From 2c4356eee5587fddaf3e1c1fa23e9fafe25ad79a Mon Sep 17 00:00:00 2001
From: jreback <jeff@reback.net>
Date: Thu, 12 Sep 2013 12:55:43 -0400
Subject: [PATCH] PERF: much faster to_datetime performance with a format of
 '%Y%m%d'

---
 doc/source/release.rst                  |  1 +
 pandas/tseries/tests/test_timeseries.py | 15 +++++++++++++++
 pandas/tseries/tools.py                 | 14 +++++++++++++-
 vb_suite/timeseries.py                  | 18 ++++++++++++++++++
 4 files changed, 47 insertions(+), 1 deletion(-)

diff --git a/doc/source/release.rst b/doc/source/release.rst
index 75194f6877a6e..5376e0396799e 100644
--- a/doc/source/release.rst
+++ b/doc/source/release.rst
@@ -105,6 +105,7 @@ Improvements to existing features
     test to vbench (:issue:`4705` and :issue:`4722`)
   - Add ``axis`` and ``level`` keywords to ``where``, so that the ``other`` argument
     can now be an alignable pandas object.
+  - ``to_datetime`` with a format of 'YYYYMMDD' now parses much faster
 
 API Changes
 ~~~~~~~~~~~
diff --git a/pandas/tseries/tests/test_timeseries.py b/pandas/tseries/tests/test_timeseries.py
index b5697a98de412..c9e643e25b761 100644
--- a/pandas/tseries/tests/test_timeseries.py
+++ b/pandas/tseries/tests/test_timeseries.py
@@ -834,6 +834,21 @@ def test_to_datetime_format(self):
                 else:
                     self.assert_(result.equals(expected))
 
+    def test_to_datetime_format_YYYYMMDD(self):
+        s = Series([19801222,19801222] + [19810105]*5)
+        expected = Series([ Timestamp(x) for x in s.apply(str) ])
+
+        result = to_datetime(s,format='%Y%m%d')
+        assert_series_equal(result, expected)
+
+        result = to_datetime(s.apply(str),format='%Y%m%d')
+        assert_series_equal(result, expected)
+
+        # with NaT
+        s[2] = np.nan
+        self.assertRaises(ValueError, to_datetime, s,format='%Y%m%d')
+        self.assertRaises(ValueError, to_datetime, s.apply(str),format='%Y%m%d')
+
     def test_to_datetime_format_microsecond(self):
         val = '01-Apr-2011 00:00:01.978'
         format = '%d-%b-%Y %H:%M:%S.%f'
diff --git a/pandas/tseries/tools.py b/pandas/tseries/tools.py
index 3087d54396691..cca4850c2c1bf 100644
--- a/pandas/tseries/tools.py
+++ b/pandas/tseries/tools.py
@@ -101,7 +101,19 @@ def _convert_listlike(arg, box):
         arg = com._ensure_object(arg)
         try:
             if format is not None:
-                result = tslib.array_strptime(arg, format)
+                result = None
+
+                # shortcut formatting here
+                if format == '%Y%m%d':
+                    try:
+                        carg = arg.astype(np.int64).astype(object)
+                        result = lib.try_parse_year_month_day(carg/10000,carg/100 % 100, carg % 100)
+                    except:
+                        raise ValueError("cannot convert the input to '%Y%m%d' date format")
+
+                # fallback
+                if result is None:
+                    result = tslib.array_strptime(arg, format)
             else:
                 result = tslib.array_to_datetime(arg, raise_=errors == 'raise',
                                                  utc=utc, dayfirst=dayfirst,
diff --git a/vb_suite/timeseries.py b/vb_suite/timeseries.py
index 4dd1dd2e96bdd..999c3869daf62 100644
--- a/vb_suite/timeseries.py
+++ b/vb_suite/timeseries.py
@@ -147,6 +147,24 @@ def date_range(start=None, end=None, periods=None, freq=None):
     Benchmark('to_datetime(strings)', setup,
               start_date=datetime(2012, 7, 11))
 
+setup = common_setup + """
+rng = date_range('1/1/2000', periods=10000, freq='D')
+strings = Series(rng.year*10000+rng.month*100+rng.day,dtype=np.int64).apply(str)
+"""
+
+timeseries_to_datetime_YYYYMMDD = \
+    Benchmark('to_datetime(strings,format="%Y%m%d")', setup,
+              start_date=datetime(2013, 9, 1))
+
+setup = common_setup + """
+rng = date_range('1/1/2000', periods=10000, freq='D')
+strings = Series(rng.year*10000+rng.month*100+rng.day,dtype=np.int64).apply(str)
+"""
+
+timeseries_to_datetime_YYYYMMDD_old = \
+    Benchmark('pandas.tslib.array_strptime(strings.values,"%Y%m%d")', setup,
+              start_date=datetime(2013, 9, 1))
+
 # ---- infer_freq
 # infer_freq