Merge branch 'master' of github.com:pydata/pandas into jcrist-period_dtime64

Author: cpcloud

ci/script.sh

@@ -16,6 +16,8 @@ fi
 "$TRAVIS_BUILD_DIR"/ci/build_docs.sh 2>&1 > /tmp/doc.log &
 # doc build log will be shown after tests
 
+pip install -U blosc  # See https://github.com/pydata/pandas/pull/9783
+python -c 'import blosc; blosc.print_versions()'
 
 echo nosetests --exe -A "$NOSE_ARGS" pandas --with-xunit --xunit-file=/tmp/nosetests.xml
 nosetests --exe -A "$NOSE_ARGS" pandas --with-xunit --xunit-file=/tmp/nosetests.xml

doc/source/indexing.rst

@@ -249,6 +249,14 @@ new column.
 If you are using the IPython environment, you may also use tab-completion to
 see these accessible attributes.
 
+You can also assign a ``dict`` to a row of a ``DataFrame``:
+
+.. ipython:: python
+
+   x = pd.DataFrame({'x': [1, 2, 3], 'y': [3, 4, 5]})
+   x.iloc[1] = dict(x=9, y=99)
+   x
+
 Slicing ranges
 --------------
 

doc/source/whatsnew/v0.16.1.txt

@@ -51,6 +51,8 @@ Enhancements
 - ``get_dummies`` function now accepts ``sparse`` keyword.  If set to ``True``, the return ``DataFrame`` is sparse, e.g. ``SparseDataFrame``. (:issue:`8823`)
 - ``Period`` now accepts ``datetime64`` as value input. (:issue:`9054`)
 
+- Allow timedelta string conversion when leading zero is missing from time definition, ie `0:00:00` vs `00:00:00`. (:issue:`9570`)
+
 .. _whatsnew_0161.api:
 
 API changes
@@ -94,34 +96,59 @@ Bug Fixes
 - Fixed bug (:issue:`9542`) where labels did not appear properly in legend of ``DataFrame.plot()``. Passing ``label=`` args also now works, and series indices are no longer mutated.
 - Bug in json serialization when frame has length zero.(:issue:`9805`)
 - Bug in `read_csv` where missing trailing delimiters would cause segfault. (:issue:`5664`)
-
-
+- Bug in retaining index name on appending (:issue:`9862`)
 - Bug in ``scatter_matrix`` draws unexpected axis ticklabels (:issue:`5662`)
-
 - Fixed bug in ``StataWriter`` resulting in changes to input ``DataFrame`` upon save (:issue:`9795`).
-
-
 - Bug in ``transform`` causing length mismatch when null entries were present and a fast aggregator was being used (:issue:`9697`)
-
 - Bug in ``equals`` causing false negatives when block order differed (:issue:`9330`)
-
+- Bug in ``read_sql_table`` error when reading postgres table with timezone (:issue:`7139`)
 - Bug in ``DataFrame`` slicing may not retain metadata (:issue:`9776`)
 - Bug where ``TimdeltaIndex`` were not properly serialized in fixed ``HDFStore`` (:issue:`9635`)
-
 - Bug in plotting continuously using ``secondary_y`` may not show legend properly. (:issue:`9610`, :issue:`9779`)
-
 - Bug in ``DataFrame.plot(kind="hist")`` results in ``TypeError`` when ``DataFrame`` contains non-numeric columns  (:issue:`9853`)
 - Bug where repeated plotting of ``DataFrame`` with a ``DatetimeIndex`` may raise ``TypeError`` (:issue:`9852`)
-
 - Bug in ``Series.quantile`` on empty Series of type ``Datetime`` or ``Timedelta`` (:issue:`9675`)
 - Bug in ``where`` causing incorrect results when upcasting was required (:issue:`9731`)
 - Bug in ``FloatArrayFormatter`` where decision boundary for displaying "small" floats in decimal format is off by one order of magnitude for a given display.precision (:issue:`9764`)
-
 - Fixed bug where ``DataFrame.plot()`` raised an error when both ``color`` and ``style`` keywords were passed and there was no color symbol in the style strings (:issue:`9671`)
 - Bug in ``read_csv`` and ``read_table`` when using ``skip_rows`` parameter if blank lines are present. (:issue:`9832`)
-
 - Bug in ``read_csv()`` interprets ``index_col=True`` as ``1`` (:issue:`9798`)
-
+- Bug in index equality comparisons using ``==`` failing on Index/MultiIndex type incompatibility (:issue:`9875`)
 - Bug in which ``SparseDataFrame`` could not take `nan` as a column name (:issue:`8822`)
-
+- Bug in ``Series.quantile`` on empty Series of type ``Datetime`` or ``Timedelta`` (:issue:`9675`)
+- Bug in ``to_msgpack`` and ``read_msgpack`` zlib and blosc compression support (:issue:`9783`)
 - Bug in unequal comparisons between a ``Series`` of dtype `"category"` and a scalar (e.g. ``Series(Categorical(list("abc"), categories=list("cba"), ordered=True)) > "b"``, which wouldn't use the order of the categories but use the lexicographical order. (:issue:`9848`)
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+- Bug in unequal comparisons between categorical data and a scalar, which was not in the categories (e.g. ``Series(Categorical(list("abc"), ordered=True)) > "d"``. This returned ``False`` for all elements, but now raises a ``TypeError``. Equality comparisons also now return ``False`` for ``==`` and ``True`` for ``!=``. (:issue:`9848`)
+
+- Bug in DataFrame ``__setitem__`` when right hand side is a dictionary (:issue:`9874`)
+
+- Bug in ``MultiIndex.sortlevel()`` results in unicode level name breaks (:issue:`9875`)
+- Bug in which ``groupby.transform`` incorrectly enforced output dtypes to match input dtypes. (:issue:`9807`)
+
+
+
+
+
+
+
+
+
+
+
+- Bug where dividing a dataframe containing values of type ``Decimal`` by another ``Decimal`` would raise. (:issue:`9787`)
+- Bug where using DataFrames asfreq would remove the name of the index. (:issue:`9885`)

