BLD: improved the bundle test suite and related adjustments

catalyst/exchange/ccxt/ccxt_exchange.py

@@ -22,8 +22,10 @@ from catalyst.exchange.exchange_errors import InvalidHistoryFrequencyError, \
     UnsupportedHistoryFrequencyError
 from catalyst.exchange.exchange_execution import ExchangeLimitOrder
 from catalyst.exchange.utils.exchange_utils import mixin_market_params, \
-    from_ms_timestamp, get_epoch, get_exchange_folder, get_catalyst_symbol, \
+    get_exchange_folder, get_catalyst_symbol, \
     get_exchange_auth
+from exchange.utils.datetime_utils import from_ms_timestamp, get_epoch, \
+    get_periods_range
 from catalyst.finance.order import Order, ORDER_STATUS
 from catalyst.finance.transaction import Transaction
 
@@ -399,7 +401,7 @@ class CCXT(Exchange):
             timeframe, source='ccxt', raise_error=raise_error
         )
 
-    def get_candles(self, freq, assets, bar_count=None, start_dt=None,
+    def get_candles(self, freq, assets, bar_count=1, start_dt=None,
                     end_dt=None):
         is_single = (isinstance(assets, TradingPair))
         if is_single:
@@ -416,9 +418,25 @@ class CCXT(Exchange):
                 freqs=freqs,
             )
 
+        if start_dt is not None and end_dt is not None:
+            raise ValueError(
+                'Please provide either start_dt or end_dt, not both.'
+            )
+
+        elif end_dt is not None:
+            dt_range = get_periods_range(
+                end_dt=end_dt,
+                periods=bar_count,
+                freq=freq,
+            )
+            # skip the left bound of the range since the open range is
+            # on the right bound
+            start_dt = dt_range[1]
+
         ms = None
         if start_dt is not None:
-            delta = start_dt - get_epoch()
+            if end_dt is not None:
+                delta = start_dt - get_epoch()
             ms = int(delta.total_seconds()) * 1000
 
         candles = dict()

catalyst/exchange/exchange.py

@@ -13,10 +13,10 @@ from catalyst.exchange.exchange_errors import MismatchingBaseCurrencies, \
     PricingDataNotLoadedError, \
     NoDataAvailableOnExchange, NoValueForField, LastCandleTooEarlyError, \
     TickerNotFoundError, NotEnoughCashError
-from catalyst.exchange.utils.bundle_utils import get_start_dt, \
-    get_delta, get_periods, get_periods_range
+from exchange.utils.datetime_utils import get_delta, get_periods_range, \
+    get_periods, get_start_dt, get_frequency
 from catalyst.exchange.utils.exchange_utils import get_exchange_symbols, \
-    get_frequency, resample_history_df, has_bundle
+    resample_history_df, has_bundle
 from logbook import Logger
 
 log = Logger('Exchange', level=LOG_LEVEL)
@@ -433,7 +433,7 @@ class Exchange:
         series = pd.Series(values, index=dates)
 
         periods = get_periods_range(
-            start_dt, end_dt, data_frequency
+            start_dt=start_dt, end_dt=end_dt, freq=data_frequency
         )
         # TODO: ensure that this working as expected, if not use fillna
         series = series.reindex(
@@ -929,8 +929,7 @@ class Exchange:
         pass
 
     @abstractmethod
-    def get_candles(self, freq, assets, bar_count=None,
-                    start_dt=None, end_dt=None):
+    def get_candles(self, freq, assets, bar_count, start_dt=None, end_dt=None):
         """
         Retrieve OHLCV candles for the given assets
 

catalyst/exchange/exchange_bundle.py

@@ -21,9 +21,9 @@ from catalyst.exchange.exchange_errors import EmptyValuesInBundleError, \
     NoDataAvailableOnExchange, \
     PricingDataNotLoadedError, DataCorruptionError, PricingDataValueError
 from catalyst.exchange.utils.bundle_utils import range_in_bundle, \
-    get_bcolz_chunk, get_month_start_end, \
-    get_year_start_end, get_df_from_arrays, get_start_dt, get_period_label, \
-    get_delta, get_assets
+    get_bcolz_chunk, get_df_from_arrays, get_assets
+from exchange.utils.datetime_utils import get_delta, get_start_dt, \
+    get_period_label, get_month_start_end, get_year_start_end
 from catalyst.exchange.utils.exchange_utils import get_exchange_folder, \
     save_exchange_symbols, mixin_market_params, get_catalyst_symbol
 from catalyst.utils.cli import maybe_show_progress

