CLN: Better exceptions and tests for core/frame

CLN: Big simplification of test_frame
TST: Add test cases for bad input, etc.
CLN: Exceptions that shouldn't happen should be assertions (mostly _apply*)
CLN/TST: Right Exceptions + remove duplicate tests

A few tests were duplicating the same thing or actually just the same
test, those have been refactored or removed. Many tests were changed to
specify the Exception they were looking for (and also to use the new
`with_statement` format for assertRaises, etc.)

CLN: Harmonize SparseDF and DF Exceptions
ENH: SparseDF apply now accepts axis='columns'

Author: jtratner

doc/source/release.rst

@@ -71,8 +71,9 @@ pandas 0.13
     when the key is a column
   - Support for using a ``DatetimeIndex/PeriodsIndex`` directly in a datelike calculation
     e.g. s-s.index (:issue:`4629`)
-  - Better/cleaned up exceptions in core/common, io/excel and core/format.
-    (:issue:`4721`, :issue:`3954`)
+  - Better/cleaned up exceptions in core/common, io/excel and core/format
+    (:issue:`4721`, :issue:`3954`), as well as cleaned up test cases in
+    tests/test_frame, tests/test_multilevel (:issue:`4732`).
 
 **API Changes**
 
@@ -143,9 +144,10 @@ pandas 0.13
       now returns a ``MultiIndex`` rather than an ``Index``. (:issue:`4039`)
 
   - Infer and downcast dtype if ``downcast='infer'`` is passed to ``fillna/ffill/bfill`` (:issue:`4604`)
-  - Factored out excel_value_to_python_value from ExcelFile::_parse_excel (:issue:`4589`)
   - ``__nonzero__`` for all NDFrame objects, will now raise a ``ValueError``, this reverts back to (:issue:`1073`, :issue:`4633`)
     behavior.
+  - ``DataFrame.update()`` no longer raises a ``DataConflictError``, it now
+    will raise a ``ValueError`` instead (if necessary) (:issue:`4732`)
 
 **Internal Refactoring**
 

pandas/core/format.py

@@ -825,9 +825,8 @@ def __init__(self, obj, path_or_buf, sep=",", na_rep='', float_format=None,
 
         # validate mi options
         if self.has_mi_columns:
-            # guarded against in to_csv itself
-            if cols is not None: # pragma: no cover
-                raise AssertionError("cannot specify cols with a multi_index on the columns")
+            if cols is not None:
+                raise TypeError("cannot specify cols with a MultiIndex on the columns")
 
         if cols is not None:
             if isinstance(cols,Index):

pandas/core/frame.py

@@ -417,7 +417,7 @@ def rename(self, *args, **kwargs):
         inplace = kwargs.get('inplace', False)
 
         if (com._count_not_none(*axes.values()) == 0):
-            raise Exception('must pass an index to rename')
+            raise TypeError('must pass an index to rename')
 
         # renamer function if passed a dict
         def _get_rename_function(mapper):
@@ -1242,7 +1242,7 @@ def filter(self, items=None, like=None, regex=None, axis=None):
             matcher = re.compile(regex)
             return self.select(lambda x: matcher.search(x) is not None, axis=axis_name)
         else:
-            raise ValueError('items was None!')
+            raise TypeError('Must pass either `items`, `like`, or `regex`')
 
     #----------------------------------------------------------------------
     # Attribute access

pandas/core/generic.py

@@ -84,7 +84,7 @@ def __setitem__(self, key, value):
 
         if isinstance(key, tuple):
             if len(key) > self.ndim:
-                raise IndexingError('only tuples of length <= %d supported',
+                raise IndexingError('only tuples of length <= %d supported' %
                                     self.ndim)
             indexer = self._convert_tuple(key, is_setter=True)
         else:

pandas/core/indexing.py

@@ -1625,7 +1625,7 @@ def set_axis(self, axis, value, maybe_rename=True, check_axis=True):
         value = _ensure_index(value)
 
         if check_axis and len(value) != len(cur_axis):
-            raise Exception('Length mismatch (%d vs %d)'
+            raise ValueError('Length mismatch (%d vs %d)'
                             % (len(value), len(cur_axis)))
 
         self.axes[axis] = value
@@ -2297,7 +2297,7 @@ def fast_2d_xs(self, loc, copy=False):
             return result
 
         if not copy:
-            raise Exception('cannot get view of mixed-type or '
+            raise TypeError('cannot get view of mixed-type or '
                             'non-consolidated DataFrame')
 
