Merge pull request #9574 from schmohlio/fix-plot-legends

fixing pandas.DataFrame.plot(): labels do not appear in legend and label kwd

Author: Tom Augspurger

Diff for: doc/source/whatsnew/v0.16.1.txt

@@ -50,6 +50,8 @@ Performance Improvements
 Bug Fixes
 ~~~~~~~~~
 
+- Fixed bug (:issue:`9542`) where labels did not appear properly in legend of ``DataFrame.plot()``. Passing ``label=`` args also now works, and series indices are no longer mutated. 
+
 
 
 

Diff for: pandas/tests/test_graphics.py

@@ -1064,12 +1064,6 @@ def test_implicit_label(self):
         ax = df.plot(x='a', y='b')
         self._check_text_labels(ax.xaxis.get_label(), 'a')
 
-    @slow
-    def test_explicit_label(self):
-        df = DataFrame(randn(10, 3), columns=['a', 'b', 'c'])
-        ax = df.plot(x='a', y='b', label='LABEL')
-        self._check_text_labels(ax.xaxis.get_label(), 'LABEL')
-
     @slow
     def test_donot_overwrite_index_name(self):
         # GH 8494
@@ -2542,6 +2536,20 @@ def test_df_legend_labels(self):
         ax = df3.plot(kind='scatter', x='g', y='h', label='data3', ax=ax)
         self._check_legend_labels(ax, labels=['data1', 'data3'])
 
+        # ensure label args pass through and
+        # index name does not mutate
+        # column names don't mutate
+        df5 = df.set_index('a')
+        ax = df5.plot(y='b')
+        self._check_legend_labels(ax, labels=['b'])
+        ax = df5.plot(y='b', label='LABEL_b')
+        self._check_legend_labels(ax, labels=['LABEL_b'])
+        self._check_text_labels(ax.xaxis.get_label(), 'a')
+        ax = df5.plot(y='c', label='LABEL_c', ax=ax)
+        self._check_legend_labels(ax, labels=['LABEL_b','LABEL_c'])
+        self.assertTrue(df5.columns.tolist() == ['b','c'])
+
+
     def test_legend_name(self):
         multi = DataFrame(randn(4, 4),
                           columns=[np.array(['a', 'a', 'b', 'b']),

Diff for: pandas/tools/plotting.py

@@ -886,10 +886,11 @@ def _iter_data(self, data=None, keep_index=False, fillna=None):
 
         from pandas.core.frame import DataFrame
         if isinstance(data, (Series, np.ndarray, Index)):
+            label = self.label if self.label is not None else data.name
             if keep_index is True:
-                yield self.label, data
+                yield label, data
             else:
-                yield self.label, np.asarray(data)
+                yield label, np.asarray(data)
         elif isinstance(data, DataFrame):
             if self.sort_columns:
                 columns = com._try_sort(data.columns)
@@ -2306,10 +2307,9 @@ def _plot(data, x=None, y=None, subplots=False,
             if y is not None:
                 if com.is_integer(y) and not data.columns.holds_integer():
                     y = data.columns[y]
-                label = x if x is not None else data.index.name
-                label = kwds.pop('label', label)
+                label = kwds['label'] if 'label' in kwds else y
                 series = data[y].copy()  # Don't modify
-                series.index.name = label
+                series.name = label
 
                 for kw in ['xerr', 'yerr']:
                     if (kw in kwds) and \