slow
+    def test_mixed_freq(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        s1 = tm.makeTimeSeries()
+        s2 = s1[[0, 5, 10, 11, 12, 13, 14, 15]]
+        s1.plot()
+        ax2 = s2.plot(style='g')
+        lines = ax2.get_lines()
+        idx1 = lines[0].get_xdata()
+        idx2 = lines[1].get_xdata()
+        self.assert_(idx1.equals(s1.index.to_period('B')))
+        self.assert_(idx2.equals(s2.index.to_period('B')))
+        left, right = ax2.get_xlim()
+        pidx = s1.index.to_period()
+        self.assert_(left == pidx[0].ordinal)
+        self.assert_(right == pidx[-1].ordinal)
+
 PNG_PATH = 'tmp.png'
 def _check_plot_works(f, freq=None, series=None, *args, **kwargs):
     import matplotlib.pyplot as plt

From 0940b78a0568020ca08ef02131f57e3b6285cf3d Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 14:36:31 -0400
Subject: [PATCH 24/42] BUG: plot mixed frequencies #1517

---
 pandas/tools/plotting.py              | 44 ++++++++++------
 pandas/tseries/plotting.py            | 66 +++++++++++++++--------
 pandas/tseries/tests/test_plotting.py | 75 ++++++++++++++++++++++++---
 3 files changed, 140 insertions(+), 45 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 8cf584155d1c8..30505cfc4d48c 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -485,13 +485,16 @@ def plt(self):
 
     _need_to_set_index = False
 
-    def _get_xticks(self):
+    def _get_xticks(self, convert_period=False):
         index = self.data.index
         is_datetype = index.inferred_type in ('datetime', 'date',
                                               'datetime64')
 
         if self.use_index:
-            if index.is_numeric() or is_datetype:
+            if convert_period and isinstance(index, PeriodIndex):
+                index = index.to_timestamp()
+                x = index._mpl_repr()
+            elif index.is_numeric() or is_datetype:
                 """
                 Matplotlib supports numeric values or datetime objects as
                 xaxis values. Taking LBYL approach here, by the time
@@ -594,19 +597,14 @@ class LinePlot(MPLPlot):
     def __init__(self, data, **kwargs):
         MPLPlot.__init__(self, data, **kwargs)
 
-    @property
-    def has_ts_index(self):
-        # TODO refactor this whole regular/irregular kludge
+    def _index_freq(self):
         from pandas.core.frame import DataFrame
         if isinstance(self.data, (Series, DataFrame)):
-            ax, _ = self._get_ax_and_style(0)
             freq = (getattr(self.data.index, 'freq', None)
-                    or getattr(self.data.index, 'inferred_freq', None)
-                    or getattr(ax, 'freq', None))
-            return (freq is not None) and  self._has_dynamic_index_freq(freq)
-        return False
+                    or getattr(self.data.index, 'inferred_freq', None))
+            return freq
 
-    def _has_dynamic_index_freq(self, freq):
+    def _is_dynamic_freq(self, freq):
         if isinstance(freq, DateOffset):
             freq = freq.rule_code
         else:
@@ -614,13 +612,26 @@ def _has_dynamic_index_freq(self, freq):
         freq = get_period_alias(freq)
         return freq is not None
 
+    def _use_dynamic_x(self):
+        freq = self._index_freq()
+
+        ax, _ = self._get_ax_and_style(0)
+        ax_freq = getattr(ax, 'freq', None)
+        if freq is None: # convert irregular if axes has freq info
+            freq = ax_freq
+        else: # do not use tsplot if irregular was plotted first
+            if (ax_freq is None) and (len(ax.get_lines()) > 0):
+                return False
+
+        return (freq is not None) and self._is_dynamic_freq(freq)
+
     def _make_plot(self):
         # this is slightly deceptive
-        if self.use_index and self.has_ts_index:
+        if self.use_index and self._use_dynamic_x():
             data = self._maybe_convert_index(self.data)
             self._make_ts_plot(data)
         else:
-            x = self._get_xticks()
+            x = self._get_xticks(convert_period=True)
 
             plotf = self._get_plot_function()
 
@@ -649,6 +660,8 @@ def _maybe_convert_index(self, data):
         if (isinstance(data.index, DatetimeIndex) and
             isinstance(data, DataFrame)):
             freq = getattr(data.index, 'freqstr', None)
+            if isinstance(freq, DateOffset):
+                freq = freq.rule_code
 
             freq = get_period_alias(freq)
 
@@ -659,9 +672,6 @@ def _maybe_convert_index(self, data):
                 ax, _ = self._get_ax_and_style(0)
                 freq = getattr(ax, 'freq', None)
 
-            if isinstance(freq, DateOffset):
-                freq = freq.rule_code
-
             data = DataFrame(data.values,
                              index=data.index.to_period(freq=freq),
                              columns=data.columns)
@@ -699,7 +709,7 @@ def _post_plot_logic(self):
             else:
                 self.axes[0].legend(loc='best')
 
-        condition = (not self.has_ts_index
+        condition = (not self._use_dynamic_x
                      and df.index.is_all_dates
                      and not self.subplots
                      or (self.subplots and self.sharex))
diff --git a/pandas/tseries/plotting.py b/pandas/tseries/plotting.py
index 7f4b4e5130c66..2e0506e6c5fc2 100644
--- a/pandas/tseries/plotting.py
+++ b/pandas/tseries/plotting.py
@@ -43,32 +43,30 @@ def tsplot(series, plotf, **kwargs):
 
     """
     # Used inferred freq is possible, need a test case for inferred
-    freq = getattr(series.index, 'freq', None)
-    if freq is None:
-        freq = getattr(series.index, 'inferred_freq', None)
-
     if 'ax' in kwargs:
         ax = kwargs.pop('ax')
     else:
         import matplotlib.pyplot as plt
         ax = plt.gca()
 
-    if freq is None:
-        freq = getattr(ax, 'freq', None)
-
-    if isinstance(freq, DateOffset):
-        freq = freq.rule_code
+    freq = _get_freq(ax, series)
+    # resample against axes freq if necessary
+    if freq is None: # pragma: no cover
+        raise ValueError('Cannot use dynamic axis without frequency info')
     else:
-        freq = frequencies.get_base_alias(freq)
+        ax_freq = getattr(ax, 'freq', None)
+        if (ax_freq is not None) and (freq != ax_freq):
+            if frequencies.is_subperiod(freq, ax_freq): # downsample
+                how = kwargs.pop('how', 'first')
+                series = series.resample(ax_freq, how=how)
+            elif frequencies.is_superperiod(freq, ax_freq):
+                series = series.resample(ax_freq)
+            freq = ax_freq
 
-    freq = frequencies.get_period_alias(freq)
     # Convert DatetimeIndex to PeriodIndex
     if isinstance(series.index, DatetimeIndex):
         series = series.to_period(freq=freq)
 
-    if freq != series.index.freq:
-        series = series.asfreq(freq)
-
     style = kwargs.pop('style', None)
 
     # Specialized ts plotting attributes for Axes
@@ -81,13 +79,7 @@ def tsplot(series, plotf, **kwargs):
     ax.date_axis_info = None
 
     # format args and lot
-    mask = isnull(series)
-    if mask.any():
-        masked_array = np.ma.array(series.values)
-        masked_array = np.ma.masked_where(mask, masked_array)
-        args = [series.index, masked_array]
-    else:
-        args = [series.index, series]
+    args = _maybe_mask(series)
 
     if style is not None:
         args.append(style)
@@ -101,6 +93,38 @@ def tsplot(series, plotf, **kwargs):
 
     return ax
 
+def _maybe_mask(series):
+    mask = isnull(series)
+    if mask.any():
+        masked_array = np.ma.array(series.values)
+        masked_array = np.ma.masked_where(mask, masked_array)
+        args = [series.index, masked_array]
+    else:
+        args = [series.index, series]
+    return args
+
+def _get_freq(ax, series):
+    # get frequency from data
+    freq = getattr(series.index, 'freq', None)
+    if freq is None:
+        freq = getattr(series.index, 'inferred_freq', None)
+
+    ax_freq = getattr(ax, 'freq', None)
+
+    # use axes freq if no data freq
+    if freq is None:
+        freq = ax_freq
+
+    # get the period frequency
+    if isinstance(freq, DateOffset):
+        freq = freq.rule_code
+    else:
+        freq = frequencies.get_base_alias(freq)
+
+    freq = frequencies.get_period_alias(freq)
+
+    return freq
+
 def _get_xlim(lines):
     left, right = np.inf, -np.inf
     for l in lines:
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index 9331d79efe15a..e0a8acacf2df3 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -6,6 +6,7 @@
 
 import numpy as np
 from numpy.testing.decorators import slow
+from numpy.testing import assert_array_equal
 
 from pandas import Index, Series, DataFrame, isnull, notnull
 
@@ -52,17 +53,22 @@ def setUp(self):
     def test_tsplot(self):
         from pandas.tseries.plotting import tsplot
         import matplotlib.pyplot as plt
+        plt.close('all')
+
         ax = plt.gca()
         ts = tm.makeTimeSeries()
         plot_ax = tsplot(ts, plt.Axes.plot)
         self.assert_(plot_ax == ax)
 
         f = lambda *args, **kwds: tsplot(s, plt.Axes.plot, *args, **kwds)
+        plt.close('all')
 
         for s in self.period_ser:
             _check_plot_works(f, s.index.freq, ax=ax, series=s)
+            plt.close('all')
         for s in self.datetime_ser:
             _check_plot_works(f, s.index.freq.rule_code, ax=ax, series=s)
+            plt.close('all')
 
         plt.close('all')
         ax = ts.plot(style='k')
@@ -442,7 +448,7 @@ def test_secondary_frame(self):
         self.assert_(axes[2].get_yaxis().get_ticks_position() == 'right')
 
