Merge remote-tracking branch 'upstream/master'

Author: erikoveson

Makefile

@@ -13,7 +13,7 @@ build: clean_pyc
 	python setup.py build_ext --inplace
 
 lint-diff:
-	git diff master --name-only -- "*.py" | grep -E "pandas|scripts" | xargs flake8
+	git diff upstream/master --name-only -- "*.py" | xargs flake8
 
 develop: build
 	-python setup.py develop

ci/deps/travis-36-doc.yaml

@@ -2,7 +2,6 @@ name: pandas
 channels:
   - defaults
   - conda-forge
-  - r
 dependencies:
   - beautifulsoup4
   - bottleneck
@@ -31,14 +30,11 @@ dependencies:
   - python-snappy
   - python=3.6*
   - pytz
-  - r
-  - rpy2
   - scipy
   - seaborn
   - sphinx
   - sqlalchemy
   - statsmodels
-  - tzlocal
   - xarray
   - xlrd
   - xlsxwriter

doc/source/contributing.rst

@@ -575,7 +575,7 @@ the `flake8 <https://pypi.org/project/flake8>`_ tool
 and report any stylistic errors in your code. Therefore, it is helpful before
 submitting code to run the check yourself on the diff::
 
-   git diff master -u -- "*.py" | flake8 --diff
+   git diff upstream/master -u -- "*.py" | flake8 --diff
 
 This command will catch any stylistic errors in your changes specifically, but
 be beware it may not catch all of them. For example, if you delete the only
@@ -584,21 +584,21 @@ unused function. However, style-checking the diff will not catch this because
 the actual import is not part of the diff. Thus, for completeness, you should
 run this command, though it will take longer::
 
-   git diff master --name-only -- "*.py" | grep "pandas/" | xargs -r flake8
+   git diff upstream/master --name-only -- "*.py" | xargs -r flake8
 
 Note that on OSX, the ``-r`` flag is not available, so you have to omit it and
 run this slightly modified command::
 
-   git diff master --name-only -- "*.py" | grep "pandas/" | xargs flake8
+   git diff upstream/master --name-only -- "*.py" | xargs flake8
 
-Windows does not support the ``grep`` and ``xargs`` commands (unless installed
-for example via the `MinGW <http://www.mingw.org/>`__ toolchain), but one can
-imitate the behaviour as follows::
+Windows does not support the ``xargs`` command (unless installed for example
+via the `MinGW <http://www.mingw.org/>`__ toolchain), but one can imitate the
+behaviour as follows::
 
-    for /f %i in ('git diff upstream/master --name-only ^| findstr pandas/') do flake8 %i
+    for /f %i in ('git diff upstream/master --name-only -- "*.py"') do flake8 %i
 
-This will also get all the files being changed by the PR (and within the
-``pandas/`` folder), and run ``flake8`` on them one after the other.
+This will get all the files being changed by the PR (and ending with ``.py``),
+and run ``flake8`` on them, one after the other.
 
 .. _contributing.import-formatting:
 

doc/source/r_interface.rst

@@ -33,21 +33,28 @@ See also the documentation of the `rpy2 <http://rpy2.bitbucket.org/>`__ project:
 
 In the remainder of this page, a few examples of explicit conversion is given. The pandas conversion of rpy2 needs first to be activated:
 
-.. ipython:: python
+.. code-block:: python
 
-   from rpy2.robjects import r, pandas2ri
-   pandas2ri.activate()
+    >>> from rpy2.robjects import pandas2ri  # doctest: +SKIP
+    >>> pandas2ri.activate()  # doctest: +SKIP
 
 Transferring R data sets into Python
 ------------------------------------
 
 Once the pandas conversion is activated (``pandas2ri.activate()``), many conversions
 of R to pandas objects will be done automatically. For example, to obtain the 'iris' dataset as a pandas DataFrame:
 
-.. ipython:: python
+.. code-block:: python
 
-    r.data('iris')
-    r['iris'].head()
+    >>> from rpy2.robjects import r  # doctest: +SKIP
+    >>> r.data('iris')  # doctest: +SKIP
+    >>> r['iris'].head()  # doctest: +SKIP
+        Sepal.Length  Sepal.Width  Petal.Length  Petal.Width Species
+    0           5.1          3.5           1.4          0.2  setosa
+    1           4.9          3.0           1.4          0.2  setosa
+    2           4.7          3.2           1.3          0.2  setosa
+    3           4.6          3.1           1.5          0.2  setosa
+    4           5.0          3.6           1.4          0.2  setosa
 
 If the pandas conversion was not activated, the above could also be accomplished
 by explicitly converting it with the ``pandas2ri.ri2py`` function
@@ -59,13 +66,19 @@ Converting DataFrames into R objects
 The ``pandas2ri.py2ri`` function support the reverse operation to convert
 DataFrames into the equivalent R object (that is, **data.frame**):
 
