API: expose pandas.errors

closes #14800

Author: Jeff Reback <[email protected]>

Closes #15541 from jreback/exceptions and squashes the following commits:

e5fbdc8 [Jeff Reback] give nicer deprecation / message on infer_dtype moving
ab4525b [Jeff Reback] typo on pandas.errors in whatsnew
d636ef7 [Jeff Reback] document removed exceptions
3dc4b9a [Jeff Reback] more docs for exceptions
2bb1fbd [Jeff Reback] remove AmbiguousIndexError, completely unused
5754630 [Jeff Reback] fix doc-string
35d225f [Jeff Reback] more examples
e91901d [Jeff Reback] DOC: better docs on infer_type
7e8432d [Jeff Reback] remove need for PandasError sub-class
92b2fdc [Jeff Reback] corrections
991fbb4 [Jeff Reback] API: expose pandas.errors
eec40cd [Jeff Reback] add pandas.api.lib add infer_dtype to pandas.api.lib

Author: jreback

doc/source/whatsnew/v0.20.0.txt

@@ -76,6 +76,28 @@ Commonly called 'unix epoch' or POSIX time.
 
    pd.to_datetime([1, 2, 3], unit='D')
 
+.. _whatsnew_0200.enhancements.errors:
+
+pandas errors
+^^^^^^^^^^^^^
+
+We are adding a standard public location for all pandas exceptions & warnings ``pandas.errors``. (:issue:`14800`). Previously
+these exceptions & warnings could be imported from ``pandas.core.common`` or ``pandas.io.common``. These exceptions and warnings
+will be removed from the ``*.common`` locations in a future release. (:issue:`15541`)
+
+The following are now part of this API:
+
+.. code-block:: python
+
+   ['DtypeWarning',
+    'EmptyDataError',
+    'OutOfBoundsDatetime',
+    'ParserError',
+    'ParserWarning',
+    'PerformanceWarning',
+    'UnsortedIndexError',
+    'UnsupportedFunctionCall']
+
 .. _whatsnew_0200.enhancements.groupby_access:
 
 Groupby Enhancements
@@ -858,6 +880,10 @@ Other API Changes
 - ``NaT`` will now correctly return ``np.nan`` for ``Timedelta`` and ``Period`` accessors such as ``days`` and ``quarter`` (:issue:`15782`)
 - ``NaT`` will now returns ``NaT`` for ``tz_localize`` and ``tz_convert``
   methods (:issue:`15830`)
+- ``DataFrame`` and ``Panel`` constructors with invalid input will now raise ``ValueError`` rather than ``PandasError``, if called with scalar inputs and not axes (:issue:`15541`)
+
+- ``DataFrame`` and ``Panel`` constructors with invalid input will now raise ``ValueError`` rather than ``pandas.core.common.PandasError``, if called with scalar inputs and not axes; The exception ``PandasError`` is removed as well. (:issue:`15541`)
+- The exception ``pandas.core.common.AmbiguousIndexError`` is removed as it is not referenced (:issue:`15541`)
 
 .. _whatsnew_0200.develop:
 

pandas/__init__.py

@@ -62,7 +62,8 @@
 
 json = _DeprecatedModule(deprmod='pandas.json', deprmodto='pandas.io.json.libjson')
 parser = _DeprecatedModule(deprmod='pandas.parser', deprmodto='pandas.io.libparsers')
-lib = _DeprecatedModule(deprmod='pandas.lib', deprmodto='pandas._libs.lib')
+lib = _DeprecatedModule(deprmod='pandas.lib', deprmodto='pandas._libs.lib',
+                        moved={'infer_dtype': 'pandas.api.lib.infer_dtype'})
 tslib = _DeprecatedModule(deprmod='pandas.tslib', deprmodto='pandas._libs.tslib')
 
