Merge pull request #2309 from quantopian/float-precision-fix

BUG: Round values before truncating when writing bcolz minute/day tables.

tests/data/test_minute_bars.py

@@ -142,6 +142,38 @@ class BcolzMinuteBarTestCase(WithTradingCalendars,
 
         self.assertEquals(50.0, volume_price)
 
+    def test_precision_after_scaling(self):
+        '''For numbers that don't have an exact float representation,
+        assert that scaling the value does not cause a loss in precision.
+        '''
+        minute = self.market_opens[self.test_calendar_start]
+        sid = 1
+        data = DataFrame(
+            data={
+                'open': [130.23],
+                'high': [130.23],
+                'low': [130.23],
+                'close': [130.23],
+                'volume': [1000]
+            },
+            index=[minute])
+        self.writer.write_sid(sid, data)
+
+        open_price = self.reader.get_value(sid, minute, 'open')
+        self.assertEquals(130.23, open_price)
+
+        high_price = self.reader.get_value(sid, minute, 'high')
+        self.assertEquals(130.23, high_price)
+
+        low_price = self.reader.get_value(sid, minute, 'low')
+        self.assertEquals(130.23, low_price)
+
+        close_price = self.reader.get_value(sid, minute, 'close')
+        self.assertEquals(130.23, close_price)
+
+        volume_price = self.reader.get_value(sid, minute, 'volume')
+        self.assertEquals(1000, volume_price)
+
     def test_write_one_ohlcv_with_ratios(self):
         minute = self.market_opens[self.test_calendar_start]
         sid = 1

tests/test_data_portal.py

@@ -406,7 +406,7 @@ class DataPortalTestBase(WithDataPortal,
         day = calendar.day
         dividend_date = self.trading_days[2]
 
-        prev_day_price = 1.005
+        prev_day_price = 1.006
         dividend_amount = 0.5  # see self.make_dividends_data
         ratio = 1.0 - dividend_amount / prev_day_price
 
@@ -541,10 +541,10 @@ class DataPortalTestBase(WithDataPortal,
 
         # Equity prices should be floored to three decimal places.
         expected_equity_values = {
-            'open': 1.005,
-            'high': 1.005,
+            'open': 1.006,
+            'high': 1.006,
             'low': 1.005,
-            'close': 1.005,
+            'close': 1.006,
             'volume': expected_equity_volume,
         }
         # Futures prices should be rounded to four decimal places.

zipline/data/bcolz_daily_bars.py

@@ -365,7 +365,7 @@ class BcolzDailyBarWriter(object):
             return raw_data
 
         winsorise_uint32(raw_data, invalid_data_behavior, 'volume', *OHLC)
-        processed = (raw_data[list(OHLC)] * 1000).astype('uint32')
+        processed = (raw_data[list(OHLC)] * 1000).round().astype('uint32')
         dates = raw_data.index.values.astype('datetime64[s]')
         check_uint32_safe(dates.max().view(np.int64), 'day')
         processed['day'] = dates.astype('uint32')

zipline/data/minute_bars.py

@@ -131,10 +131,10 @@ def convert_cols(cols, scale_factor, sid, invalid_data_behavior):
         If 'warn', logs a warning and filters out incompatible values.
         If 'ignore', silently filters out incompatible values.
     """
-    scaled_opens = np.nan_to_num(cols['open']) * scale_factor
-    scaled_highs = np.nan_to_num(cols['high']) * scale_factor
-    scaled_lows = np.nan_to_num(cols['low']) * scale_factor
-    scaled_closes = np.nan_to_num(cols['close']) * scale_factor
+    scaled_opens = (np.nan_to_num(cols['open']) * scale_factor).round()
+    scaled_highs = (np.nan_to_num(cols['high']) * scale_factor).round()
+    scaled_lows = (np.nan_to_num(cols['low']) * scale_factor).round()
+    scaled_closes = (np.nan_to_num(cols['close']) * scale_factor).round()
 
     exclude_mask = np.zeros_like(scaled_opens, dtype=bool)