BUG: wide_to_long should check for unique id vars (pandas-dev#16382) (pandas-dev#16403)

* BUG: wide_to_long should check for unique id vars (pandas-dev#16382)

* Fix uncaught lint error

* Add whatsnew note (bug fix)

(cherry picked from commit 04356a8)

Author: erikcs

doc/source/whatsnew/v0.20.2.txt

@@ -83,6 +83,7 @@ Reshaping
 - Bug in construction of a ``DataFrame`` with mixed dtypes including an all-NaT column. (:issue:`16395`)
 
 
+
 Numeric
 ^^^^^^^
 

pandas/core/reshape/reshape.py

@@ -1046,6 +1046,9 @@ def melt_stub(df, stub, i, j, value_vars, sep):
     else:
         i = list(i)
 
+    if df[i].duplicated().any():
+        raise ValueError("the id variables need to uniquely identify each row")
+
     value_vars = list(map(lambda stub:
                           get_var_names(df, stub, sep, suffix), stubnames))
 

pandas/tests/reshape/test_reshape.py

@@ -976,3 +976,14 @@ def test_multiple_id_columns(self):
         exp_frame = exp_frame.set_index(['famid', 'birth', 'age'])[['ht']]
         long_frame = wide_to_long(df, 'ht', i=['famid', 'birth'], j='age')
         tm.assert_frame_equal(long_frame, exp_frame)
+
+    def test_non_unique_idvars(self):
+        # GH16382
+        # Raise an error message if non unique id vars (i) are passed
+        df = pd.DataFrame({
+            'A_A1': [1, 2, 3, 4, 5],
+            'B_B1': [1, 2, 3, 4, 5],
+            'x': [1, 1, 1, 1, 1]
+        })
+        with pytest.raises(ValueError):
+            wide_to_long(df, ['A_A', 'B_B'], i='x', j='colname')