keep index name with to_datetime (dask#4905)

bolliger32 · jcrist · commit 7a9cfaf35f94 · 2019-06-12T13:44:25.000-05:00
* keep index name with to_datetime

* allow users to pass meta

* Update dask/dataframe/core.py

put meta as explicit kwarg

Co-Authored-By: Matthew Rocklin &lt;mrocklin@gmail.com&gt;

* Update dask/dataframe/core.py

remove meta kwargs.pop

Co-Authored-By: Matthew Rocklin &lt;mrocklin@gmail.com&gt;

* remove test for index

* allow index
diff --git a/dask/dataframe/core.py b/dask/dataframe/core.py
@@ -4877,8 +4877,15 @@ def maybe_shift_divisions(df, periods, freq):
 
 
 @wraps(pd.to_datetime)
-def to_datetime(arg, **kwargs):
-    meta = pd.Series([pd.Timestamp('2000')])
+def to_datetime(arg, meta=None, **kwargs):
+    if meta is None:
+        if isinstance(arg, Index):
+            meta = pd.DatetimeIndex([])
+            meta.name = arg.name
+        else:
+            meta = pd.Series([pd.Timestamp('2000')])
+            meta.index = meta.index.astype(arg.index.dtype)
+            meta.index.name = arg.index.name
     return map_partitions(pd.to_datetime, arg, meta=meta, **kwargs)
 
 
diff --git a/dask/dataframe/tests/test_dataframe.py b/dask/dataframe/tests/test_dataframe.py
@@ -3261,15 +3261,20 @@ def test_to_datetime():
     df = pd.DataFrame({'year': [2015, 2016],
                        'month': [2, 3],
                        'day': [4, 5]})
+    df.index.name = 'ix'
     ddf = dd.from_pandas(df, npartitions=2)
 
     assert_eq(pd.to_datetime(df), dd.to_datetime(ddf))
 
     s = pd.Series(['3/11/2000', '3/12/2000', '3/13/2000'] * 100)
-    ds = dd.from_pandas(s, npartitions=10)
+    s.index = s.values
+    ds = dd.from_pandas(s, npartitions=10, sort=False)
 
     assert_eq(pd.to_datetime(s, infer_datetime_format=True),
               dd.to_datetime(ds, infer_datetime_format=True))
+    assert_eq(pd.to_datetime(s.index, infer_datetime_format=True),
+              dd.to_datetime(ds.index, infer_datetime_format=True),
+              check_divisions=False)
 
 
 def test_to_timedelta():
