Merge pull request #5584 from jreback/copy_warn

BUG/TST: reset setitem_copy on object enlargement

Author: jreback

pandas/computation/tests/test_eval.py

@@ -1,6 +1,5 @@
 #!/usr/bin/env python
 
-import unittest
 import functools
 from itertools import product
 
@@ -104,10 +103,11 @@ def _is_py3_complex_incompat(result, expected):
 _good_arith_ops = com.difference(_arith_ops_syms, _special_case_arith_ops_syms)
 
 
-class TestEvalNumexprPandas(unittest.TestCase):
+class TestEvalNumexprPandas(tm.TestCase):
 
     @classmethod
     def setUpClass(cls):
+        super(TestEvalNumexprPandas, cls).setUpClass()
         skip_if_no_ne()
         import numexpr as ne
         cls.ne = ne
@@ -116,6 +116,7 @@ def setUpClass(cls):
 
     @classmethod
     def tearDownClass(cls):
+        super(TestEvalNumexprPandas, cls).tearDownClass()
         del cls.engine, cls.parser
         if hasattr(cls, 'ne'):
             del cls.ne
@@ -707,6 +708,7 @@ class TestEvalNumexprPython(TestEvalNumexprPandas):
 
     @classmethod
     def setUpClass(cls):
+        super(TestEvalNumexprPython, cls).setUpClass()
         skip_if_no_ne()
         import numexpr as ne
         cls.ne = ne
@@ -733,6 +735,7 @@ class TestEvalPythonPython(TestEvalNumexprPython):
 
     @classmethod
     def setUpClass(cls):
+        super(TestEvalPythonPython, cls).setUpClass()
         cls.engine = 'python'
         cls.parser = 'python'
 
@@ -761,6 +764,7 @@ class TestEvalPythonPandas(TestEvalPythonPython):
 
     @classmethod
     def setUpClass(cls):
+        super(TestEvalPythonPandas, cls).setUpClass()
         cls.engine = 'python'
         cls.parser = 'pandas'
 
@@ -1024,17 +1028,19 @@ def test_performance_warning_for_poor_alignment(self):
 #------------------------------------
 # slightly more complex ops
 
-class TestOperationsNumExprPandas(unittest.TestCase):
+class TestOperationsNumExprPandas(tm.TestCase):
 
     @classmethod
     def setUpClass(cls):
+        super(TestOperationsNumExprPandas, cls).setUpClass()
         skip_if_no_ne()
         cls.engine = 'numexpr'
         cls.parser = 'pandas'
         cls.arith_ops = expr._arith_ops_syms + expr._cmp_ops_syms
 
     @classmethod
     def tearDownClass(cls):
+        super(TestOperationsNumExprPandas, cls).tearDownClass()
         del cls.engine, cls.parser
 
     def eval(self, *args, **kwargs):
@@ -1337,6 +1343,7 @@ class TestOperationsNumExprPython(TestOperationsNumExprPandas):
 
     @classmethod
     def setUpClass(cls):
+        super(TestOperationsNumExprPython, cls).setUpClass()
         if not _USE_NUMEXPR:
             raise nose.SkipTest("numexpr engine not installed")
         cls.engine = 'numexpr'
@@ -1404,6 +1411,7 @@ class TestOperationsPythonPython(TestOperationsNumExprPython):
 
     @classmethod
     def setUpClass(cls):
+        super(TestOperationsPythonPython, cls).setUpClass()
         cls.engine = cls.parser = 'python'
         cls.arith_ops = expr._arith_ops_syms + expr._cmp_ops_syms
         cls.arith_ops = filter(lambda x: x not in ('in', 'not in'),
@@ -1414,6 +1422,7 @@ class TestOperationsPythonPandas(TestOperationsNumExprPandas):
 
     @classmethod
     def setUpClass(cls):
+        super(TestOperationsPythonPandas, cls).setUpClass()
         cls.engine = 'python'
         cls.parser = 'pandas'
         cls.arith_ops = expr._arith_ops_syms + expr._cmp_ops_syms

pandas/core/frame.py

@@ -1563,7 +1563,7 @@ def _ixs(self, i, axis=0, copy=False):
 