     @slow
-    def test_mixed_freq(self):
+    def test_mixed_freq_regular_first(self):
         import matplotlib.pyplot as plt
         plt.close('all')
         s1 = tm.makeTimeSeries()
@@ -458,6 +464,58 @@ def test_mixed_freq(self):
         pidx = s1.index.to_period()
         self.assert_(left == pidx[0].ordinal)
         self.assert_(right == pidx[-1].ordinal)
+        plt.close('all')
+
+    @slow
+    def test_mixed_freq_irregular_first(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        s1 = tm.makeTimeSeries()
+        s2 = s1[[0, 5, 10, 11, 12, 13, 14, 15]]
+        s2.plot(style='g')
+        ax = s1.plot()
+        self.assert_(not hasattr(ax, 'freq'))
+        lines = ax.get_lines()
+        x1 = lines[0].get_xdata()
+        assert_array_equal(x1, s2.index.asobject.values)
+        x2 = lines[1].get_xdata()
+        assert_array_equal(x2, s1.index.asobject.values)
+        plt.close('all')
+
+    @slow
+    def test_mixed_freq_hf_first(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        idxh = date_range('1/1/1999', periods=365, freq='D')
+        idxl = date_range('1/1/1999', periods=12, freq='M')
+        high = Series(np.random.randn(len(idxh)), idxh)
+        low = Series(np.random.randn(len(idxl)), idxl)
+        high.plot()
+        ax = low.plot()
+        for l in ax.get_lines():
+            self.assert_(l.get_xdata().freq == 'D')
+
+    @slow
+    def test_mixed_freq_lf_first(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        idxh = date_range('1/1/1999', periods=365, freq='D')
+        idxl = date_range('1/1/1999', periods=12, freq='M')
+        high = Series(np.random.randn(len(idxh)), idxh)
+        low = Series(np.random.randn(len(idxl)), idxl)
+        low.plot()
+        ax = high.plot()
+        for l in ax.get_lines():
+            self.assert_(l.get_xdata().freq == 'M')
+
+    @slow
+    def test_mixed_freq_irreg_period(self):
+        ts = tm.makeTimeSeries()
+        irreg = ts[[0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 16, 17, 18, 29]]
+        rng = period_range('1/3/2000', periods=30, freq='B')
+        ps = Series(np.random.randn(len(rng)), rng)
+        irreg.plot()
+        ps.plot()
 
 PNG_PATH = 'tmp.png'
 def _check_plot_works(f, freq=None, series=None, *args, **kwargs):
@@ -466,19 +524,22 @@ def _check_plot_works(f, freq=None, series=None, *args, **kwargs):
     fig = plt.gcf()
     plt.clf()
     ax = fig.add_subplot(211)
+    orig_ax = kwargs.pop('ax', plt.gca())
+    orig_axfreq = getattr(orig_ax, 'freq', None)
+
     ret = f(*args, **kwargs)
     assert(ret is not None)  # do something more intelligent
 
-    orig_ax = kwargs.pop('ax', plt.gca())
-    if series is not None: # non-business
+    ax = kwargs.pop('ax', plt.gca())
+    if series is not None:
         dfreq = series.index.freq
         if isinstance(dfreq, DateOffset):
             dfreq = dfreq.rule_code
-        #dfreq = frequencies.offset_to_period_alias(dfreq)
-        assert(orig_ax.freq == dfreq)
+        if orig_axfreq is None:
+            assert(ax.freq == dfreq)
 
-    if freq is not None:
-        assert(orig_ax.freq == freq)
+    if freq is not None and orig_axfreq is None:
+        assert(ax.freq == freq)
 
     ax = fig.add_subplot(212)
     try:

From 07e4ecd5c99ba799eebc3e31338e81780eb616b7 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 15:10:21 -0400
Subject: [PATCH 25/42] ENH: handle weekly resampling via daily

---
 pandas/tseries/plotting.py            |  6 +++++-
 pandas/tseries/tests/test_plotting.py | 26 ++++++++++++++++++++++++++
 2 files changed, 31 insertions(+), 1 deletion(-)

diff --git a/pandas/tseries/plotting.py b/pandas/tseries/plotting.py
index 2e0506e6c5fc2..3e3540271153a 100644
--- a/pandas/tseries/plotting.py
+++ b/pandas/tseries/plotting.py
@@ -57,10 +57,14 @@ def tsplot(series, plotf, **kwargs):
         ax_freq = getattr(ax, 'freq', None)
         if (ax_freq is not None) and (freq != ax_freq):
             if frequencies.is_subperiod(freq, ax_freq): # downsample
-                how = kwargs.pop('how', 'first')
+                how = kwargs.pop('how', 'last')
                 series = series.resample(ax_freq, how=how)
             elif frequencies.is_superperiod(freq, ax_freq):
                 series = series.resample(ax_freq)
+            else: # one freq is weekly
+                how = kwargs.pop('how', 'last')
+                series = series.resample('D', how=how, fill_method='pad')
+                series = series.resample(ax_freq, how=how, fill_method='pad')
             freq = ax_freq
 
     # Convert DatetimeIndex to PeriodIndex
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index e0a8acacf2df3..718249330ec92 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -517,6 +517,32 @@ def test_mixed_freq_irreg_period(self):
         irreg.plot()
         ps.plot()
 
+    @slow
+    def test_to_weekly_resampling(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        idxh = date_range('1/1/1999', periods=52, freq='W')
+        idxl = date_range('1/1/1999', periods=12, freq='M')
+        high = Series(np.random.randn(len(idxh)), idxh)
+        low = Series(np.random.randn(len(idxl)), idxl)
+        high.plot()
+        ax = low.plot()
+        for l in ax.get_lines():
+            self.assert_(l.get_xdata().freq.startswith('W'))
+
+    @slow
+    def test_from_weekly_resampling(self):
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        idxh = date_range('1/1/1999', periods=52, freq='W')
+        idxl = date_range('1/1/1999', periods=12, freq='M')
+        high = Series(np.random.randn(len(idxh)), idxh)
+        low = Series(np.random.randn(len(idxl)), idxl)
+        low.plot()
+        ax = high.plot()
+        for l in ax.get_lines():
+            self.assert_(l.get_xdata().freq == 'M')
+
 PNG_PATH = 'tmp.png'
 def _check_plot_works(f, freq=None, series=None, *args, **kwargs):
     import matplotlib.pyplot as plt

From 791e7042b43b7a4246b9736d5c8cd1d7f5a4df1b Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 19:39:08 -0400
Subject: [PATCH 26/42] ENH: column aliases for to_csv/to_excel #921

---
 pandas/core/frame.py       | 31 ++++++++++++++++++++++---------
 pandas/tests/test_frame.py | 19 +++++++++++++++++++
 2 files changed, 41 insertions(+), 9 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index b51fab43c9e33..c8cbed8fee44f 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -1024,7 +1024,8 @@ def to_panel(self):
 
     to_wide = deprecate('to_wide', to_panel)
 
-    def _helper_csvexcel(self, writer, na_rep=None, cols=None, header=True,
+    def _helper_csvexcel(self, writer, na_rep=None, cols=None,
+                         col_aliases=None, header=True,
                          index=True, index_label=None):
         if cols is None:
             cols = self.columns
@@ -1053,7 +1054,15 @@ def _helper_csvexcel(self, writer, na_rep=None, cols=None, header=True,
                     index_label = [index_label]
 
                 encoded_labels = list(index_label)
-                encoded_cols = list(cols)
+                if col_aliases is not None:
+                    if len(col_aliases) != len(cols):
+                        raise ValueError(('Writing %d cols but got %d aliases'
+                                          % (len(cols), len(col_aliases))))
+                    else:
+                        write_cols = col_aliases
+                else:
+                    write_cols = cols
+                encoded_cols = list(write_cols)
 
                 writer.writerow(encoded_labels + encoded_cols)
             else:
@@ -1078,8 +1087,8 @@ def _helper_csvexcel(self, writer, na_rep=None, cols=None, header=True,
             writer.writerow(row_fields)
 
     def to_csv(self, path_or_buf, sep=",", na_rep='', cols=None,
-               header=True, index=True, index_label=None, mode='w',
-               nanRep=None, encoding=None):
+               col_aliases=None, header=True, index=True, index_label=None,
+               mode='w', nanRep=None, encoding=None):
         """
         Write DataFrame to a comma-separated values (csv) file
 