pandas/core/base.py

@@ -11,7 +11,7 @@
 import pandas.lib as lib
 from pandas.util.decorators import Appender, cache_readonly
 from pandas.core.strings import StringMethods
-
+from pandas.core.common import AbstractMethodError
 
 _shared_docs = dict()
 _indexops_doc_kwargs = dict(klass='IndexOpsMixin', inplace='',
@@ -32,7 +32,7 @@ class StringMixin(object):
     # Formatting
 
     def __unicode__(self):
-        raise NotImplementedError
+        raise AbstractMethodError(self)
 
     def __str__(self):
         """
@@ -566,4 +566,4 @@ def duplicated(self, take_last=False):
     # abstracts
 
     def _update_inplace(self, result, **kwargs):
-        raise NotImplementedError
+        raise AbstractMethodError(self)

pandas/core/categorical.py

@@ -61,7 +61,14 @@ def f(self, other):
                 i = self.categories.get_loc(other)
                 return getattr(self._codes, op)(i)
             else:
-                return np.repeat(False, len(self))
+                if op == '__eq__':
+                    return np.repeat(False, len(self))
+                elif op == '__ne__':
+                    return np.repeat(True, len(self))
+                else:
+                    msg  = "Cannot compare a Categorical for op {op} with a scalar, " \
+                           "which is not a category."
+                    raise TypeError(msg.format(op=op))
         else:
 
             # allow categorical vs object dtype array comparisons for equality
@@ -1159,7 +1166,8 @@ def fillna(self, fill_value=None, method=None, limit=None):
         if fill_value is None:
             fill_value = np.nan
         if limit is not None:
-            raise NotImplementedError
+            raise NotImplementedError("specifying a limit for fillna has not "
+                                      "been implemented yet")
 
         values = self._codes
 

pandas/core/common.py

@@ -39,6 +39,17 @@ class AmbiguousIndexError(PandasError, KeyError):
     pass
 
 
+class AbstractMethodError(NotImplementedError):
+    """Raise this error instead of NotImplementedError for abstract methods
+    while keeping compatibility with Python 2 and Python 3.
+    """
+    def __init__(self, class_instance):
+        self.class_instance = class_instance
+
+    def __str__(self):
+        return "This method must be defined on the concrete class of " \
+               + self.class_instance.__class__.__name__
+
 _POSSIBLY_CAST_DTYPES = set([np.dtype(t).name
                              for t in ['O', 'int8',
                                        'uint8', 'int16', 'uint16', 'int32',
@@ -1397,14 +1408,19 @@ def _fill_zeros(result, x, y, name, fill):
 