                     # a location index by definition
                     i = _maybe_convert_indices(i, len(self._get_axis(axis)))
-                    return self.reindex(i, takeable=True)
+                    return self.reindex(i, takeable=True)._setitem_copy(True)
                 else:
                     new_values, copy = self._data.fast_2d_xs(i, copy=copy)
                     return Series(new_values, index=self.columns,
@@ -2714,7 +2714,7 @@ def trans(v):
 
             self._clear_item_cache()
         else:
-            return self.take(indexer, axis=axis, convert=False)
+            return self.take(indexer, axis=axis, convert=False, is_copy=False)
 
     def sortlevel(self, level=0, axis=0, ascending=True, inplace=False):
         """
@@ -2760,7 +2760,7 @@ def sortlevel(self, level=0, axis=0, ascending=True, inplace=False):
 
             self._clear_item_cache()
         else:
-            return self.take(indexer, axis=axis, convert=False)
+            return self.take(indexer, axis=axis, convert=False, is_copy=False)
 
     def swaplevel(self, i, j, axis=0):
         """

pandas/core/generic.py

@@ -1064,7 +1064,7 @@ def __delitem__(self, key):
         except KeyError:
             pass
 
-    def take(self, indices, axis=0, convert=True):
+    def take(self, indices, axis=0, convert=True, is_copy=True):
         """
         Analogous to ndarray.take
 
@@ -1073,6 +1073,7 @@ def take(self, indices, axis=0, convert=True):
         indices : list / array of ints
         axis : int, default 0
         convert : translate neg to pos indices (default)
+        is_copy : mark the returned frame as a copy
 
         Returns
         -------
@@ -1090,12 +1091,17 @@ def take(self, indices, axis=0, convert=True):
             labels = self._get_axis(axis)
             new_items = labels.take(indices)
             new_data = self._data.reindex_axis(new_items, indexer=indices,
-                                               axis=0)
+                                               axis=baxis)
         else:
-            new_data = self._data.take(indices, axis=baxis, verify=convert)
-        return self._constructor(new_data)\
-                   ._setitem_copy(True)\
-                   .__finalize__(self)
+            new_data = self._data.take(indices, axis=baxis)
+
+        result = self._constructor(new_data).__finalize__(self)
+
+        # maybe set copy if we didn't actually change the index
+        if is_copy and not result._get_axis(axis).equals(self._get_axis(axis)):
+            result = result._setitem_copy(is_copy)
+
+        return result
 
     # TODO: Check if this was clearer in 0.12
     def select(self, crit, axis=0):

pandas/core/indexing.py

@@ -209,6 +209,7 @@ def _setitem_with_indexer(self, indexer, value):
                     labels = _safe_append_to_index(index, key)
                     self.obj._data = self.obj.reindex_axis(labels, i)._data
                     self.obj._maybe_update_cacher(clear=True)
+                    self.obj._setitem_copy(False)
 
                     if isinstance(labels, MultiIndex):
                         self.obj.sortlevel(inplace=True)

pandas/io/tests/test_clipboard.py

@@ -1,5 +1,3 @@
-import unittest
-
 import numpy as np
 from numpy.random import randint
 
@@ -18,9 +16,10 @@
     raise nose.SkipTest("no clipboard found")
 
 
-class TestClipboard(unittest.TestCase):
+class TestClipboard(tm.TestCase):
     @classmethod
     def setUpClass(cls):
+        super(TestClipboard, cls).setUpClass()
         cls.data = {}
         cls.data['string'] = mkdf(5, 3, c_idx_type='s', r_idx_type='i',
                                   c_idx_names=[None], r_idx_names=[None])
@@ -43,6 +42,7 @@ def setUpClass(cls):
 