@@ -1091,6 +1100,8 @@ def to_csv(self, path_or_buf, sep=",", na_rep='', cols=None,
             Missing data representation
         cols : sequence, optional
             Columns to write
+        col_aliases : sequence, default None
+            Optional column aliases to be written instead of column names
         header : boolean, default True
             Write out column names
         index : boolean, default True
@@ -1126,14 +1137,16 @@ def to_csv(self, path_or_buf, sep=",", na_rep='', cols=None,
             else:
                 csvout = csv.writer(f, lineterminator='\n', delimiter=sep)
             self._helper_csvexcel(csvout, na_rep=na_rep, cols=cols,
-                                  header=header, index=index,
-                                  index_label=index_label)
+                                  col_aliases=col_aliases, header=header,
+                                  index=index, index_label=index_label)
+
         finally:
             if close:
                 f.close()
 
     def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
-                 cols=None, header=True, index=True, index_label=None):
+                 cols=None, col_aliases=None, header=True, index=True,
+                 index_label=None):
         """
         Write DataFrame to a excel sheet
 
@@ -1173,8 +1186,8 @@ def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
             need_save = True
         excel_writer.cur_sheet = sheet_name
         self._helper_csvexcel(excel_writer, na_rep=na_rep, cols=cols,
-                              header=header, index=index,
-                              index_label=index_label)
+                              col_aliases=col_aliases, header=header,
+                              index=index, index_label=index_label)
         if need_save:
             excel_writer.save()
 
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 15d465e8df966..273e5bbd75ab7 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -3100,6 +3100,16 @@ def test_to_csv_from_csv(self):
         result = DataFrame.from_csv(path, index_col=[0, 1, 2],
                                     parse_dates=False)
         assert_frame_equal(result, df)
+
+        # column aliases
+        col_aliases = Index(['AA', 'X', 'Y', 'Z'])
+        self.frame2.to_csv(path, header=True, col_aliases=col_aliases)
+        rs = DataFrame.from_csv(path)
+        xp = self.frame2.copy()
+        xp.columns = col_aliases
+
+        assert_frame_equal(xp, rs)
+
         os.remove(path)
 
     def test_to_csv_multiindex(self):
@@ -3284,6 +3294,15 @@ def test_to_excel_from_excel(self):
             np.testing.assert_equal('test1', reader.sheet_names[0])
             np.testing.assert_equal('test2', reader.sheet_names[1])
 
+            # column aliases
+            col_aliases = Index(['AA', 'X', 'Y', 'Z'])
+            self.frame2.to_excel(path, 'test1', header=True,
+                                 col_aliases=col_aliases)
+            reader = ExcelFile(path)
+            rs = reader.parse('test1')
+            xp = self.frame2.copy()
+            xp.columns = col_aliases
+            assert_frame_equal(xp, rs)
 
             os.remove(path)
 

From 7923fa5466320551269cbfb4cd4492c135a553cd Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Mon, 25 Jun 2012 10:07:12 -0400
Subject: [PATCH 27/42] overload header keyword instead of extra col_aliases
 keyword

---
 pandas/core/frame.py       | 40 +++++++++++++++++++-------------------
 pandas/tests/test_frame.py | 10 ++++++----
 2 files changed, 26 insertions(+), 24 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index c8cbed8fee44f..a517df708fdd4 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -1025,15 +1025,16 @@ def to_panel(self):
     to_wide = deprecate('to_wide', to_panel)
 
     def _helper_csvexcel(self, writer, na_rep=None, cols=None,
-                         col_aliases=None, header=True,
-                         index=True, index_label=None):
+                         header=True, index=True, index_label=None):
         if cols is None:
             cols = self.columns
 
         series = {}
         for k, v in self._series.iteritems():
             series[k] = v.values
-        if header:
+
+        has_aliases = isinstance(header, (tuple, list, np.ndarray))
+        if has_aliases or header:
             if index:
                 # should write something for index label
                 if index_label is None:
@@ -1054,12 +1055,12 @@ def _helper_csvexcel(self, writer, na_rep=None, cols=None,
                     index_label = [index_label]
 
                 encoded_labels = list(index_label)
-                if col_aliases is not None:
-                    if len(col_aliases) != len(cols):
+                if has_aliases:
+                    if len(header) != len(cols):
                         raise ValueError(('Writing %d cols but got %d aliases'
-                                          % (len(cols), len(col_aliases))))
+                                          % (len(cols), len(header))))
                     else:
-                        write_cols = col_aliases
+                        write_cols = header
                 else:
                     write_cols = cols
                 encoded_cols = list(write_cols)
@@ -1087,7 +1088,7 @@ def _helper_csvexcel(self, writer, na_rep=None, cols=None,
             writer.writerow(row_fields)
 
     def to_csv(self, path_or_buf, sep=",", na_rep='', cols=None,
-               col_aliases=None, header=True, index=True, index_label=None,
+               header=True, index=True, index_label=None,
                mode='w', nanRep=None, encoding=None):
         """
         Write DataFrame to a comma-separated values (csv) file
@@ -1100,10 +1101,9 @@ def to_csv(self, path_or_buf, sep=",", na_rep='', cols=None,
             Missing data representation
         cols : sequence, optional
             Columns to write
-        col_aliases : sequence, default None
-            Optional column aliases to be written instead of column names
-        header : boolean, default True
-            Write out column names
+        header : boolean or list of string, default True
+            Write out column names. If a list of string is given it is
+            assumed to be aliases for the column names
         index : boolean, default True
             Write row names (index)
         index_label : string or sequence, default None
@@ -1137,16 +1137,15 @@ def to_csv(self, path_or_buf, sep=",", na_rep='', cols=None,
             else:
                 csvout = csv.writer(f, lineterminator='\n', delimiter=sep)
             self._helper_csvexcel(csvout, na_rep=na_rep, cols=cols,
-                                  col_aliases=col_aliases, header=header,
-                                  index=index, index_label=index_label)
+                                  header=header, index=index,
+                                  index_label=index_label)
 
         finally:
             if close:
                 f.close()
 
     def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
-                 cols=None, col_aliases=None, header=True, index=True,
-                 index_label=None):
+                 cols=None, header=True, index=True, index_label=None):
         """
         Write DataFrame to a excel sheet
 
@@ -1160,8 +1159,9 @@ def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
             Missing data rep'n
         cols : sequence, optional
             Columns to write
-        header : boolean, default True
-            Write out column names
+        header : boolean or list of string, default True
+            Write out column names. If a list of string is given it is
+            assumed to be aliases for the column names
         index : boolean, default True
             Write row names (index)
         index_label : string or sequence, default None
@@ -1186,8 +1186,8 @@ def to_excel(self, excel_writer, sheet_name='sheet1', na_rep='',
             need_save = True
         excel_writer.cur_sheet = sheet_name
         self._helper_csvexcel(excel_writer, na_rep=na_rep, cols=cols,
-                              col_aliases=col_aliases, header=header,
-                              index=index, index_label=index_label)
+                              header=header, index=index,
+                              index_label=index_label)
         if need_save:
             excel_writer.save()
 
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 273e5bbd75ab7..8c1c2169ed0d7 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -3103,13 +3103,16 @@ def test_to_csv_from_csv(self):
 
         # column aliases
         col_aliases = Index(['AA', 'X', 'Y', 'Z'])
-        self.frame2.to_csv(path, header=True, col_aliases=col_aliases)
+        self.frame2.to_csv(path, header=col_aliases)
         rs = DataFrame.from_csv(path)
         xp = self.frame2.copy()
         xp.columns = col_aliases
 
         assert_frame_equal(xp, rs)
 
+        self.assertRaises(ValueError, self.frame2.to_csv, path,
+                          header=['AA', 'X'])
+
         os.remove(path)
 
     def test_to_csv_multiindex(self):
@@ -3296,10 +3299,9 @@ def test_to_excel_from_excel(self):
 
             # column aliases
             col_aliases = Index(['AA', 'X', 'Y', 'Z'])
-            self.frame2.to_excel(path, 'test1', header=True,
-                                 col_aliases=col_aliases)
+            self.frame2.to_excel(path, 'test1', header=col_aliases)
             reader = ExcelFile(path)
-            rs = reader.parse('test1')
+            rs = reader.parse('test1', index_col=0)
             xp = self.frame2.copy()
             xp.columns = col_aliases
             assert_frame_equal(xp, rs)

From 0d91eda8edd6a7ce73c2414eee0137be107c8a38 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 12:23:04 -0400
Subject: [PATCH 28/42] DOC: release notes re: #921

---
 RELEASE.rst | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/RELEASE.rst b/RELEASE.rst
index 79b980694860a..af0d06caa1825 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -132,6 +132,8 @@ pandas 0.8.0
     404s (#826)
   - Improve performance of reshaping by using O(N) categorical sorting
   - Series names will be used for index of DataFrame if no index passed (#1494)
+  - Header argument in DataFrame.to_csv can accept a list of column names to
+    use instead of the object's columns (#921)
 