catalyst/exchange/exchange_data_portal.py

@@ -9,8 +9,8 @@ from catalyst.exchange.exchange_bundle import ExchangeBundle
 from catalyst.exchange.exchange_errors import (
     ExchangeRequestError,
     PricingDataNotLoadedError)
-from catalyst.exchange.utils.exchange_utils import get_frequency, \
-    resample_history_df, group_assets_by_exchange
+from catalyst.exchange.utils.exchange_utils import resample_history_df, group_assets_by_exchange
+from exchange.utils.datetime_utils import get_frequency
 from logbook import Logger
 from redo import retry
 

catalyst/exchange/utils/bundle_utils.py

@@ -1,11 +1,19 @@
-import calendar
 import os
 import tarfile
-from datetime import timedelta, datetime, date
+from datetime import datetime
 
 import numpy as np
 import pandas as pd
-import pytz
+
+from catalyst.data.bundles.core import download_without_progress
+from catalyst.exchange.utils.exchange_utils import get_exchange_bundles_folder
+import os
+import tarfile
+from datetime import datetime
+
+import numpy as np
+import pandas as pd
+
 from catalyst.data.bundles.core import download_without_progress
 from catalyst.exchange.utils.exchange_utils import get_exchange_bundles_folder
 
@@ -13,41 +21,6 @@ EXCHANGE_NAMES = ['bitfinex', 'bittrex', 'poloniex']
 API_URL = 'http://data.enigma.co/api/v1'
 
 
-def get_date_from_ms(ms):
-    """
-    The date from the number of miliseconds from the epoch.
-
-    Parameters
-    ----------
-    ms: int
-
-    Returns
-    -------
-    datetime
-
-    """
-    return datetime.fromtimestamp(ms / 1000.0)
-
-
-def get_seconds_from_date(date):
-    """
-    The number of seconds from the epoch.
-
-    Parameters
-    ----------
-    date: datetime
-
-    Returns
-    -------
-    int
-
-    """
-    epoch = datetime.utcfromtimestamp(0)
-    epoch = epoch.replace(tzinfo=pytz.UTC)
-
-    return int((date - epoch).total_seconds())
-
-
 def get_bcolz_chunk(exchange_name, symbol, data_frequency, period):
     """
     Download and extract a bcolz bundle.
@@ -77,8 +50,8 @@ def get_bcolz_chunk(exchange_name, symbol, data_frequency, period):
     if not os.path.isdir(path):
         url = 'https://s3.amazonaws.com/enigmaco/catalyst-bundles/' \
               'exchange-{exchange}/{name}.tar.gz'.format(
-                exchange=exchange_name,
-                name=name)
+            exchange=exchange_name,
+            name=name)
 
         bytes = download_without_progress(url)
         with tarfile.open('r', fileobj=bytes) as tar:
@@ -87,178 +60,6 @@ def get_bcolz_chunk(exchange_name, symbol, data_frequency, period):
     return path
 