-.. ipython:: python
+.. code-block:: python
+
+   >>> df = pd.DataFrame({'A': [1, 2, 3], 'B': [4, 5, 6], 'C': [7, 8, 9]},
+   ...                   index=["one", "two", "three"])  # doctest: +SKIP
+   >>> r_dataframe = pandas2ri.py2ri(df)  # doctest: +SKIP
+   >>> print(type(r_dataframe))  # doctest: +SKIP
+   <class 'rpy2.robjects.vectors.DataFrame'>
+   >>> print(r_dataframe)  # doctest: +SKIP
+         A B C
+   one   1 4 7
+   two   2 5 8
+   three 3 6 9
 
-   df = pd.DataFrame({'A': [1, 2, 3], 'B': [4, 5, 6], 'C':[7,8,9]},
-                     index=["one", "two", "three"])
-   r_dataframe = pandas2ri.py2ri(df)
-   print(type(r_dataframe))
-   print(r_dataframe)
 
 The DataFrame's index is stored as the ``rownames`` attribute of the
 data.frame instance.

doc/source/whatsnew/v0.24.0.rst

@@ -466,6 +466,7 @@ For situations where you need an ``ndarray`` of ``Interval`` objects, use
    np.asarray(idx)
    idx.values.astype(object)
 
+
 .. _whatsnew_0240.api.timezone_offset_parsing:
 
 Parsing Datetime Strings with Timezone Offsets
@@ -1442,6 +1443,7 @@ Reshaping
 - Bug in :func:`merge_asof` where confusing error message raised when attempting to merge with missing values (:issue:`23189`)
 - Bug in :meth:`DataFrame.nsmallest` and :meth:`DataFrame.nlargest` for dataframes that have a :class:`MultiIndex` for columns (:issue:`23033`).
 - Bug in :meth:`DataFrame.append` with a :class:`Series` with a dateutil timezone would raise a ``TypeError`` (:issue:`23682`)
+- Bug in ``Series`` construction when passing no data and ``dtype=str`` (:issue:`22477`)
 
 .. _whatsnew_0240.bug_fixes.sparse:
 
@@ -1474,6 +1476,7 @@ Other
 - :meth:`DataFrame.nlargest` and :meth:`DataFrame.nsmallest` now returns the correct n values when keep != 'all' also when tied on the first columns (:issue:`22752`)
 - :meth:`~pandas.io.formats.style.Styler.bar` now also supports tablewise application (in addition to rowwise and columnwise) with ``axis=None`` and setting clipping range with ``vmin`` and ``vmax`` (:issue:`21548` and :issue:`21526`). ``NaN`` values are also handled properly.
 - Logical operations ``&, |, ^`` between :class:`Series` and :class:`Index` will no longer raise ``ValueError`` (:issue:`22092`)
+- Checking PEP 3141 numbers in :func:`~pandas.api.types.is_scalar` function returns ``True`` (:issue:`22903`)
 - Bug in :meth:`DataFrame.combine_first` in which column types were unexpectedly converted to float (:issue:`20699`)
 