 **API Changes**
 

From 0b30cacaa643e0bc911d676d96924292fd879692 Mon Sep 17 00:00:00 2001
From: Skipper Seabold <jsseabold@gmail.com>
Date: Mon, 25 Jun 2012 09:59:52 -0400
Subject: [PATCH 29/42] ENH: Add raise on conflict keyword to update

---
 pandas/core/frame.py       | 15 ++++++++++++++-
 pandas/tests/test_frame.py | 11 +++++++++++
 2 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index a517df708fdd4..9a3c47302707f 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -166,6 +166,9 @@
 merged : DataFrame
 """
 
+# Custom error class for update
+
+class DataConflictError(Exception): pass
 
 #----------------------------------------------------------------------
 # Factory helper methods
@@ -3168,7 +3171,8 @@ def combine_first(self, other):
         combiner = lambda x, y: np.where(isnull(x), y, x)
         return self.combine(other, combiner)
 
-    def update(self, other, join='left', overwrite=True, filter_func=None):
+    def update(self, other, join='left', overwrite=True, filter_func=None,
+                     raise_conflict=False):
         """
         Modify DataFrame in place using non-NA values from passed
         DataFrame. Aligns on indices
@@ -3182,6 +3186,9 @@ def update(self, other, join='left', overwrite=True, filter_func=None):
         filter_func : callable(1d-array) -> 1d-array<boolean>, default None
             Can choose to replace values other than NA. Return True for values
             that should be updated
+        raise_conflict : bool
+            If True, will raise an error if the DataFrame and other both
+            contain data in the same place.
         """
         if join != 'left':
             raise NotImplementedError
@@ -3193,6 +3200,12 @@ def update(self, other, join='left', overwrite=True, filter_func=None):
             if filter_func is not None:
                 mask = -filter_func(this) | isnull(that)
             else:
+                if raise_conflict:
+                    mask_this = notnull(that)
+                    mask_that = notnull(this)
+                    if any(mask_this & mask_that):
+                        raise DataConflictError("Data overlaps.")
+
                 if overwrite:
                     mask = isnull(that)
                 else:
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 8c1c2169ed0d7..6ccd13911d24f 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -5347,6 +5347,17 @@ def test_update_filtered(self):
                               [1.5, nan, 7.]])
         assert_frame_equal(df, expected)
 
+    def test_update_raise(self):
+        df = DataFrame([[1.5, 1, 3.],
+                        [1.5, nan, 3.],
+                        [1.5, nan, 3],
+                        [1.5, nan, 3]])
+
+        other = DataFrame([[2., nan],
+                           [nan, 7]], index=[1, 3], columns=[1,2])
+
+        np.testing.assert_raises(Exception, df.update, *(other,),
+                **{'raise_conflict' : True})
 
     def test_combineAdd(self):
         # trivial

From 7a233cd402590a42b3f8ede7c679f8d06fb9b6ef Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 12:35:53 -0400
Subject: [PATCH 30/42] DOC: release notes

---
 RELEASE.rst | 1 +
 1 file changed, 1 insertion(+)

diff --git a/RELEASE.rst b/RELEASE.rst
index af0d06caa1825..d9e5f6abe4119 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -134,6 +134,7 @@ pandas 0.8.0
   - Series names will be used for index of DataFrame if no index passed (#1494)
   - Header argument in DataFrame.to_csv can accept a list of column names to
     use instead of the object's columns (#921)
+  - Add ``raise_conflict`` argument to DataFrame.update (#1526)
 
 **API Changes**
 

From 10284c15efa42aebbf9b2972e4e2f6c75a807126 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Mon, 25 Jun 2012 14:00:07 -0400
Subject: [PATCH 31/42] BUG: timedelta.total_seconds only in 2.7 and 3.2

---
 pandas/tseries/offsets.py | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/pandas/tseries/offsets.py b/pandas/tseries/offsets.py
index 8f62c3304cb53..f35a85970136a 100644
--- a/pandas/tseries/offsets.py
+++ b/pandas/tseries/offsets.py
@@ -280,7 +280,6 @@ def freqstr(self):
         return fstr
 
     def _offset_str(self):
-
         def get_str(td):
             off_str = ''
             if td.days > 0:
@@ -302,10 +301,10 @@ def get_str(td):
             return off_str
 
         if isinstance(self.offset, timedelta):
-            tot_sec = self.offset.total_seconds()
-            if tot_sec > 0:
+            zero = timedelta(0, 0, 0)
+            if self.offset >= zero:
                 off_str = '+' + get_str(self.offset)
-            if tot_sec < 0:
+            else:
                 off_str = '-' + get_str(-self.offset)
             return off_str
         else:

From b225898728b708c22ee068ba499b052b692b3b7c Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Mon, 25 Jun 2012 14:26:43 -0400
Subject: [PATCH 32/42] BUG: DataFrame plotting with inferred freq

---
 pandas/tools/plotting.py              | 10 +++++++---
 pandas/tseries/tests/test_plotting.py | 16 ++++++++++++++++
 2 files changed, 23 insertions(+), 3 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 30505cfc4d48c..985d58895d8f6 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -660,18 +660,22 @@ def _maybe_convert_index(self, data):
         if (isinstance(data.index, DatetimeIndex) and
             isinstance(data, DataFrame)):
             freq = getattr(data.index, 'freqstr', None)
+
+            if freq is None:
+                freq = getattr(data.index, 'inferred_freq', None)
+
             if isinstance(freq, DateOffset):
                 freq = freq.rule_code
 
             freq = get_period_alias(freq)
 
-            if freq is None:
-                freq = getattr(data.index, 'inferred_freq', None)
-
             if freq is None:
                 ax, _ = self._get_ax_and_style(0)
                 freq = getattr(ax, 'freq', None)
 
+            if freq is None:
+                raise ValueError('Could not get frequency alias for plotting')
+
             data = DataFrame(data.values,
                              index=data.index.to_period(freq=freq),
                              columns=data.columns)
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index 718249330ec92..c88102a31c8f4 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -49,6 +49,22 @@ def setUp(self):
                                     columns=['A', 'B', 'C'])
                                     for x in idx]
 
+    @slow
+    def test_frame_inferred(self):
+        # inferred freq
+        import matplotlib.pyplot as plt
+        plt.close('all')
+        idx = date_range('1/1/1987', freq='MS', periods=100)
+        idx = DatetimeIndex(idx.values, freq=None)
+        df = DataFrame(np.random.randn(len(idx), 3), index=idx)
+        df.plot()
+
+        # axes freq
+        idx = idx[0:40] + idx[45:99]
+        df2 = DataFrame(np.random.randn(len(idx), 3), index=idx)
+        df2.plot()
+        plt.close('all')
+
     @slow
     def test_tsplot(self):
         from pandas.tseries.plotting import tsplot

From 483348be900785905e3f6be195dfdce7c0571a44 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Mon, 25 Jun 2012 14:49:19 -0400
Subject: [PATCH 33/42] BUG: plotting DataFrame with freq with offset

---
 pandas/tools/plotting.py              | 2 +-
 pandas/tseries/tests/test_plotting.py | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 985d58895d8f6..37b3f6e641202 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -659,7 +659,7 @@ def _maybe_convert_index(self, data):
         from pandas.core.frame import DataFrame
         if (isinstance(data.index, DatetimeIndex) and
             isinstance(data, DataFrame)):
-            freq = getattr(data.index, 'freqstr', None)
+            freq = getattr(data.index, 'freq', None)
 
             if freq is None:
                 freq = getattr(data.index, 'inferred_freq', None)
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index c88102a31c8f4..6e246fb8ed0ec 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -42,7 +42,7 @@ def setUp(self):
                                     columns=['A', 'B', 'C'])
                                     for x in idx]
 
-        freq = ['S', 'T', 'H', 'D', 'W', 'M', 'Q-DEC', 'A']
+        freq = ['S', 'T', 'H', 'D', 'W', 'M', 'Q-DEC', 'A', '1B30Min']
         idx = [date_range('12/31/1999', freq=x, periods=100) for x in freq]
         self.datetime_ser = [Series(np.random.randn(len(x)), x) for x in idx]
         self.datetime_df = [DataFrame(np.random.randn(len(x), 3), index=x,

From 6832a94e2c87267a0ae5d19e4ceddd5950a650b6 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 17:44:34 -0400
Subject: [PATCH 34/42] ENH: experimental lreshape function

---
 pandas/core/api.py           |  3 +-
 pandas/core/common.py        | 12 +++++
 pandas/core/reshape.py       | 51 +++++++++++++++++++++
 pandas/tests/test_reshape.py | 88 +++++++++++++++++++++++++++++++++++-
 4 files changed, 151 insertions(+), 3 deletions(-)

diff --git a/pandas/core/api.py b/pandas/core/api.py
index df164575e3b04..8cf3b7f4cbda4 100644
--- a/pandas/core/api.py
+++ b/pandas/core/api.py
@@ -15,7 +15,8 @@
 from pandas.core.frame import DataFrame
 from pandas.core.panel import Panel
 from pandas.core.groupby import groupby
-from pandas.core.reshape import pivot_simple as pivot, get_dummies
+from pandas.core.reshape import (pivot_simple as pivot, get_dummies,
+                                 lreshape)
 
