From ca54721058afbfce8cf25713446f6c88624b7bf0 Mon Sep 17 00:00:00 2001
From: Scott Sanderson <ssanderson@quantopian.com>
Date: Wed, 20 Jul 2016 20:03:35 -0400
Subject: [PATCH] MAINT: Pandas compat for rolling_*.

---
 tests/pipeline/test_engine.py | 36 ++++++++++-----
 zipline/utils/pandas_utils.py | 82 +++++++++++++++++++++++++++++++++++
 2 files changed, 107 insertions(+), 11 deletions(-)

diff --git a/tests/pipeline/test_engine.py b/tests/pipeline/test_engine.py
index e656cbae..335cda44 100644
--- a/tests/pipeline/test_engine.py
+++ b/tests/pipeline/test_engine.py
@@ -25,12 +25,8 @@ from pandas import (
     Categorical,
     DataFrame,
     date_range,
-    ewma,
-    ewmstd,
     Int64Index,
     MultiIndex,
-    rolling_apply,
-    rolling_mean,
     Series,
     Timestamp,
 )
@@ -84,6 +80,12 @@ from zipline.testing.fixtures import (
 )
 from zipline.utils.memoize import lazyval
 from zipline.utils.numpy_utils import bool_dtype, datetime64ns_dtype
+from zipline.utils.pandas_utils import (
+    ewma,
+    ewmstd,
+    rolling_apply,
+    rolling_mean,
+)
 
 
 class RollingSumDifference(CustomFactor):
@@ -1009,14 +1011,16 @@ class SyntheticBcolzTestCase(WithAdjustmentReader,
         # computed results to be computed using values anchored on the
         # **previous** day's data.
         expected_raw = rolling_mean(
-            expected_bar_values_2d(
-                dates - self.trading_calendar.day,
-                self.equity_info,
-                'close',
+            DataFrame(
+                expected_bar_values_2d(
+                    dates - self.trading_calendar.day,
+                    self.equity_info,
+                    'close',
+                ),
             ),
             window_length,
             min_periods=1,
-        )
+        ).values
 
         expected = DataFrame(
             # Truncate off the extra rows needed to compute the SMAs.
@@ -1122,19 +1126,29 @@ class ParameterizedFactorTestCase(WithTradingEnvironment, ZiplineTestCase):
     def expected_ewma(self, window_length, decay_rate):
         alpha = 1 - decay_rate
         span = (2 / alpha) - 1
+
+        # XXX: This is a comically inefficient way to compute a windowed EWMA.
+        # Don't use it outside of testing.  We're using rolling-apply of an
+        # ewma (which is itself a rolling-window function) because we only want
+        # to look at ``window_length`` rows at a time.
         return rolling_apply(
             self.raw_data,
             window_length,
-            lambda window: ewma(window, span=span)[-1],
+            lambda window: ewma(DataFrame(window), span=span).values[-1],
         )[window_length:]
 
     def expected_ewmstd(self, window_length, decay_rate):
         alpha = 1 - decay_rate
         span = (2 / alpha) - 1
+
+        # XXX: This is a comically inefficient way to compute a windowed EWMSTD.
+        # Don't use it outside of testing.  We're using rolling-apply of an
+        # ewma (which is itself a rolling-window function) because we only want
+        # to look at ``window_length`` rows at a time.
         return rolling_apply(
             self.raw_data,
             window_length,
-            lambda window: ewmstd(window, span=span)[-1],
+            lambda window: ewmstd(DataFrame(window), span=span).values[-1],
         )[window_length:]
 
     @parameterized.expand([
diff --git a/zipline/utils/pandas_utils.py b/zipline/utils/pandas_utils.py
index 3e58230a..a38b437a 100644
--- a/zipline/utils/pandas_utils.py
+++ b/zipline/utils/pandas_utils.py
@@ -160,3 +160,85 @@ def timedelta_to_integral_minutes(delta):
     Convert a pd.Timedelta to a number of minutes as an int.
     """
     return timedelta_to_integral_seconds(delta) // 60
+
+
+# Remove when we drop support for 0.17
+if pandas_version >= StrictVersion('0.18'):
+    def rolling_mean(arg,
+                     window,
+                     min_periods=None,
+                     freq=None,
+                     center=False,
+                     **kwargs):
+        return arg.rolling(
+            window,
+            min_periods=min_periods,
+            freq=freq,
+            center=center,
+            **kwargs
+        ).mean()
+
+    def rolling_apply(arg,
+                      window,
+                      func,
+                      min_periods=None,
+                      freq=None,
+                      center=False,
+                      **kwargs):
+        return arg.rolling(
+            window,
+            min_periods=min_periods,
+            freq=freq,
+            center=center,
+            **kwargs
+        ).apply(func)
+
+    def ewma(arg,
+             com=None,
+             span=None,
+             halflife=None,
+             alpha=None,
+             min_periods=0,
+             freq=None,
+             adjust=True,
+             how=None,
+             ignore_na=False):
+
+        return arg.ewm(
+            com=com,
+            span=span,
+            halflife=halflife,
+            alpha=alpha,
+            min_periods=min_periods,
+            freq=freq,
+            adjust=adjust,
+            ignore_na=ignore_na,
+        ).mean()
+
+    def ewmstd(arg,
+               com=None,
+               span=None,
+               halflife=None,
+               alpha=None,
+               min_periods=0,
+               freq=None,
+               adjust=True,
+               how=None,
+               ignore_na=False):
+
+        return arg.ewm(
+            com=com,
+            span=span,
+            halflife=halflife,
+            alpha=alpha,
+            min_periods=min_periods,
+            freq=freq,
+            adjust=adjust,
+            ignore_na=ignore_na,
+        ).std()
+
+else:
+    rolling_mean = pd.rolling_mean
+    rolling_apply = pd.rolling_apply
+    ewma = pd.ewma
+    ewmstd = pd.ewmstd