 
-def get_delta(periods, data_frequency):
-    """
-    Get a time delta based on the specified data frequency.
-
-    Parameters
-    ----------
-    periods: int
-    data_frequency: str
-
-    Returns
-    -------
-    timedelta
-
-    """
-    return timedelta(minutes=periods) \
-        if data_frequency == 'minute' else timedelta(days=periods)
-
-
-def get_periods_range(start_dt, end_dt, freq):
-    """
-    Get a date range for the specified parameters.
-
-    Parameters
-    ----------
-    start_dt: datetime
-    end_dt: datetime
-    freq: str
-
-    Returns
-    -------
-    DateTimeIndex
-
-    """
-    if freq == 'minute':
-        freq = 'T'
-
-    elif freq == 'daily':
-        freq = 'D'
-
-    return pd.date_range(start_dt, end_dt, freq=freq)
-
-
-def get_periods(start_dt, end_dt, freq):
-    """
-    The number of periods in the specified range.
-
-    Parameters
-    ----------
-    start_dt: datetime
-    end_dt: datetime
-    freq: str
-
-    Returns
-    -------
-    int
-
-    """
-    return len(get_periods_range(start_dt, end_dt, freq))
-
-
-def get_start_dt(end_dt, bar_count, data_frequency, include_first=True):
-    """
-    The start date based on specified end date and data frequency.
-
-    Parameters
-    ----------
-    end_dt: datetime
-    bar_count: int
-    data_frequency: str
-
-    Returns
-    -------
-    datetime
-
-    """
-    periods = bar_count
-    if periods > 1:
-        delta = get_delta(periods, data_frequency)
-        start_dt = end_dt - delta
-
-        if not include_first:
-            start_dt += get_delta(1, data_frequency)
-    else:
-        start_dt = end_dt
-
-    return start_dt
-
-
-def get_period_label(dt, data_frequency):
-    """
-    The period label for the specified date and frequency.
-
-    Parameters
-    ----------
-    dt: datetime
-    data_frequency: str
-
-    Returns
-    -------
-    str
-
-    """
-    if data_frequency == 'minute':
-        return '{}-{:02d}'.format(dt.year, dt.month)
-    else:
-        return '{}'.format(dt.year)
-
-
-def get_month_start_end(dt, first_day=None, last_day=None):
-    """
-    The first and last day of the month for the specified date.
-
-    Parameters
-    ----------
-    dt: datetime
-    first_day: datetime
-    last_day: datetime
-
-    Returns
-    -------
-    datetime, datetime
-
-    """
-    month_range = calendar.monthrange(dt.year, dt.month)
-
-    if first_day:
-        month_start = first_day
-    else:
-        month_start = pd.to_datetime(datetime(
-            dt.year, dt.month, 1, 0, 0, 0, 0
-        ), utc=True)
-
-    if last_day:
-        month_end = last_day
-    else:
-        month_end = pd.to_datetime(datetime(
-            dt.year, dt.month, month_range[1], 23, 59, 0, 0
-        ), utc=True)
-
-        if month_end > pd.Timestamp.utcnow():
-            month_end = pd.Timestamp.utcnow().floor('1D')
-
-    return month_start, month_end
-
-
-def get_year_start_end(dt, first_day=None, last_day=None):
-    """
-    The first and last day of the year for the specified date.
-
-    Parameters
-    ----------
-
-    dt: datetime
-    first_day: datetime
-    last_day: datetime
-
-    Returns
-    -------
-    datetime, datetime
-
-    """
-    year_start = first_day if first_day \
-        else pd.to_datetime(date(dt.year, 1, 1), utc=True)
-    year_end = last_day if last_day \
-        else pd.to_datetime(date(dt.year, 12, 31), utc=True)
-
-    if year_end > pd.Timestamp.utcnow():
-        year_end = pd.Timestamp.utcnow().floor('1D')
-
-    return year_start, year_end
-
-
 def get_df_from_arrays(arrays, periods):
     """
     A DataFrame from the specified OHCLV arrays.