 WidePanel = Panel
 
diff --git a/pandas/core/common.py b/pandas/core/common.py
index 7d94e55c6d597..4db15c6c69101 100644
--- a/pandas/core/common.py
+++ b/pandas/core/common.py
@@ -914,3 +914,15 @@ def writerow(self, row):
             self.stream.write(data)
             # empty queue
             self.queue.truncate(0)
+
+
+_NS_DTYPE = np.dtype('M8[ns]')
+
+def _concat_compat(to_concat):
+    if all(x.dtype == _NS_DTYPE for x in to_concat):
+        # work around NumPy 1.6 bug
+        new_values = np.concatenate([x.view(np.int64) for x in to_concat])
+        return new_values.view(_NS_DTYPE)
+    else:
+        return np.concatenate(to_concat)
+
diff --git a/pandas/core/reshape.py b/pandas/core/reshape.py
index 08d45ddf63bd5..8703de4b31c0d 100644
--- a/pandas/core/reshape.py
+++ b/pandas/core/reshape.py
@@ -12,6 +12,7 @@
 from pandas.core.common import notnull, _ensure_platform_int
 from pandas.core.groupby import (get_group_index, _compress_group_index,
                                  decons_group_index)
+import pandas.core.common as com
 import pandas.lib as lib
 
 
@@ -559,6 +560,56 @@ def melt(frame, id_vars=None, value_vars=None):
     mdata['variable'] = np.asarray(frame.columns).repeat(N)
     return DataFrame(mdata, columns=mcolumns)
 
+
+def lreshape(data, groups, dropna=True, label=None):
+    """
+    Reshape long-format data to wide. Generalized inverse of DataFrame.pivot
+
+    Parameters
+    ----------
+    data :
+    groups : dict
+    na_action :
+
+    Returns
+    -------
+    reshaped : DataFrame
+    """
+    if isinstance(groups, dict):
+        keys = groups.keys()
+        values = groups.values()
+    else:
+        keys, values = zip(*groups)
+
+    all_cols = list(set.union(*[set(x) for x in values]))
+    id_cols = list(data.columns.diff(all_cols))
+
+    K = len(values[0])
+
+    for seq in values:
+        if len(seq) != K:
+            raise ValueError('All column lists must be same length')
+
+    mdata = {}
+    pivot_cols = []
+
+    for target, names in zip(keys, values):
+        mdata[target] = com._concat_compat([data[col].values for col in names])
+        pivot_cols.append(target)
+
+    for col in id_cols:
+        mdata[col] = np.tile(data[col].values, K)
+
+    if dropna:
+        mask = np.ones(len(mdata[pivot_cols[0]]), dtype=bool)
+        for c in pivot_cols:
+            mask &= notnull(mdata[c])
+        if not mask.all():
+            mdata = dict((k, v[mask]) for k, v in mdata.iteritems())
+
+    return DataFrame(mdata, columns=id_cols + pivot_cols)
+
+
 def convert_dummies(data, cat_variables, prefix_sep='_'):
     """
     Compute DataFrame with specified columns converted to dummy variables (0 /
diff --git a/pandas/tests/test_reshape.py b/pandas/tests/test_reshape.py
index 036ee8677b41c..e37fd13b5758b 100644
--- a/pandas/tests/test_reshape.py
+++ b/pandas/tests/test_reshape.py
@@ -1,8 +1,20 @@
+# pylint: disable-msg=W0612,E1101
+from copy import deepcopy
+from datetime import datetime, timedelta
+from StringIO import StringIO
+import cPickle as pickle
+import operator
+import os
+import unittest
+
+import nose
+
 from pandas import DataFrame
 
+from numpy import nan
 import numpy as np
 
-from pandas.core.reshape import melt, convert_dummies
+from pandas.core.reshape import melt, convert_dummies, lreshape
 import pandas.util.testing as tm
 
 def test_melt():
@@ -38,8 +50,80 @@ def test_convert_dummies():
     tm.assert_frame_equal(result, expected)
     tm.assert_frame_equal(result2, expected2)
 
+class Test_lreshape(unittest.TestCase):
+
+    def test_pairs(self):
+        data = {'birthdt': ['08jan2009', '20dec2008', '30dec2008',
+                            '21dec2008', '11jan2009'],
+                'birthwt': [1766, 3301, 1454, 3139, 4133],
+                'id': [101, 102, 103, 104, 105],
+                'sex': ['Male', 'Female', 'Female', 'Female', 'Female'],
+                'visitdt1': ['11jan2009', '22dec2008', '04jan2009',
+                             '29dec2008', '20jan2009'],
+                'visitdt2': ['21jan2009', nan, '22jan2009', '31dec2008', '03feb2009'],
+                'visitdt3': ['05feb2009', nan, nan, '02jan2009', '15feb2009'],
+                'wt1': [1823, 3338, 1549, 3298, 4306],
+                'wt2': [2011.0, nan, 1892.0, 3338.0, 4575.0],
+                'wt3': [2293.0, nan, nan, 3377.0, 4805.0]}
+
+        df = DataFrame(data)
+
+        spec = {'visitdt': ['visitdt%d' % i for i in range(1, 4)],
+                'wt': ['wt%d' % i for i in range(1, 4)]}
+        result = lreshape(df, spec)
+
+        exp_data = {'birthdt': ['08jan2009', '20dec2008', '30dec2008',
+                                '21dec2008', '11jan2009', '08jan2009',
+                                '30dec2008', '21dec2008', '11jan2009',
+                                '08jan2009', '21dec2008', '11jan2009'],
+                    'birthwt': [1766, 3301, 1454, 3139, 4133, 1766,
+                                1454, 3139, 4133, 1766, 3139, 4133],
+                    'id': [101, 102, 103, 104, 105, 101,
+                           103, 104, 105, 101, 104, 105],
+                    'sex': ['Male', 'Female', 'Female', 'Female', 'Female',
+                            'Male', 'Female', 'Female', 'Female', 'Male',
+                            'Female', 'Female'],
+                    'visitdt': ['11jan2009', '22dec2008', '04jan2009', '29dec2008',
+                                '20jan2009', '21jan2009', '22jan2009', '31dec2008',
+                                '03feb2009', '05feb2009', '02jan2009', '15feb2009'],
+                    'wt': [1823.0, 3338.0, 1549.0, 3298.0, 4306.0, 2011.0,
+                           1892.0, 3338.0, 4575.0, 2293.0, 3377.0, 4805.0]}
+        exp = DataFrame(exp_data, columns=result.columns)
+        tm.assert_frame_equal(result, exp)
+
+        result = lreshape(df, spec, dropna=False)
+        exp_data = {'birthdt': ['08jan2009', '20dec2008', '30dec2008',
+                                '21dec2008', '11jan2009',
+                                '08jan2009', '20dec2008', '30dec2008',
+                                '21dec2008', '11jan2009',
+                                '08jan2009', '20dec2008', '30dec2008',
+                                '21dec2008', '11jan2009'],
+                    'birthwt': [1766, 3301, 1454, 3139, 4133,
+                                1766, 3301, 1454, 3139, 4133,
+                                1766, 3301, 1454, 3139, 4133],
+                    'id': [101, 102, 103, 104, 105,
+                           101, 102, 103, 104, 105,
+                           101, 102, 103, 104, 105],
+                    'sex': ['Male', 'Female', 'Female', 'Female', 'Female',
+                            'Male', 'Female', 'Female', 'Female', 'Female',
+                            'Male', 'Female', 'Female', 'Female', 'Female'],
+                    'visitdt': ['11jan2009', '22dec2008', '04jan2009',
+                                '29dec2008', '20jan2009',
+                                '21jan2009', nan, '22jan2009',
+                                '31dec2008', '03feb2009',
+                                '05feb2009', nan, nan, '02jan2009', '15feb2009'],
+                    'wt': [1823.0, 3338.0, 1549.0, 3298.0, 4306.0, 2011.0,
+                           nan, 1892.0, 3338.0, 4575.0, 2293.0, nan, nan,
+                           3377.0, 4805.0]}
+        exp = DataFrame(exp_data, columns=result.columns)
+        tm.assert_frame_equal(result, exp)
+
+        spec = {'visitdt': ['visitdt%d' % i for i in range(1, 3)],
+                'wt': ['wt%d' % i for i in range(1, 4)]}
+        self.assertRaises(ValueError, lreshape, df, spec)
+
+
 if __name__ == '__main__':
-    import nose
     nose.runmodule(argv=[__file__,'-vvs','-x','--pdb', '--pdb-failure'],
                    exit=False)
 

From 0ade7fa670981b8e11239c1b2c39dd44c7cf883d Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 18:05:16 -0400
Subject: [PATCH 35/42] DOC: lreshape docstring, release note