 .. _whatsnew_0.24.0.contributors:
@@ -1482,3 +1485,4 @@ Contributors
 ~~~~~~~~~~~~
 
 .. contributors:: v0.23.4..HEAD
+

pandas/_libs/lib.pyx

@@ -1,5 +1,8 @@
 # -*- coding: utf-8 -*-
 from decimal import Decimal
+from fractions import Fraction
+from numbers import Number
+
 import sys
 
 import cython
@@ -15,7 +18,6 @@ from cpython.datetime cimport (PyDateTime_Check, PyDate_Check,
                                PyDateTime_IMPORT)
 PyDateTime_IMPORT
 
-
 import numpy as np
 cimport numpy as cnp
 from numpy cimport (ndarray, PyArray_GETITEM,
@@ -105,23 +107,54 @@ def is_scalar(val: object) -> bool:
     """
     Return True if given value is scalar.
 
-    This includes:
-    - numpy array scalar (e.g. np.int64)
-    - Python builtin numerics
-    - Python builtin byte arrays and strings
-    - None
-    - instances of datetime.datetime
-    - instances of datetime.timedelta
-    - Period
-    - instances of decimal.Decimal
-    - Interval
-    - DateOffset
+    Parameters
+    ----------
+    val : object
+        This includes:
+
+        - numpy array scalar (e.g. np.int64)
+        - Python builtin numerics
+        - Python builtin byte arrays and strings
+        - None
+        - datetime.datetime
+        - datetime.timedelta
+        - Period
+        - decimal.Decimal
+        - Interval
+        - DateOffset
+        - Fraction
+        - Number
+
+    Returns
+    -------
+    bool
+        Return True if given object is scalar, False otherwise
+
+    Examples
+    --------
+    >>> dt = pd.datetime.datetime(2018, 10, 3)
+    >>> pd.is_scalar(dt)
+    True
+
+    >>> pd.api.types.is_scalar([2, 3])
+    False
+
+    >>> pd.api.types.is_scalar({0: 1, 2: 3})
+    False
+
+    >>> pd.api.types.is_scalar((0, 2))
+    False
+
+    pandas supports PEP 3141 numbers:
 
+    >>> from fractions import Fraction
+    >>> pd.api.types.is_scalar(Fraction(3, 5))
+    True
     """
 
     return (cnp.PyArray_IsAnyScalar(val)
             # As of numpy-1.9, PyArray_IsAnyScalar misses bytearrays on Py3.
-            or isinstance(val, bytes)
+            or isinstance(val, (bytes, Fraction, Number))
             # We differ from numpy (as of 1.10), which claims that None is
             # not scalar in np.isscalar().
             or val is None

pandas/core/arrays/timedeltas.py

@@ -1,5 +1,8 @@
 # -*- coding: utf-8 -*-
+from __future__ import division
+
 from datetime import timedelta
+import operator
 import warnings
 
 import numpy as np
@@ -16,9 +19,11 @@
     _TD_DTYPE, ensure_int64, is_datetime64_dtype, is_float_dtype,
     is_integer_dtype, is_list_like, is_object_dtype, is_string_dtype,
     is_timedelta64_dtype)
-from pandas.core.dtypes.generic import ABCSeries, ABCTimedeltaIndex
+from pandas.core.dtypes.generic import (
+    ABCDataFrame, ABCIndexClass, ABCSeries, ABCTimedeltaIndex)
 from pandas.core.dtypes.missing import isna
 
+from pandas.core import ops
 from pandas.core.algorithms import checked_add_with_arr
 import pandas.core.common as com
 
@@ -101,8 +106,32 @@ def wrapper(self, other):
     return compat.set_function_name(wrapper, opname, cls)
 
 
+def _wrap_tdi_op(op):
+    """
+    Instead of re-implementing multiplication/division etc operations
+    in the Array class, for now we dispatch to the TimedeltaIndex
+    implementations.
+    """
+    # TODO: implement directly here and wrap in TimedeltaIndex, instead of
+    #  the other way around
+    def method(self, other):
+        if isinstance(other, (ABCSeries, ABCDataFrame, ABCIndexClass)):
+            return NotImplemented
+
+        from pandas import TimedeltaIndex
+        obj = TimedeltaIndex(self)
+        result = op(obj, other)
+        if is_timedelta64_dtype(result):
+            return type(self)(result)
+        return np.array(result)
+
+    method.__name__ = '__{name}__'.format(name=op.__name__)
+    return method
+
+
 class TimedeltaArrayMixin(dtl.DatetimeLikeArrayMixin):
     _typ = "timedeltaarray"
+    __array_priority__ = 1000
 
     @property
     def _box_func(self):
@@ -293,11 +322,25 @@ def _evaluate_with_timedelta_like(self, other, op):
 
         return NotImplemented
 
+    __mul__ = _wrap_tdi_op(operator.mul)
+    __rmul__ = __mul__
+    __truediv__ = _wrap_tdi_op(operator.truediv)
+    __floordiv__ = _wrap_tdi_op(operator.floordiv)
+    __rfloordiv__ = _wrap_tdi_op(ops.rfloordiv)
+
+    if compat.PY2:
+        __div__ = __truediv__
+
+    # Note: TimedeltaIndex overrides this in call to cls._add_numeric_methods
     def __neg__(self):
         if self.freq is not None:
             return type(self)(-self._data, freq=-self.freq)
         return type(self)(-self._data)
 
+    def __abs__(self):
+        # Note: freq is not preserved
+        return type(self)(np.abs(self._data))
+
     # ----------------------------------------------------------------
     # Conversion Methods - Vectorized analogues of Timedelta methods
 

pandas/core/dtypes/cast.py

@@ -6,7 +6,7 @@
 
 from pandas._libs import lib, tslib, tslibs
 from pandas._libs.tslibs import OutOfBoundsDatetime, Period, iNaT
-from pandas.compat import PY3, string_types, text_type
+from pandas.compat import PY3, string_types, text_type, to_str
 
 from .common import (
     _INT64_DTYPE, _NS_DTYPE, _POSSIBLY_CAST_DTYPES, _TD_DTYPE, _string_dtypes,
@@ -1216,11 +1216,16 @@ def construct_1d_arraylike_from_scalar(value, length, dtype):
         if not isinstance(dtype, (np.dtype, type(np.dtype))):
             dtype = dtype.dtype
 
-        # coerce if we have nan for an integer dtype
-        # GH 22858: only cast to float if an index
-        # (passed here as length) is specified
         if length and is_integer_dtype(dtype) and isna(value):
-            dtype = np.float64
+            # coerce if we have nan for an integer dtype
+            dtype = np.dtype('float64')
+        elif isinstance(dtype, np.dtype) and dtype.kind in ("U", "S"):
+            # we need to coerce to object dtype to avoid
+            # to allow numpy to take our string as a scalar value
+            dtype = object
+            if not isna(value):
+                value = to_str(value)
+
         subarr = np.empty(length, dtype=dtype)
         subarr.fill(value)
 

pandas/core/dtypes/common.py

@@ -419,7 +419,7 @@ def is_datetime64_dtype(arr_or_dtype):
         return False
     try:
         tipo = _get_dtype_type(arr_or_dtype)
-    except TypeError:
+    except (TypeError, UnicodeEncodeError):
         return False
     return issubclass(tipo, np.datetime64)