catalyst/exchange/utils/datetime_utils.py

@@ -0,0 +1,327 @@
+import calendar
+import re
+from datetime import datetime, timedelta, date
+
+import pandas as pd
+import pytz
+
+from exchange.exchange_errors import InvalidHistoryFrequencyError, \
+    InvalidHistoryFrequencyAlias
+
+
+def get_date_from_ms(ms):
+    """
+    The date from the number of miliseconds from the epoch.
+
+    Parameters
+    ----------
+    ms: int
+
+    Returns
+    -------
+    datetime
+
+    """
+    return datetime.fromtimestamp(ms / 1000.0)
+
+
+def get_seconds_from_date(date):
+    """
+    The number of seconds from the epoch.
+
+    Parameters
+    ----------
+    date: datetime
+
+    Returns
+    -------
+    int
+
+    """
+    epoch = datetime.utcfromtimestamp(0)
+    epoch = epoch.replace(tzinfo=pytz.UTC)
+
+    return int((date - epoch).total_seconds())
+
+
+def get_delta(periods, data_frequency):
+    """
+    Get a time delta based on the specified data frequency.
+
+    Parameters
+    ----------
+    periods: int
+    data_frequency: str
+
+    Returns
+    -------
+    timedelta
+
+    """
+    return timedelta(minutes=periods) \
+        if data_frequency == 'minute' else timedelta(days=periods)
+
+
+def get_periods_range(freq, start_dt=None, end_dt=None, periods=None):
+    """
+    Get a date range for the specified parameters.
+
+    Parameters
+    ----------
+    start_dt: datetime
+    end_dt: datetime
+    freq: str
+
+    Returns
+    -------
+    DateTimeIndex
+
+    """
+    if freq == 'minute':
+        freq = 'T'
+
+    elif freq == 'daily':
+        freq = 'D'
+
+    if start_dt is not None and end_dt is not None and periods is None:
+
+        return pd.date_range(start_dt, end_dt, freq=freq)
+
+    elif periods is not None and (start_dt is not None or end_dt is not None):
+        _, unit_periods, unit, _ = get_frequency(freq)
+        adj_periods = periods * unit_periods
+
+        # TODO: standardize time aliases to avoid any mapping
+        unit = 'd' if unit == 'D' else 'm'
+        delta = pd.Timedelta(adj_periods, unit)
+
+        if start_dt is not None:
+            return pd.date_range(
+                start=start_dt,
+                end=start_dt + delta,
+                freq=freq,
+                closed='left',
+            )
+
+        else:
+            return pd.date_range(
+                start=end_dt - delta,
+                end=end_dt,
+                freq=freq,
+            )
+
+    else:
+        raise ValueError(
+            'Choose only two parameters between start_dt, end_dt '
+            'and periods.'
+        )
+
+
+def get_periods(start_dt, end_dt, freq):
+    """
+    The number of periods in the specified range.
+
+    Parameters
+    ----------
+    start_dt: datetime
+    end_dt: datetime
+    freq: str
+
+    Returns
+    -------
+    int
+
+    """
+    return len(get_periods_range(start_dt=start_dt, end_dt=end_dt, freq=freq))
+
+
+def get_start_dt(end_dt, bar_count, data_frequency, include_first=True):
+    """
+    The start date based on specified end date and data frequency.
+
+    Parameters
+    ----------
+    end_dt: datetime
+    bar_count: int
+    data_frequency: str
+    include_first
+
+    Returns
+    -------
+    datetime
+
+    """
+    periods = bar_count
+    if periods > 1:
+        delta = get_delta(periods, data_frequency)
+        start_dt = end_dt - delta
+
+        if not include_first:
+            start_dt += get_delta(1, data_frequency)
+    else:
+        start_dt = end_dt
+
+    return start_dt
+
+
+def get_period_label(dt, data_frequency):
+    """
+    The period label for the specified date and frequency.
+
+    Parameters
+    ----------
+    dt: datetime
+    data_frequency: str
+
+    Returns
+    -------
+    str
+