---
 RELEASE.rst             |  1 +
 pandas/core/reshape.py  | 20 ++++++++++++++++++--
 pandas/tseries/index.py | 12 ++----------
 3 files changed, 21 insertions(+), 12 deletions(-)

diff --git a/RELEASE.rst b/RELEASE.rst
index d9e5f6abe4119..c5bb354c8bdc9 100644
--- a/RELEASE.rst
+++ b/RELEASE.rst
@@ -89,6 +89,7 @@ pandas 0.8.0
   - Add support for Categorical use in GroupBy (#292)
   - Add ``any`` and ``all`` methods to DataFrame (#1416)
   - Add ``secondary_y`` option to Series.plot
+  - Add experimental ``lreshape`` function for reshaping wide to long
 
 **Improvements to existing features**
 
diff --git a/pandas/core/reshape.py b/pandas/core/reshape.py
index 8703de4b31c0d..475a6822edfea 100644
--- a/pandas/core/reshape.py
+++ b/pandas/core/reshape.py
@@ -567,9 +567,25 @@ def lreshape(data, groups, dropna=True, label=None):
 
     Parameters
     ----------
-    data :
+    data : DataFrame
     groups : dict
-    na_action :
+        {new_name : list_of_columns}
+    dropna : boolean, default True
+
+    Examples
+    --------
+    >>> data
+       hr1  hr2     team  year1  year2
+    0  514  545  Red Sox   2007   2008
+    1  573  526  Yankees   2007   2008
+
+    >>> pd.lreshape(data, {'year': ['year1', 'year2'],
+                           'hr': ['hr1', 'hr2']})
+          team   hr  year
+    0  Red Sox  514  2007
+    1  Yankees  573  2007
+    2  Red Sox  545  2008
+    3  Yankees  526  2008
 
     Returns
     -------
diff --git a/pandas/tseries/index.py b/pandas/tseries/index.py
index 1c4af41ff90e3..ae212902fb494 100644
--- a/pandas/tseries/index.py
+++ b/pandas/tseries/index.py
@@ -484,7 +484,7 @@ def append(self, other):
         to_concat = [x.values if isinstance(x, Index) else x
                      for x in to_concat]
 
-        return Index(_concat(to_concat), name=name)
+        return Index(com._concat_compat(to_concat), name=name)
 
     def get_duplicates(self):
         values = Index.get_duplicates(self)
@@ -775,7 +775,7 @@ def _fast_union(self, other):
             if left_end < right_end:
                 loc = right.searchsorted(left_end, side='right')
                 right_chunk = right.values[loc:]
-                dates = _concat((left.values, right_chunk))
+                dates = com._concat_compat((left.values, right_chunk))
                 return self._view_like(dates)
             else:
                 return left
@@ -1371,11 +1371,3 @@ def _time_to_micros(time):
     return 1000000 * seconds + time.microsecond
 
 
-def _concat(to_concat):
-    if all(x.dtype == _NS_DTYPE for x in to_concat):
-        # work around NumPy 1.6 bug
-        new_values = np.concatenate([x.view('i8') for x in to_concat])
-        return new_values.view(_NS_DTYPE)
-    else:
-        return np.concatenate(to_concat)
-

From 3453d446d6e1755c807b6b293f020bb2ca5da81c Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Sun, 24 Jun 2012 16:34:19 -0400
Subject: [PATCH 36/42] BUG: workaround vstack/concat bug in numpy 1.6 #1518

---
 pandas/core/internals.py   | 10 +++++++++-
 pandas/tests/test_frame.py | 32 +++++++++++++++++++++++++++++++-
 2 files changed, 40 insertions(+), 2 deletions(-)

diff --git a/pandas/core/internals.py b/pandas/core/internals.py
index 83278e3c994d0..1966b51f5b2e7 100644
--- a/pandas/core/internals.py
+++ b/pandas/core/internals.py
@@ -1389,7 +1389,7 @@ def _consolidate(blocks, items):
 def _merge_blocks(blocks, items):
     if len(blocks) == 1:
         return blocks[0]
-    new_values = np.vstack([b.values for b in blocks])
+    new_values = _vstack([b.values for b in blocks])
     new_items = blocks[0].items.append([b.items for b in blocks[1:]])
     new_block = make_block(new_values, new_items, items,
                            do_integrity_check=True)
@@ -1422,3 +1422,11 @@ def _union_items_slow(all_items):
         else:
             seen = seen.union(items)
     return seen
+
+def _vstack(to_stack):
+    if all(x.dtype == _NS_DTYPE for x in to_stack):
+        # work around NumPy 1.6 bug
+        new_values = np.vstack([x.view('i8') for x in to_stack])
+        return new_values.view(_NS_DTYPE)
+    else:
+        return np.vstack(to_stack)
diff --git a/pandas/tests/test_frame.py b/pandas/tests/test_frame.py
index 6ccd13911d24f..e3794222d0481 100644
--- a/pandas/tests/test_frame.py
+++ b/pandas/tests/test_frame.py
@@ -13,6 +13,7 @@
 from numpy.random import randn
 import numpy as np
 import numpy.ma as ma
+from numpy.testing import assert_array_equal
 
 import pandas as pan
 import pandas.core.nanops as nanops
@@ -21,7 +22,7 @@
 import pandas.core.datetools as datetools
 from pandas.core.api import (DataFrame, Index, Series, notnull, isnull,
                              MultiIndex, DatetimeIndex)
-from pandas.io.parsers import (ExcelFile, ExcelWriter)
+from pandas.io.parsers import (ExcelFile, ExcelWriter, read_csv)
 
 from pandas.util.testing import (assert_almost_equal,
                                  assert_series_equal,
@@ -6393,6 +6394,35 @@ def test_any_all(self):
         self._check_bool_op('any', np.any, has_skipna=True, has_bool_only=True)
         self._check_bool_op('all', np.all, has_skipna=True, has_bool_only=True)
 
+    def test_consolidate_datetime64(self):
+        # numpy vstack bug
+
+        data = """\
+starting,ending,measure
+2012-06-21 00:00,2012-06-23 07:00,77
+2012-06-23 07:00,2012-06-23 16:30,65
+2012-06-23 16:30,2012-06-25 08:00,77
+2012-06-25 08:00,2012-06-26 12:00,0
+2012-06-26 12:00,2012-06-27 08:00,77
+"""
+        df = read_csv(StringIO(data), parse_dates=[0,1])
+
+        ser_starting = df.starting
+        ser_starting.index = ser_starting.values
+        ser_starting = ser_starting.tz_localize('US/Eastern')
+        ser_starting = ser_starting.tz_convert('UTC')
+
+        ser_ending = df.ending
+        ser_ending.index = ser_ending.values
+        ser_ending = ser_ending.tz_localize('US/Eastern')
+        ser_ending = ser_ending.tz_convert('UTC')
+
+        df.starting = ser_starting.index
+        df.ending = ser_ending.index
+
+        assert_array_equal(df.starting.values, ser_starting.index.values)
+        assert_array_equal(df.ending.values, ser_ending.index.values)
+
     def _check_bool_op(self, name, alternative, frame=None, has_skipna=True,
                        has_bool_only=False):
         if frame is None:

From eb7a751fa3396d1e194f319128be7dd4513512b7 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 18:32:55 -0400
Subject: [PATCH 37/42] BUG: repr of pre-1900 datetime64 values in a DataFrame
 column close #1518

---
 pandas/core/format.py                   | 14 +-------------
 pandas/src/datetime.pyx                 | 24 +++++++++++++++---------
 pandas/tests/test_reshape.py            |  1 -
 pandas/tseries/tests/test_timeseries.py |  6 ++++++
 4 files changed, 22 insertions(+), 23 deletions(-)

diff --git a/pandas/core/format.py b/pandas/core/format.py
index 59ab7b473d99a..776a0fd80291f 100644
--- a/pandas/core/format.py
+++ b/pandas/core/format.py
@@ -594,19 +594,7 @@ def _format_datetime64(x, tz=None):
         return 'NaT'
 
     stamp = lib.Timestamp(x, tz=tz)
-    base = stamp.strftime('%Y-%m-%d %H:%M:%S')
-
-    fraction = stamp.microsecond * 1000 + stamp.nanosecond
-    digits = 9
-
-    if fraction == 0:
-        return base
-
-    while (fraction % 10) == 0:
-        fraction /= 10
-        digits -= 1
-
-    return base + ('.%%.%id' % digits) % fraction
+    return stamp._repr_base
 
 
 def _make_fixed_width(strings, justify='right'):
diff --git a/pandas/src/datetime.pyx b/pandas/src/datetime.pyx
index 17f35a73501aa..f7e325c4ae718 100644
--- a/pandas/src/datetime.pyx
+++ b/pandas/src/datetime.pyx
@@ -102,15 +102,7 @@ class Timestamp(_Timestamp):
         return ts_base
 
     def __repr__(self):
-        result = '%d-%.2d-%.2d %.2d:%.2d:%.2d' % (self.year, self.month,
-                                                  self.day, self.hour,
-                                                  self.minute, self.second)
-
-        if self.nanosecond != 0:
-            nanos = self.nanosecond + 1000 * self.microsecond
-            result += '.%.9d' % nanos
-        elif self.microsecond != 0:
-            result += '.%.6d' % self.microsecond
+        result = self._repr_base
 