         items = self.items
@@ -2483,7 +2483,8 @@ def _set_item(item, arr):
     def insert(self, loc, item, value, allow_duplicates=False):
 
         if not allow_duplicates and item in self.items:
-            raise Exception('cannot insert %s, already exists' % item)
+            # Should this be a different kind of error??
+            raise ValueError('cannot insert %s, already exists' % item)
 
         try:
             new_items = self.items.insert(loc, item)
@@ -2826,7 +2827,8 @@ def _maybe_rename_join(self, other, lsuffix, rsuffix, copydata=True):
         to_rename = self.items.intersection(other.items)
         if len(to_rename) > 0:
             if not lsuffix and not rsuffix:
-                raise Exception('columns overlap: %s' % to_rename)
+                raise ValueError('columns overlap but no suffix specified: %s'
+                                 % to_rename)
 
             def lrenamer(x):
                 if x in to_rename:

pandas/core/internals.py

@@ -21,10 +21,6 @@
 from pandas.core.index import Index, MultiIndex
 
 
-class ReshapeError(Exception):
-    pass
-
-
 class _Unstacker(object):
     """
     Helper class to unstack data / pivot with multi-level index
@@ -129,8 +125,8 @@ def _make_selectors(self):
         mask.put(selector, True)
 
         if mask.sum() < len(self.index):
-            raise ReshapeError('Index contains duplicate entries, '
-                               'cannot reshape')
+            raise ValueError('Index contains duplicate entries, '
+                             'cannot reshape')
 
         self.group_index = comp_index
         self.mask = mask

pandas/core/reshape.py

@@ -2377,7 +2377,7 @@ def sort(self, axis=0, kind='quicksort', order=None, ascending=True):
 
         if (true_base is not None and
                 (true_base.ndim != 1 or true_base.shape != self.shape)):
-            raise Exception('This Series is a view of some other array, to '
+            raise TypeError('This Series is a view of some other array, to '
                             'sort in-place you must create a copy')
 
         self[:] = sortedSeries
@@ -2542,7 +2542,7 @@ def sortlevel(self, level=0, ascending=True):
         sorted : Series
         """
         if not isinstance(self.index, MultiIndex):
-            raise Exception('can only sort by level with a hierarchical index')
+            raise TypeError('can only sort by level with a hierarchical index')
 
         new_index, indexer = self.index.sortlevel(level, ascending=ascending)
         new_values = self.values.take(indexer)

pandas/core/series.py

@@ -600,7 +600,8 @@ def _reindex_with_indexers(self, reindexers, method=None, fill_value=np.nan, lim
     def _join_compat(self, other, on=None, how='left', lsuffix='', rsuffix='',
                      sort=False):
         if isinstance(other, Series):
-            assert(other.name is not None)
+            if other.name is None:
+                raise ValueError('Other Series must have a name')
             other = SparseDataFrame({other.name: other},
                                     default_fill_value=self._default_fill_value)
         if on is not None:
@@ -627,19 +628,19 @@ def _join_index(self, other, how, lsuffix, rsuffix):
         return concat([this, other], axis=1, verify_integrity=True)
 
     def _maybe_rename_join(self, other, lsuffix, rsuffix):
-        intersection = self.columns.intersection(other.columns)
-
-        if len(intersection) > 0:
+        to_rename = self.columns.intersection(other.columns)
+        if len(to_rename) > 0:
             if not lsuffix and not rsuffix:
-                raise Exception('columns overlap: %s' % intersection)
+                raise ValueError('columns overlap but no suffix specified: %s'
+                                 % to_rename)
 
             def lrenamer(x):
-                if x in intersection:
+                if x in to_rename:
                     return '%s%s' % (x, lsuffix)
                 return x
 
             def rrenamer(x):
-                if x in intersection:
+                if x in to_rename:
                     return '%s%s' % (x, rsuffix)
                 return x
 
@@ -687,7 +688,7 @@ def apply(self, func, axis=0, broadcast=False, reduce=False):
         ----------
         func : function
             Function to apply to each column
-        axis : {0, 1}
+        axis : {0, 1, 'index', 'columns'}
         broadcast : bool, default False
             For aggregation functions, return object of same size with values
             propagated
@@ -698,6 +699,7 @@ def apply(self, func, axis=0, broadcast=False, reduce=False):
         """
         if not len(self.columns):
             return self
+        axis = self._get_axis_number(axis)
 
         if isinstance(func, np.ufunc):
             new_series = {}