Merge pull request pandas-dev#6677 from cpcloud/index-resolvers-after-columns

ENH: define the order of resolution for index vs columns in query/eval

Author: cpcloud

doc/source/indexing.rst

@@ -784,6 +784,29 @@ If instead you don't want to or cannot name your index, you can use the name
    del old_index
 
 
+.. note::
+
+   If the name of your index overlaps with a column name, the column name is
+   given precedence. For example,
+
+   .. ipython:: python
+
+      df = DataFrame({'a': randint(5, size=5)})
+      df.index.name = 'a'
+      df.query('a > 2') # uses the column 'a', not the index
+
+   You can still use the index in a query expression by using the special
+   identifier 'index':
+
+   .. ipython:: python
+
+      df.query('index > 2')
+
+   If for some reason you have a column named ``index``, then you can refer to
+   the index as ``ilevel_0`` as well, but at this point you should consider
+   renaming your columns to something less ambiguous.
+
+
 :class:`~pandas.MultiIndex` :meth:`~pandas.DataFrame.query` Syntax
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 

doc/source/release.rst

@@ -143,6 +143,8 @@ API Changes
 - Following keywords are now acceptable for :meth:`DataFrame.plot(kind='bar')` and :meth:`DataFrame.plot(kind='barh')`.
   - `width`: Specify the bar width. In previous versions, static value 0.5 was passed to matplotlib and it cannot be overwritten.
   - `position`: Specify relative alignments for bar plot layout. From 0 (left/bottom-end) to 1(right/top-end). Default is 0.5 (center). (:issue:`6604`)
+  - Define and document the order of column vs index names in query/eval
+    (:issue:`6676`)
 
 Deprecations
 ~~~~~~~~~~~~

pandas/core/frame.py

@@ -1855,7 +1855,7 @@ def eval(self, expr, **kwargs):
         kwargs['level'] = kwargs.pop('level', 0) + 1
         if resolvers is None:
             index_resolvers = self._get_index_resolvers()
-            resolvers = index_resolvers, dict(self.iteritems())
+            resolvers = dict(self.iteritems()), index_resolvers
         kwargs['target'] = self
         kwargs['resolvers'] = kwargs.get('resolvers', ()) + resolvers
         return _eval(expr, **kwargs)

pandas/tests/test_frame.py

@@ -12583,7 +12583,7 @@ def setUpClass(cls):
         super(TestDataFrameQueryNumExprPandas, cls).setUpClass()
         cls.engine = 'numexpr'
         cls.parser = 'pandas'
-        tm.skip_if_no_ne()
+        tm.skip_if_no_ne(cls.engine)
 
     @classmethod
     def tearDownClass(cls):
@@ -12867,6 +12867,31 @@ def test_query_undefined_local(self):
                                    "local variable 'c' is not defined"):
             df.query('a == @c', engine=engine, parser=parser)
 
+    def test_index_resolvers_come_after_columns_with_the_same_name(self):
+        n = 1
+        a = np.r_[20:101:20]
+
+        df = DataFrame({'index': a, 'b': np.random.randn(a.size)})
+        df.index.name = 'index'
+        result = df.query('index > 5', engine=self.engine, parser=self.parser)
+        expected = df[df['index'] > 5]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({'index': a, 'b': np.random.randn(a.size)})
+        result = df.query('ilevel_0 > 5', engine=self.engine, parser=self.parser)
+        expected = df.loc[df.index[df.index > 5]]
+        tm.assert_frame_equal(result, expected)
+
+        df = DataFrame({'a': a, 'b': np.random.randn(a.size)})
+        df.index.name = 'a'
+        result = df.query('a > 5', engine=self.engine, parser=self.parser)
+        expected = df[df.a > 5]
+        tm.assert_frame_equal(result, expected)
+
+        result = df.query('index > 5', engine=self.engine, parser=self.parser)
+        expected = df.loc[df.index[df.index > 5]]
+        tm.assert_frame_equal(result, expected)
+
 
 class TestDataFrameQueryNumExprPython(TestDataFrameQueryNumExprPandas):