+    """
+    if data_frequency == 'minute':
+        return '{}-{:02d}'.format(dt.year, dt.month)
+    else:
+        return '{}'.format(dt.year)
+
+
+def get_month_start_end(dt, first_day=None, last_day=None):
+    """
+    The first and last day of the month for the specified date.
+
+    Parameters
+    ----------
+    dt: datetime
+    first_day: datetime
+    last_day: datetime
+
+    Returns
+    -------
+    datetime, datetime
+
+    """
+    month_range = calendar.monthrange(dt.year, dt.month)
+
+    if first_day:
+        month_start = first_day
+    else:
+        month_start = pd.to_datetime(datetime(
+            dt.year, dt.month, 1, 0, 0, 0, 0
+        ), utc=True)
+
+    if last_day:
+        month_end = last_day
+    else:
+        month_end = pd.to_datetime(datetime(
+            dt.year, dt.month, month_range[1], 23, 59, 0, 0
+        ), utc=True)
+
+        if month_end > pd.Timestamp.utcnow():
+            month_end = pd.Timestamp.utcnow().floor('1D')
+
+    return month_start, month_end
+
+
+def get_year_start_end(dt, first_day=None, last_day=None):
+    """
+    The first and last day of the year for the specified date.
+
+    Parameters
+    ----------
+
+    dt: datetime
+    first_day: datetime
+    last_day: datetime
+
+    Returns
+    -------
+    datetime, datetime
+
+    """
+    year_start = first_day if first_day \
+        else pd.to_datetime(date(dt.year, 1, 1), utc=True)
+    year_end = last_day if last_day \
+        else pd.to_datetime(date(dt.year, 12, 31), utc=True)
+
+    if year_end > pd.Timestamp.utcnow():
+        year_end = pd.Timestamp.utcnow().floor('1D')
+
+    return year_start, year_end
+
+
+def get_frequency(freq, data_frequency=None):
+    """
+    Get the frequency parameters.
+
+    Notes
+    -----
+    We're trying to use Pandas convention for frequency aliases.
+
+    Parameters
+    ----------
+    freq: str
+    data_frequency: str
+
+    Returns
+    -------
+    str, int, str, str
+
+    """
+    if data_frequency is None:
+        data_frequency = 'daily' if freq.upper().endswith('D') else 'minute'
+
+    if freq == 'minute':
+        unit = 'T'
+        candle_size = 1
+
+    elif freq == 'daily':
+        unit = 'D'
+        candle_size = 1
+
+    else:
+        freq_match = re.match(r'([0-9].*)?(m|M|d|D|h|H|T)', freq, re.M | re.I)
+        if freq_match:
+            candle_size = int(freq_match.group(1)) if freq_match.group(1) \
+                else 1
+            unit = freq_match.group(2)
+
+        else:
+            raise InvalidHistoryFrequencyError(frequency=freq)
+
+    # TODO: some exchanges support H and W frequencies but not bundles
+    # Find a way to pass-through these parameters to exchanges
+    # but resample from minute or daily in backtest mode
+    # see catalyst/exchange/ccxt/ccxt_exchange.py:242 for mapping between
+    # Pandas offet aliases (used by Catalyst) and the CCXT timeframes
+    if unit.lower() == 'd':
+        unit = 'D'
+        alias = '{}D'.format(candle_size)
+
+        if data_frequency == 'minute':
+            data_frequency = 'daily'
+
+    elif unit.lower() == 'm' or unit == 'T':
+        unit = 'T'
+        alias = '{}T'.format(candle_size)
+
+        if data_frequency == 'daily':
+            data_frequency = 'minute'
+
+    # elif unit.lower() == 'h':
+    #     candle_size = candle_size * 60
+    #
+    #     alias = '{}T'.format(candle_size)
+    #     if data_frequency == 'daily':
+    #         data_frequency = 'minute'
+
+    else:
+        raise InvalidHistoryFrequencyAlias(freq=freq)
+
+    return alias, candle_size, unit, data_frequency
+
+
+def from_ms_timestamp(ms):
+    return pd.to_datetime(ms, unit='ms', utc=True)
+
+
+def get_epoch():
+    return pd.to_datetime('1970-1-1', utc=True)