     @classmethod
     def tearDownClass(cls):
+        super(TestClipboard, cls).tearDownClass()
         del cls.data_types, cls.data
 
     def check_round_trip_frame(self, data_type, excel=None, sep=None):

pandas/io/tests/test_cparser.py

@@ -9,7 +9,6 @@
 import os
 import sys
 import re
-import unittest
 
 import nose
 
@@ -32,7 +31,7 @@
 import pandas.parser as parser
 
 
-class TestCParser(unittest.TestCase):
+class TestCParser(tm.TestCase):
 
     def setUp(self):
         self.dirpath = tm.get_data_path()
@@ -132,7 +131,7 @@ def test_integer_thousands(self):
 
         expected = [123456, 12500]
         tm.assert_almost_equal(result[0], expected)
-        
+
     def test_integer_thousands_alt(self):
         data = '123.456\n12.500'
 

pandas/io/tests/test_data.py

@@ -1,6 +1,5 @@
 from __future__ import print_function
 from pandas import compat
-import unittest
 import warnings
 import nose
 from nose.tools import assert_equal
@@ -39,15 +38,17 @@ def assert_n_failed_equals_n_null_columns(wngs, obj, cls=SymbolWarning):
     assert msgs.str.contains('|'.join(failed_symbols)).all()
 
 
-class TestGoogle(unittest.TestCase):
+class TestGoogle(tm.TestCase):
     @classmethod
     def setUpClass(cls):
+        super(TestGoogle, cls).setUpClass()
         cls.locales = tm.get_locales(prefix='en_US')
         if not cls.locales:
             raise nose.SkipTest("US English locale not available for testing")
 
     @classmethod
     def tearDownClass(cls):
+        super(TestGoogle, cls).tearDownClass()
         del cls.locales
 
     @network
@@ -109,9 +110,10 @@ def test_get_multi2(self):
                 assert_n_failed_equals_n_null_columns(w, result)
 
 
-class TestYahoo(unittest.TestCase):
+class TestYahoo(tm.TestCase):
     @classmethod
     def setUpClass(cls):
+        super(TestYahoo, cls).setUpClass()
         _skip_if_no_lxml()
 
     @network
@@ -228,9 +230,10 @@ def test_get_date_ret_index(self):
         assert np.issubdtype(pan.values.dtype, np.floating)
 
 
-class TestYahooOptions(unittest.TestCase):
+class TestYahooOptions(tm.TestCase):
     @classmethod
     def setUpClass(cls):
+        super(TestYahooOptions, cls).setUpClass()
         _skip_if_no_lxml()
 
         # aapl has monthlies
@@ -245,6 +248,7 @@ def setUpClass(cls):
 
     @classmethod
     def tearDownClass(cls):
+        super(TestYahooOptions, cls).tearDownClass()
         del cls.aapl, cls.expiry
 
     @network
@@ -287,9 +291,10 @@ def test_get_put_data(self):
             assert len(puts)>1
 
 
-class TestOptionsWarnings(unittest.TestCase):
+class TestOptionsWarnings(tm.TestCase):
     @classmethod
     def setUpClass(cls):
+        super(TestOptionsWarnings, cls).setUpClass()
         _skip_if_no_lxml()
 
         with assert_produces_warning(FutureWarning):
@@ -304,6 +309,7 @@ def setUpClass(cls):
 
     @classmethod
     def tearDownClass(cls):
+        super(TestOptionsWarnings, cls).tearDownClass()
         del cls.aapl, cls.year, cls.month
 
     @network
@@ -346,7 +352,7 @@ def test_get_put_data_warning(self):
                 warnings.warn("IndexError thrown no tables found")
 
 
-class TestDataReader(unittest.TestCase):
+class TestDataReader(tm.TestCase):
     def test_is_s3_url(self):
         from pandas.io.common import _is_s3_url
         self.assert_(_is_s3_url("s3://pandas/somethingelse.com"))
@@ -376,7 +382,7 @@ def test_read_famafrench(self):
             assert isinstance(ff, dict)
 
 
-class TestFred(unittest.TestCase):
+class TestFred(tm.TestCase):
     @network
     def test_fred(self):
         """

pandas/io/tests/test_date_converters.py

@@ -4,7 +4,6 @@
 import os
 import sys
 import re
-import unittest
 
 import nose
 
@@ -22,9 +21,9 @@
 from pandas import compat
 from pandas.lib import Timestamp
 import pandas.io.date_converters as conv
+import pandas.util.testing as tm
 
-
-class TestConverters(unittest.TestCase):
+class TestConverters(tm.TestCase):
 
     def setUp(self):
         self.years = np.array([2007, 2008])