 # use the closest tagged version if possible

pandas/_libs/src/inference.pyx

@@ -218,9 +218,91 @@ cdef _try_infer_map(v):
     return None
 
 
-def infer_dtype(object _values):
+def infer_dtype(object value):
     """
-    we are coercing to an ndarray here
+    Effeciently infer the type of a passed val, or list-like
+    array of values. Return a string describing the type.
+
+    Parameters
+    ----------
+    value : scalar, list, ndarray, or pandas type
+
+    Returns
+    -------
+    string describing the common type of the input data.
+    Results can include:
+
+    - string
+    - unicode
+    - bytes
+    - floating
+    - integer
+    - mixed-integer
+    - mixed-integer-float
+    - complex
+    - categorical
+    - boolean
+    - datetime64
+    - datetime
+    - date
+    - timedelta64
+    - timedelta
+    - time
+    - period
+    - mixed
+
+    Raises
+    ------
+    TypeError if ndarray-like but cannot infer the dtype
+
+    Notes
+    -----
+    - 'mixed' is the catchall for anything that is not otherwise
+      specialized
+    - 'mixed-integer-float' are floats and integers
+    - 'mixed-integer' are integers mixed with non-integers
+
+    Examples
+    --------
+    >>> infer_dtype(['foo', 'bar'])
+    'string'
+
+    >>> infer_dtype([b'foo', b'bar'])
+    'bytes'
+
+    >>> infer_dtype([1, 2, 3])
+    'integer'
+
+    >>> infer_dtype([1, 2, 3.5])
+    'mixed-integer-float'
+
+    >>> infer_dtype([1.0, 2.0, 3.5])
+    'floating'
+
+    >>> infer_dtype(['a', 1])
+    'mixed-integer'
+
+    >>> infer_dtype([True, False])
+    'boolean'
+
+    >>> infer_dtype([True, False, np.nan])
+    'mixed'
+
+    >>> infer_dtype([pd.Timestamp('20130101')])
+    'datetime'
+
+    >>> infer_dtype([datetime.date(2013, 1, 1)])
+    'date'
+
+    >>> infer_dtype([np.datetime64('2013-01-01')])
+    'datetime64'
+
+    >>> infer_dtype([datetime.timedelta(0, 1, 1)])
+    'timedelta'
+
+    >>> infer_dtype(pd.Series(list('aabc')).astype('category'))
+    'categorical'
+
     """
 
     cdef:
@@ -229,27 +311,27 @@ def infer_dtype(object _values):
         ndarray values
         bint seen_pdnat = False, seen_val = False
 
-    if isinstance(_values, np.ndarray):
-        values = _values
-    elif hasattr(_values, 'dtype'):
+    if isinstance(value, np.ndarray):
+        values = value
+    elif hasattr(value, 'dtype'):
 
         # this will handle ndarray-like
         # e.g. categoricals
         try:
-            values = getattr(_values, '_values', getattr(
-                _values, 'values', _values))
+            values = getattr(value, '_values', getattr(
+                value, 'values', value))
         except:
-            val = _try_infer_map(_values)
-            if val is not None:
-                return val
+            value = _try_infer_map(value)
+            if value is not None:
+                return value
 
             # its ndarray like but we can't handle
-            raise ValueError("cannot infer type for {0}".format(type(_values)))
+            raise ValueError("cannot infer type for {0}".format(type(value)))
 
     else:
-        if not isinstance(_values, list):
-            _values = list(_values)
-        values = list_to_object_array(_values)
+        if not isinstance(value, list):
+            value = list(value)
+        values = list_to_object_array(value)
 
     values = getattr(values, 'values', values)
     val = _try_infer_map(values)

pandas/api/lib/__init__.py

@@ -0,0 +1,5 @@
+# flake8: noqa
+
+""" public toolkit API """
+
+from pandas._libs.lib import infer_dtype

pandas/compat/numpy/function.py

@@ -21,7 +21,7 @@
 from numpy import ndarray
 from pandas.util.validators import (validate_args, validate_kwargs,
                                     validate_args_and_kwargs)
-from pandas.core.common import UnsupportedFunctionCall
+from pandas.errors import UnsupportedFunctionCall
 from pandas.types.common import is_integer, is_bool
 from pandas.compat import OrderedDict
 

pandas/computation/align.py

@@ -9,7 +9,8 @@
 
 import pandas as pd
 from pandas import compat
-import pandas.core.common as com
+from pandas.errors import PerformanceWarning
+from pandas.core.common import flatten
 from pandas.computation.common import _result_type_many
 
 
@@ -101,7 +102,7 @@ def _align_core(terms):
                                   'than an order of magnitude on term {1!r}, '
                                   'by more than {2:.4g}; performance may '
                                   'suffer'.format(axis, terms[i].name, ordm),
-                                  category=pd.core.common.PerformanceWarning,
+                                  category=PerformanceWarning,
                                   stacklevel=6)
 
                 if transpose:
@@ -121,7 +122,7 @@ def _align(terms):
     """Align a set of terms"""
     try:
         # flatten the parse tree (a nested list, really)
-        terms = list(com.flatten(terms))
+        terms = list(flatten(terms))
     except TypeError:
         # can't iterate so it must just be a constant or single variable
         if isinstance(terms.value, pd.core.generic.NDFrame):

pandas/core/common.py

@@ -20,6 +20,10 @@
 from pandas.api import types
 from pandas.types import common
 