     mask the nan's from x
     """
-
     if fill is None or is_float_dtype(result):
         return result
 
     if name.startswith(('r', '__r')):
         x,y = y,x
 
-    if np.isscalar(y):
+    is_typed_variable = (hasattr(y, 'dtype') or hasattr(y,'type'))
+    is_scalar = lib.isscalar(y)
+
+    if not is_typed_variable and not is_scalar:
+        return result
+
+    if is_scalar:
         y = np.array(y)
 
     if is_integer_dtype(y):
@@ -2439,7 +2455,10 @@ def _get_dtype_type(arr_or_dtype):
         return np.dtype(arr_or_dtype).type
     elif isinstance(arr_or_dtype, CategoricalDtype):
         return CategoricalDtypeType
-    return arr_or_dtype.dtype.type
+    try:
+        return arr_or_dtype.dtype.type
+    except AttributeError:
+        raise ValueError('%r is not a dtype' % arr_or_dtype)
 
 
 def is_any_int_dtype(arr_or_dtype):
@@ -2510,7 +2529,11 @@ def is_floating_dtype(arr_or_dtype):
 
 
 def is_bool_dtype(arr_or_dtype):
-    tipo = _get_dtype_type(arr_or_dtype)
+    try:
+        tipo = _get_dtype_type(arr_or_dtype)
+    except ValueError:
+        # this isn't even a dtype
+        return False
     return issubclass(tipo, np.bool_)
 
 def is_categorical(array):

pandas/core/frame.py

@@ -4414,12 +4414,12 @@ def mode(self, axis=0, numeric_only=False):
         """
         Gets the mode(s) of each element along the axis selected. Empty if nothing
         has 2+ occurrences. Adds a row for each mode per label, fills in gaps
-        with nan. 
-        
+        with nan.
+
         Note that there could be multiple values returned for the selected
-        axis (when more than one item share the maximum frequency), which is the 
-        reason why a dataframe is returned. If you want to impute missing values 
-        with the mode in a dataframe ``df``, you can just do this: 
+        axis (when more than one item share the maximum frequency), which is the
+        reason why a dataframe is returned. If you want to impute missing values
+        with the mode in a dataframe ``df``, you can just do this:
         ``df.fillna(df.mode().iloc[0])``
 
         Parameters

pandas/core/generic.py

@@ -21,7 +21,8 @@
 from pandas.core.common import (isnull, notnull, is_list_like,
                                 _values_from_object, _maybe_promote,
                                 _maybe_box_datetimelike, ABCSeries,
-                                SettingWithCopyError, SettingWithCopyWarning)
+                                SettingWithCopyError, SettingWithCopyWarning,
+                                AbstractMethodError)
 import pandas.core.nanops as nanops
 from pandas.util.decorators import Appender, Substitution, deprecate_kwarg
 from pandas.core import config
@@ -137,7 +138,7 @@ def _init_mgr(self, mgr, axes=None, dtype=None, copy=False):
 
     @property
     def _constructor(self):
-        raise NotImplementedError
+        raise AbstractMethodError(self)
 
     def __unicode__(self):
         # unicode representation based upon iterating over self
@@ -152,7 +153,7 @@ def _local_dir(self):
 
     @property
     def _constructor_sliced(self):
-        raise NotImplementedError
+        raise AbstractMethodError(self)
 
     #----------------------------------------------------------------------
     # Axis
@@ -1100,7 +1101,7 @@ def _iget_item_cache(self, item):
         return lower
 
     def _box_item_values(self, key, values):
-        raise NotImplementedError
+        raise AbstractMethodError(self)
 
     def _maybe_cache_changed(self, item, value):
         """
@@ -3057,7 +3058,8 @@ def first(self, offset):
         """
         from pandas.tseries.frequencies import to_offset
         if not isinstance(self.index, DatetimeIndex):
-            raise NotImplementedError
+            raise NotImplementedError("'first' only supports a DatetimeIndex "
+                                      "index")
 
         if len(self.index) == 0:
             return self
@@ -3091,7 +3093,8 @@ def last(self, offset):
         """
         from pandas.tseries.frequencies import to_offset
         if not isinstance(self.index, DatetimeIndex):
-            raise NotImplementedError
+            raise NotImplementedError("'last' only supports a DatetimeIndex "
+                                      "index")
 
         if len(self.index) == 0:
             return self