         try:
             result += self.strftime('%z')
@@ -124,6 +116,20 @@ class Timestamp(_Timestamp):
 
         return '<Timestamp: %s>' % result
 
+    @property
+    def _repr_base(self):
+        result = '%d-%.2d-%.2d %.2d:%.2d:%.2d' % (self.year, self.month,
+                                                  self.day, self.hour,
+                                                  self.minute, self.second)
+
+        if self.nanosecond != 0:
+            nanos = self.nanosecond + 1000 * self.microsecond
+            result += '.%.9d' % nanos
+        elif self.microsecond != 0:
+            result += '.%.6d' % self.microsecond
+
+        return result
+
     @property
     def tz(self):
         """
diff --git a/pandas/tests/test_reshape.py b/pandas/tests/test_reshape.py
index e37fd13b5758b..3cb600ff69513 100644
--- a/pandas/tests/test_reshape.py
+++ b/pandas/tests/test_reshape.py
@@ -126,4 +126,3 @@ def test_pairs(self):
 if __name__ == '__main__':
     nose.runmodule(argv=[__file__,'-vvs','-x','--pdb', '--pdb-failure'],
                    exit=False)
-
diff --git a/pandas/tseries/tests/test_timeseries.py b/pandas/tseries/tests/test_timeseries.py
index f04f0a5b8259c..5f6a00a27699f 100644
--- a/pandas/tseries/tests/test_timeseries.py
+++ b/pandas/tseries/tests/test_timeseries.py
@@ -384,6 +384,12 @@ def test_frame_add_datetime64_column(self):
         df['A'] = rng
         self.assert_(np.issubdtype(df['A'].dtype, np.dtype('M8[ns]')))
 
+    def test_frame_datetime64_pre1900_repr(self):
+        df = DataFrame({'year': date_range('1/1/1700', periods=50,
+                                           freq='A-DEC')})
+        # it works!
+        repr(df)
+
     def test_frame_add_datetime64_col_other_units(self):
         n = 100
 

From 462c731ece74ad0886188ac1f2c100baf4a15dc9 Mon Sep 17 00:00:00 2001
From: Wes McKinney <wesmckinn@gmail.com>
Date: Mon, 25 Jun 2012 18:34:47 -0400
Subject: [PATCH 38/42] DOC: small doc for #1450

---
 doc/source/io.rst | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

diff --git a/doc/source/io.rst b/doc/source/io.rst
index 50a90f601c573..8c67b7d96d395 100644
--- a/doc/source/io.rst
+++ b/doc/source/io.rst
@@ -83,8 +83,10 @@ data into a DataFrame object. They can take a number of arguments:
     as the index.
   - ``names``: List of column names to use. If passed, header will be
     implicitly set to None.
-  - ``na_values``: optional list of strings to recognize as NaN (missing values),
-    in addition to a default set.
+  - ``na_values``: optional list of strings to recognize as NaN (missing
+    values), in addition to a default set. If you pass an empty list or an
+    empty list for a particular column, no values (including empty strings)
+    will be considered NA
   - ``parse_dates``: if True then index will be parsed as dates
     (False by default). You can specify more complicated options to parse
     a subset of columns or a combination of columns into a single date column

From 4c860de3486e7410321c3403f655689aa73c4cc5 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Tue, 26 Jun 2012 11:52:07 -0400
Subject: [PATCH 39/42] ENH: handle datetime.date in Period constructor

---
 pandas/tseries/period.py            |  7 +++++--
 pandas/tseries/tests/test_period.py | 10 +++++++++-
 2 files changed, 14 insertions(+), 3 deletions(-)

diff --git a/pandas/tseries/period.py b/pandas/tseries/period.py
index 22e571acef416..0a8c54e1c257f 100644
--- a/pandas/tseries/period.py
+++ b/pandas/tseries/period.py
@@ -1,6 +1,5 @@
 # pylint: disable=E1101,E1103,W0232
-
-from datetime import datetime
+from datetime import datetime, date
 import numpy as np
 
 from pandas.tseries.frequencies import (get_freq_code as _gfc, to_offset,
@@ -104,6 +103,10 @@ def __init__(self, value=None, freq=None, ordinal=None,
             dt = value
             if freq is None:
                 raise ValueError('Must supply freq for datetime value')
+        elif isinstance(value, date):
+            dt = datetime(year=value.year, month=value.month, day=value.day)
+            if freq is None:
+                raise ValueError('Must supply freq for datetime value')
         else:
             msg = "Value must be Period, string, integer, or datetime"
             raise ValueError(msg)
diff --git a/pandas/tseries/tests/test_period.py b/pandas/tseries/tests/test_period.py
index c65615cd1d03e..fb09fec8bfd1d 100644
--- a/pandas/tseries/tests/test_period.py
+++ b/pandas/tseries/tests/test_period.py
@@ -7,7 +7,7 @@
 """
 
 from unittest import TestCase
-from datetime import datetime, timedelta
+from datetime import datetime, date, timedelta
 import unittest
 
 from numpy.ma.testutils import assert_equal
@@ -168,6 +168,14 @@ def test_period_constructor(self):
         i1 = Period(ordinal=200701, freq='M')
         self.assertEqual(i1.year, 18695)
 
+        i1 = Period(datetime(2007, 1, 1), freq='M')
+        i2 = Period('200701', freq='M')
+        self.assertEqual(i1, i2)
+
+        i1 = Period(date(2007, 1, 1), freq='M')
+        i2 = Period(datetime(2007, 1, 1), freq='M')
+        self.assertEqual(i1, i2)
+
         self.assertRaises(ValueError, Period, ordinal=200701)
 
     def test_freq_str(self):

From 14759715101b0d38dedebd2588c587f6f1c11aee Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Tue, 26 Jun 2012 11:57:34 -0400
Subject: [PATCH 40/42] ENH: register converters with matplotlib for better
 datetime convesion

---
 pandas/tools/plotting.py              | 53 +++++++++++++++++----------
 pandas/tseries/plotting.py            | 23 +++++++++---
 pandas/tseries/tests/test_plotting.py | 21 ++++++++++-
 3 files changed, 70 insertions(+), 27 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 37b3f6e641202..1f116529def2d 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -1,6 +1,7 @@
 # being a bit too dynamic
 # pylint: disable=E1101
 from itertools import izip
+import datetime
 
 import numpy as np
 
@@ -13,6 +14,7 @@
 from pandas.tseries.frequencies import get_period_alias, get_base_alias
 from pandas.tseries.offsets import DateOffset
 import pandas.tseries.tools as datetools
+import pandas.lib as lib
 
 def _get_standard_kind(kind):
     return {'density' : 'kde'}.get(kind, kind)
@@ -573,24 +575,39 @@ def _post_plot_logic(self):
         if self.subplots and self.legend:
             self.axes[0].legend(loc='best')
 
-class DatetimeConverter(object):
-
-    @classmethod
-    def convert(cls, values, units, axis):
-        def try_parse(values):
-            try:
-                return datetools.to_datetime(values).toordinal()
-            except Exception:
+try:
+    import matplotlib.units as units
+    import matplotlib.dates as dates
+
+    class DatetimeConverter(dates.DateConverter):
+
+        @staticmethod
+        def convert(values, unit, axis):
+            def try_parse(values):
+                try:
+                    return datetools.to_datetime(values).toordinal()
+                except Exception:
+                    return values
+
+            if isinstance(values, (datetime.datetime, datetime.date)):
+                return values.toordinal()
+            elif isinstance(values, (datetime.time)):
+                return dates.date2num(values)
+            elif (com.is_integer(values) or com.is_float(values)):
                 return values
-
-        if (com.is_integer(values) or
-            com.is_float(values)):
+            elif isinstance(values, str):
+                return try_parse(values)
+            elif isinstance(values, Index):
+                return values.map(try_parse)
+            elif isinstance(values, (list, tuple, np.ndarray)):
+                return [try_parse(x) for x in values]
             return values
-        elif isinstance(values, str):
-            return try_parse(values)
-        elif isinstance(values, Index):
-            return values.map(try_parse)
-        return map(try_parse, values)
+
+    units.registry[lib.Timestamp] = DatetimeConverter()
+    units.registry[datetime.date] = DatetimeConverter()
+    units.registry[datetime.datetime] = DatetimeConverter()
+except ImportError:
+    pass
 
 class LinePlot(MPLPlot):
 
