BUG: fix DataFrame.info with non-unique columns close #1700

Author: wesm

RELEASE.rst

@@ -50,6 +50,7 @@ pandas 0.8.2
   - Fixed GroupBy bugs interacting with DatetimeIndex asof / map methods (#1677)
   - Handle factors with NAs in pandas.rpy (#1615)
   - Fix statsmodels import in pandas.stats.var (#1734)
+  - Fix DataFrame repr/info summary with non-unique columns (#1700)
 
 pandas 0.8.1
 ============

pandas/core/frame.py

@@ -1340,7 +1340,8 @@ def convert_objects(self):
 
     def get_dtype_counts(self):
         counts = {}
-        for _, series in self.iterkv():
+        for i in range(len(self.columns)):
+            series = self.icol(i)
             # endianness can cause dtypes to look different
             dtype_str = str(series.dtype)
             if dtype_str in counts:

pandas/tests/test_frame.py

@@ -3491,6 +3491,14 @@ def test_info(self):
         frame.info(verbose=False)
         sys.stdout = sys.__stdout__
 
+    def test_info_duplicate_columns(self):
+        io = StringIO()
+
+        # it works!
+        frame = DataFrame(np.random.randn(1500, 4),
+                          columns=['a', 'a', 'b', 'b'])
+        frame.info(buf=io)
+
     def test_dtypes(self):
         self.mixed_frame['bool'] = self.mixed_frame['A'] > 0
         result = self.mixed_frame.dtypes