From a776b0ce33f28ed50dd42f4946f2239e2cadb95e Mon Sep 17 00:00:00 2001
From: jreback <jeff@reback.net>
Date: Sat, 19 Jan 2013 23:00:16 -0500
Subject: [PATCH 1/3] DOC: docs for enhancement for using np.nan with
 datetim64[ns]

---
 doc/source/v0.10.1.txt | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/doc/source/v0.10.1.txt b/doc/source/v0.10.1.txt
index 8aa2dad2b35a0..e4083f745e77e 100644
--- a/doc/source/v0.10.1.txt
+++ b/doc/source/v0.10.1.txt
@@ -14,6 +14,21 @@ API changes
 New features
 ~~~~~~~~~~~~
 
+Datetime64[ns] columns in a DataFrame (or a Series) allow the use of ``np.nan`` to indicate a nan value, in additional to the traditional ``NaT``, or not-a-time. This allows convenient nan setting in a generic way. Furthermore datetime64 columns are created by default, when using ``datetime.datetime`` or ``Timestamp`` objects, rather than default as object dtype.
+
+.. ipython:: python
+
+   df = DataFrame(randn(6,2),date_range('20010102',periods=6),columns=['A','B'])
+   df['timestamp'] = Timestamp('20010103')
+   df
+
+   # datetime64[ns] out of the box
+   df.get_dtype_counts()
+
+   # use the traditional nan, which is mapped to NaT internally
+   df.ix[2:4,['A','timestamp']] = np.nan
+   df
+
 HDFStore
 ~~~~~~~~
 
@@ -59,7 +74,7 @@ You can now store ``datetime64`` in data columns
 
     df_mixed               = df.copy()
     df_mixed['datetime64'] = Timestamp('20010102')
-    df_mixed.ix[3:4,['A','B']] = np.nan
+    df_mixed.ix[3:4,['A','B','datetime64']] = np.nan
 
     store.append('df_mixed', df_mixed)
     df_mixed1 = store.select('df_mixed')

From 44564168ab30d787ed7fb53cb25c8c2a18fe73a4 Mon Sep 17 00:00:00 2001
From: jreback <jeff@reback.net>
Date: Mon, 21 Jan 2013 09:00:54 -0500
Subject: [PATCH 2/3] CLN: changed datetime64 testing in applymap to use
 is_datetime64_dtype idiom

---
 pandas/core/frame.py | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/pandas/core/frame.py b/pandas/core/frame.py
index 301ea9d28d001..4c3922d313684 100644
--- a/pandas/core/frame.py
+++ b/pandas/core/frame.py
@@ -4287,9 +4287,9 @@ def applymap(self, func):
         applied : DataFrame
         """
 
-        # if we have a dtype == 'M8[ns]', provide boxed values
+        # if we have a datetime64 type, provide boxed values
         def infer(x):
-            if x.dtype == 'M8[ns]':
+            if com.is_datetime64_dtype(x):
                 x = lib.map_infer(x, lib.Timestamp)
             return lib.map_infer(x, func)
         return self.apply(infer)

From 83cceab68b82c42a96b9e5393e800ceae0d039d2 Mon Sep 17 00:00:00 2001
From: jreback <jeff@reback.net>
Date: Mon, 21 Jan 2013 21:25:03 -0500
Subject: [PATCH 3/3] DOC: fixed v0.10.1 whatsnew for datetime64[ns] series

---
 doc/source/v0.10.1.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/source/v0.10.1.txt b/doc/source/v0.10.1.txt
index e4083f745e77e..a2b99bf6c4174 100644
--- a/doc/source/v0.10.1.txt
+++ b/doc/source/v0.10.1.txt
@@ -14,7 +14,7 @@ API changes
 New features
 ~~~~~~~~~~~~
 
-Datetime64[ns] columns in a DataFrame (or a Series) allow the use of ``np.nan`` to indicate a nan value, in additional to the traditional ``NaT``, or not-a-time. This allows convenient nan setting in a generic way. Furthermore datetime64 columns are created by default, when using ``datetime.datetime`` or ``Timestamp`` objects, rather than default as object dtype.
+Datetime64[ns] columns in a DataFrame (or a Series) allow the use of ``np.nan`` to indicate a nan value, in addition to the traditional ``NaT``, or not-a-time. This allows convenient nan setting in a generic way. Furthermore datetime64 columns are created by default, when using ``datetime.datetime`` or ``Timestamp`` objects, rather than default as object dtype.
 
 .. ipython:: python