+# compat
+from pandas.errors import (  # noqa
+    PerformanceWarning, UnsupportedFunctionCall, UnsortedIndexError)
+
 # back-compat of public API
 # deprecate these functions
 m = sys.modules['pandas.core.common']
@@ -73,14 +77,6 @@ def array_equivalent(*args, **kwargs):
     return missing.array_equivalent(*args, **kwargs)
 
 
-class PandasError(Exception):
-    pass
-
-
-class PerformanceWarning(Warning):
-    pass
-
-
 class SettingWithCopyError(ValueError):
     pass
 
@@ -89,24 +85,6 @@ class SettingWithCopyWarning(Warning):
     pass
 
 
-class AmbiguousIndexError(PandasError, KeyError):
-    pass
-
-
-class UnsupportedFunctionCall(ValueError):
-    pass
-
-
-class UnsortedIndexError(KeyError):
-    """ Error raised when attempting to get a slice of a MultiIndex
-    and the index has not been lexsorted. Subclass of `KeyError`.
-
-    .. versionadded:: 0.20.0
-
-    """
-    pass
-
-
 class AbstractMethodError(NotImplementedError):
     """Raise this error instead of NotImplementedError for abstract methods
     while keeping compatibility with Python 2 and Python 3.

pandas/core/frame.py

@@ -56,7 +56,7 @@
                                  is_named_tuple)
 from pandas.types.missing import isnull, notnull
 
-from pandas.core.common import (PandasError, _try_sort,
+from pandas.core.common import (_try_sort,
                                 _default_index,
                                 _values_from_object,
                                 _maybe_box_datetimelike,
@@ -347,7 +347,7 @@ def __init__(self, data=None, index=None, columns=None, dtype=None,
                 mgr = self._init_ndarray(values, index, columns, dtype=dtype,
                                          copy=False)
             else:
-                raise PandasError('DataFrame constructor not properly called!')
+                raise ValueError('DataFrame constructor not properly called!')
 
         NDFrame.__init__(self, mgr, fastpath=True)
 

pandas/core/indexing.py

@@ -1140,7 +1140,7 @@ def _convert_to_indexer(self, obj, axis=0, is_setter=False):
         ix[['foo', 'bar', 'baz']] -> [i, j, k] (indices of foo, bar, baz)
 
         Going by Zen of Python?
-        "In the face of ambiguity, refuse the temptation to guess."
+        'In the face of ambiguity, refuse the temptation to guess.'
         raise AmbiguousIndexError with integer labels?
         - No, prefer label-based indexing
         """

pandas/core/ops.py

@@ -21,8 +21,8 @@
 from pandas.compat import bind_method
 import pandas.core.missing as missing
 
-from pandas.core.common import (_values_from_object, _maybe_match_name,
-                                PerformanceWarning)
+from pandas.errors import PerformanceWarning
+from pandas.core.common import _values_from_object, _maybe_match_name
 from pandas.types.missing import notnull, isnull
 from pandas.types.common import (needs_i8_conversion,
                                  is_datetimelike_v_numeric,

pandas/core/panel.py

@@ -21,7 +21,7 @@
 from pandas import compat
 from pandas.compat import (map, zip, range, u, OrderedDict, OrderedDefaultdict)
 from pandas.compat.numpy import function as nv
-from pandas.core.common import PandasError, _try_sort, _default_index
+from pandas.core.common import _try_sort, _default_index
 from pandas.core.frame import DataFrame
 from pandas.core.generic import NDFrame, _shared_docs
 from pandas.core.index import (Index, MultiIndex, _ensure_index,
@@ -174,7 +174,7 @@ def _init_data(self, data, copy, dtype, **kwargs):
                                     copy=False)
             copy = False
         else:  # pragma: no cover
-            raise PandasError('Panel constructor not properly called!')
+            raise ValueError('Panel constructor not properly called!')
 
         NDFrame.__init__(self, mgr, axes=axes, copy=copy, dtype=dtype)
 
@@ -1150,8 +1150,8 @@ def _construct_return_type(self, result, axes=None):
             return self._constructor_sliced(
                 result, **self._extract_axes_for_slice(self, axes))
 
-        raise PandasError('invalid _construct_return_type [self->%s] '
-                          '[result->%s]' % (self, result))
+        raise ValueError('invalid _construct_return_type [self->%s] '
+                         '[result->%s]' % (self, result))
 
     def _wrap_result(self, result, axis):
         axis = self._get_axis_name(axis)