ENH: Enable minute bar data with TALib transform.

Add a `bars` keyword arg, as is used with BatchTransform.

Also, instead of overwriting the window_length kwarg with timeperiod,
always use the lookback value from the created TALib function,
as timeperiod will be an input into that value if it exists.

Calculate `window_length` in minute mode so that there are enough
days to cover the minutes in the timeperiod.

tests/test_transforms.py

@@ -372,3 +372,24 @@ class TestTALIB(TestCase):
             talib_data['close'] = data['price'][0].values
             expected_result = talib_fn(talib_data, **t.call_kwargs)[-1]
             np.testing.assert_allclose(talib_result, expected_result)
+
+    def test_talib_with_minute_data(self):
+
+        ma_one_day_minutes = ta.MA(timeperiod=10, bars='minute')
+
+        # Assert that the BatchTransform window length is enough to cover
+        # the amount of minutes in the timeperiod.
+
+        # Here, 10 minutes only needs a window length of 1.
+        self.assertEquals(1, ma_one_day_minutes.window_length)
+
+        # With minutes greater than the 390, i.e. one trading day, we should
+        # have a window_length of two days.
+        ma_two_day_minutes = ta.MA(timeperiod=490, bars='minute')
+        self.assertEquals(2, ma_two_day_minutes.window_length)
+
+        # TODO: Ensure that the lookback into the datapanel is returning
+        # expected results.
+        # Requires supplying minute instead of day data to the unit test.
+        # When adding test data, should add more minute events than the
+        # timeperiod to ensure that lookback is behaving properly.

zipline/transforms/ta.py

@@ -12,6 +12,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
+import math
 
 import numpy as np
 import talib
@@ -86,6 +87,7 @@ def make_transform(talib_fn, name):
                      low='low',
                      volume='volume',
                      refresh_period=0,
+                     bars='daily',
                      **kwargs):
 
             key_map = {'high': high,
@@ -94,11 +96,6 @@ def make_transform(talib_fn, name):
                        'volume': volume,
                        'close': close}
 
-            # Rename timeperiod to window_length to conform with
-            # TALib interface.
-            if 'timeperiod' in kwargs:
-                kwargs['window_length'] = kwargs['timeperiod']
-
             self.call_kwargs = kwargs
 
             # Make deepcopy of talib abstract function.
@@ -119,6 +116,15 @@ def make_transform(talib_fn, name):
             # get the lookback
             self.lookback = self.talib_fn.lookback
 
+            self.bars = bars
+            if bars == 'daily':
+                lookback = self.lookback + 1
+            elif bars == 'minute':
+                lookback = int(math.ceil(self.lookback / (6.5 * 60)))
+
+            # Ensure that window_length is at least 1 day's worth of data.
+            window_length = max(lookback, 1)
+
             def zipline_wrapper(data):
                 # get required TA-Lib input names
                 if 'price' in self.talib_fn.input_names:
@@ -157,7 +163,7 @@ def make_transform(talib_fn, name):
             super(TALibTransform, self).__init__(
                 func=zipline_wrapper,
                 refresh_period=refresh_period,
-                window_length=max(1, self.lookback + 1))
+                window_length=window_length)
 
         def __repr__(self):
             return 'Zipline BatchTransform: {0}'.format(