@@ -648,10 +665,6 @@ def _make_plot(self):
                     y = np.ma.masked_where(mask, y)
                 plotf(ax, x, y, style, label=label, **self.kwds)
                 ax.grid(self.grid)
-                idx = getattr(self.data, 'index', None)
-                if isinstance(idx, DatetimeIndex) or (idx is not None and
-                    idx.inferred_type == 'datetime'):
-                    ax.get_xaxis().converter = DatetimeConverter
 
     def _maybe_convert_index(self, data):
         # tsplot converts automatically, but don't want to convert index
diff --git a/pandas/tseries/plotting.py b/pandas/tseries/plotting.py
index 3e3540271153a..152f677a5b03e 100644
--- a/pandas/tseries/plotting.py
+++ b/pandas/tseries/plotting.py
@@ -77,7 +77,6 @@ def tsplot(series, plotf, **kwargs):
     ax.freq = freq
     xaxis = ax.get_xaxis()
     xaxis.freq = freq
-    xaxis.converter = DateConverter
     ax.legendlabels = [kwargs.get('label', None)]
     ax.view_interval = None
     ax.date_axis_info = None
@@ -739,12 +738,24 @@ def format_dateaxis(subplot, freq):
     subplot.xaxis.set_minor_formatter(minformatter)
     pylab.draw_if_interactive()
 
-class DateConverter(object):
 
-    @classmethod
-    def convert(cls, values, units, axis):
-        if isinstance(values, (int, float, str, datetime, Period)):
+import matplotlib.units as units
+import matplotlib.dates as dates
+
+class PeriodConverter(dates.DateConverter):
+
+    @staticmethod
+    def convert(values, units, axis):
+        if not hasattr(axis, 'freq'):
+            raise TypeError('Axis must have `freq` set to convert to Periods')
+        valid_types = (str, datetime, Period, pydt.date, pydt.time)
+        if (isinstance(values, valid_types) or com.is_integer(values) or
+            com.is_float(values)):
             return get_datevalue(values, axis.freq)
         if isinstance(values, Index):
             return values.map(lambda x: get_datevalue(x, axis.freq))
-        return map(lambda x: get_datevalue(x, axis.freq), values)
+        if isinstance(values, (list, tuple, np.ndarray)):
+            return [get_datevalue(x, axis.freq) for x in values]
+        return values
+
+units.registry[Period] = PeriodConverter()
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index 6e246fb8ed0ec..58ffd2c8c4e38 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -1,5 +1,5 @@
 import os
-from datetime import datetime, timedelta
+from datetime import datetime, timedelta, date, time
 
 import unittest
 import nose
@@ -559,6 +559,25 @@ def test_from_weekly_resampling(self):
         for l in ax.get_lines():
             self.assert_(l.get_xdata().freq == 'M')
 
+    @slow
+    def test_irreg_dtypes(self):
+        #date
+        idx = [date(2000, 1, 1), date(2000, 1, 5), date(2000, 1, 20)]
+        df = DataFrame(np.random.randn(len(idx), 3), Index(idx, dtype=object))
+        _check_plot_works(df.plot)
+
+        #np.datetime64
+        idx = date_range('1/1/2000', periods=10)
+        idx = idx[[0, 2, 5, 9]].asobject
+        df = DataFrame(np.random.randn(len(idx), 3), idx)
+        _check_plot_works(df.plot)
+
+        #time
+        inc = Series(np.random.randint(1, 6, 9)).cumsum().values
+        idx = [time(1, 1, i) for i in inc]
+        df = DataFrame(np.random.randn(len(idx), 3), idx)
+        _check_plot_works(df.plot)
+
 PNG_PATH = 'tmp.png'
 def _check_plot_works(f, freq=None, series=None, *args, **kwargs):
     import matplotlib.pyplot as plt

From f89ced44269746125a06482d8947bde12fa373bc Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Tue, 26 Jun 2012 15:23:41 -0400
Subject: [PATCH 41/42] BUG: incorrect tick label positions #1531 (zooming is
 still wrong)

---
 pandas/tools/plotting.py              |  8 ++++++--
 pandas/tseries/tests/test_plotting.py | 11 +++++++++--
 2 files changed, 15 insertions(+), 4 deletions(-)

diff --git a/pandas/tools/plotting.py b/pandas/tools/plotting.py
index 1f116529def2d..e5ba885c8e28e 100644
--- a/pandas/tools/plotting.py
+++ b/pandas/tools/plotting.py
@@ -461,9 +461,12 @@ def _adorn_subplots(self):
                 self.ax.set_title(self.title)
 
         if self._need_to_set_index:
-            xticklabels = [_stringify(key) for key in self.data.index]
+            labels = [_stringify(key) for key in self.data.index]
+            labels = dict(zip(range(len(self.data.index)), labels))
+
             for ax_ in self.axes:
                 # ax_.set_xticks(self.xticks)
+                xticklabels = [labels.get(x, '') for x in ax_.get_xticks()]
                 ax_.set_xticklabels(xticklabels, rotation=self.rot)
 
     @property
@@ -575,7 +578,7 @@ def _post_plot_logic(self):
         if self.subplots and self.legend:
             self.axes[0].legend(loc='best')
 
-try:
+try: # matplotlib is optional dependency
     import matplotlib.units as units
     import matplotlib.dates as dates
 
@@ -663,6 +666,7 @@ def _make_plot(self):
                 if mask.any():
                     y = np.ma.array(y)
                     y = np.ma.masked_where(mask, y)
+
                 plotf(ax, x, y, style, label=label, **self.kwds)
                 ax.grid(self.grid)
 
diff --git a/pandas/tseries/tests/test_plotting.py b/pandas/tseries/tests/test_plotting.py
index 58ffd2c8c4e38..af399ee83e185 100644
--- a/pandas/tseries/tests/test_plotting.py
+++ b/pandas/tseries/tests/test_plotting.py
@@ -561,6 +561,7 @@ def test_from_weekly_resampling(self):
 
     @slow
     def test_irreg_dtypes(self):
+        import matplotlib.pyplot as plt
         #date
         idx = [date(2000, 1, 1), date(2000, 1, 5), date(2000, 1, 20)]
         df = DataFrame(np.random.randn(len(idx), 3), Index(idx, dtype=object))
@@ -573,10 +574,16 @@ def test_irreg_dtypes(self):
         _check_plot_works(df.plot)
 
         #time
-        inc = Series(np.random.randint(1, 6, 9)).cumsum().values
+        plt.close('all')
+        inc = Series(np.random.randint(1, 15, 3)).cumsum().values
         idx = [time(1, 1, i) for i in inc]
         df = DataFrame(np.random.randn(len(idx), 3), idx)
-        _check_plot_works(df.plot)
+        ax = df.plot()
+        ticks = ax.get_xticks()
+        labels = ax.get_xticklabels()
+        td = dict(zip(ticks, labels))
+        for i in range(3):
+            self.assert_(td[i].get_text() == str(idx[i]))
 
 PNG_PATH = 'tmp.png'
 def _check_plot_works(f, freq=None, series=None, *args, **kwargs):

From 76c63517c8d5437abfdfcce958213cce01897c86 Mon Sep 17 00:00:00 2001
From: Chang She <chang@lambdafoundry.com>
Date: Tue, 26 Jun 2012 20:21:46 -0400
Subject: [PATCH 42/42] BUG/TST: typo caused read_csv to lose index name #1536

---
 pandas/io/parsers.py            |  2 +-
 pandas/io/tests/test_parsers.py | 17 +++++++++++++++++
 2 files changed, 18 insertions(+), 1 deletion(-)

diff --git a/pandas/io/parsers.py b/pandas/io/parsers.py
index 3b2fedd65b45f..fe07c37a49f36 100644
--- a/pandas/io/parsers.py
+++ b/pandas/io/parsers.py
@@ -662,7 +662,7 @@ def _get_index_name(self, columns=None):
             index_name = None
         elif np.isscalar(self.index_col):
             if isinstance(self.index_col, basestring):
-                index_names = self.index_col
+                index_name = self.index_col
                 for i, c in enumerate(list(columns)):
                     if c == self.index_col:
                         self.index_col = i
diff --git a/pandas/io/tests/test_parsers.py b/pandas/io/tests/test_parsers.py
index 1f14df528af8f..84a3b165a0af8 100644
--- a/pandas/io/tests/test_parsers.py
+++ b/pandas/io/tests/test_parsers.py
@@ -334,6 +334,23 @@ def test_index_col_named(self):
         self.assertRaises(ValueError, read_csv, StringIO(no_header),
                           index_col='ID')
 
+        data = """\
+1,2,3,4,hello
+5,6,7,8,world
+9,10,11,12,foo
+"""
+        names = ['a', 'b', 'c', 'd', 'message']
+        xp = DataFrame({'a' : [1, 5, 9], 'b' : [2, 6, 10], 'c' : [3, 7, 11],
+                        'd' : [4, 8, 12]},
+                       index=Index(['hello', 'world', 'foo'], name='message'))
+        rs = read_csv(StringIO(data), names=names, index_col=['message'])
+        assert_frame_equal(xp, rs)
+        self.assert_(xp.index.name == rs.index.name)
+
+        rs = read_csv(StringIO(data), names=names, index_col='message')
+        assert_frame_equal(xp, rs)
+        self.assert_(xp.index.name == rs.index.name)
+
     def test_multiple_skts_example(self):
         data = "year, month, a, b\n 2001, 01, 0.0, 10.\n 2001, 02, 1.1, 11."
         pass