catalyst/exchange/utils/exchange_utils.py

@@ -2,21 +2,20 @@ import hashlib
 import json
 import os
 import pickle
-import re
 import shutil
 from datetime import date, datetime
 
 import pandas as pd
 from catalyst.assets._assets import TradingPair
+from six import string_types
+from six.moves.urllib import request
+
 from catalyst.constants import DATE_FORMAT, SYMBOLS_URL
-from catalyst.exchange.exchange_errors import ExchangeSymbolsNotFound, \
-    InvalidHistoryFrequencyError, InvalidHistoryFrequencyAlias
+from catalyst.exchange.exchange_errors import ExchangeSymbolsNotFound
 from catalyst.exchange.utils.serialization_utils import ExchangeJSONEncoder, \
     ExchangeJSONDecoder
 from catalyst.utils.paths import data_root, ensure_directory, \
     last_modified_time
-from six import string_types
-from six.moves.urllib import request
 
 
 def get_sid(symbol):
@@ -513,72 +512,6 @@ def get_common_assets(exchanges):
     return assets
 
 
-def get_frequency(freq, data_frequency):
-    """
-    Get the frequency parameters.
-
-    Notes
-    -----
-    We're trying to use Pandas convention for frequency aliases.
-
-    Parameters
-    ----------
-    freq: str
-    data_frequency: str
-
-    Returns
-    -------
-    str, int, str, str
-
-    """
-    if freq == 'minute':
-        unit = 'T'
-        candle_size = 1
-
-    elif freq == 'daily':
-        unit = 'D'
-        candle_size = 1
-
-    else:
-        freq_match = re.match(r'([0-9].*)?(m|M|d|D|h|H|T)', freq, re.M | re.I)
-        if freq_match:
-            candle_size = int(freq_match.group(1)) if freq_match.group(1) \
-                else 1
-            unit = freq_match.group(2)
-
-        else:
-            raise InvalidHistoryFrequencyError(frequency=freq)
-
-    # TODO: some exchanges support H and W frequencies but not bundles
-    # Find a way to pass-through these parameters to exchanges
-    # but resample from minute or daily in backtest mode
-    # see catalyst/exchange/ccxt/ccxt_exchange.py:242 for mapping between
-    # Pandas offet aliases (used by Catalyst) and the CCXT timeframes
-    if unit.lower() == 'd':
-        alias = '{}D'.format(candle_size)
-
-        if data_frequency == 'minute':
-            data_frequency = 'daily'
-
-    elif unit.lower() == 'm' or unit == 'T':
-        alias = '{}T'.format(candle_size)
-
-        if data_frequency == 'daily':
-            data_frequency = 'minute'
-
-    # elif unit.lower() == 'h':
-    #     candle_size = candle_size * 60
-    #
-    #     alias = '{}T'.format(candle_size)
-    #     if data_frequency == 'daily':
-    #         data_frequency = 'minute'
-
-    else:
-        raise InvalidHistoryFrequencyAlias(freq=freq)
-
-    return alias, candle_size, unit, data_frequency
-
-
 def resample_history_df(df, freq, field):
     """
     Resample the OHCLV DataFrame using the specified frequency.
@@ -652,14 +585,6 @@ def mixin_market_params(exchange_name, params, market):
                 params['lot'] = params['min_trade_size']
 
 
-def from_ms_timestamp(ms):
-    return pd.to_datetime(ms, unit='ms', utc=True)
-
-
-def get_epoch():
-    return pd.to_datetime('1970-1-1', utc=True)
-
-
 def group_assets_by_exchange(assets):
     exchange_assets = dict()
     for asset in assets:

catalyst/exchange/utils/test_utils.py

@@ -62,14 +62,14 @@ def output_df(df, assets, name=None):
 
     """
     if isinstance(assets, TradingPair):
-        exchange_folder = assets.exchange
-        asset_folder = assets.symbol
+        asset_folder = '{}_{}'.format(assets.exchange, assets.symbol)
     else:
-        exchange_folder = ','.join([asset.exchange for asset in assets])
-        asset_folder = ','.join([asset.symbol for asset in assets])
+        asset_folder = ','.join(
+            ['{}_{}'.format(a.exchange, a.symbol) for a in assets]
+        )
 
     folder = os.path.join(
-        tempfile.gettempdir(), 'catalyst', exchange_folder, asset_folder
+        tempfile.gettempdir(), 'catalyst', asset_folder
     )
     ensure_directory(folder)
 
@@ -79,4 +79,4 @@ def output_df(df, assets, name=None):
     path = os.path.join(folder, '{}.csv'.format(name))
     df.to_csv(path)
 
-    return path
+    return path, folder

tests/exchange/test_bundle.py

@@ -10,7 +10,8 @@ from catalyst.exchange.exchange_bcolz import BcolzExchangeBarReader, \
 from catalyst.exchange.exchange_bundle import ExchangeBundle, \
     BUNDLE_NAME_TEMPLATE
 from catalyst.exchange.utils.bundle_utils import get_bcolz_chunk, \
-    get_start_dt, get_df_from_arrays
+    get_df_from_arrays
+from exchange.utils.datetime_utils import get_start_dt
 from catalyst.exchange.utils.exchange_utils import get_exchange_folder
 from catalyst.exchange.utils.factory import get_exchange
 from catalyst.exchange.utils.stats_utils import df_to_string

tests/exchange/test_ccxt.py

@@ -1,6 +1,8 @@
 import pandas as pd
 from logbook import Logger
 
+from catalyst.testing import ZiplineTestCase
+from catalyst.testing.fixtures import WithLogger
 from .base import BaseExchangeTestCase
 from catalyst.exchange.ccxt.ccxt_exchange import CCXT
 from catalyst.exchange.exchange_execution import ExchangeLimitOrder
@@ -59,7 +61,7 @@ class TestCCXT(BaseExchangeTestCase):
             freq='5T',
             assets=[self.exchange.get_asset('eth_btc')],
             bar_count=200,
-            start_dt=pd.to_datetime('2017-01-01', utc=True)
+            start_dt=pd.to_datetime('2017-09-01', utc=True)
         )
 
         for asset in candles:

tests/exchange/test_suites/test_suite_bundle.py

@@ -1,5 +1,6 @@
 import random
 
+import os
 import pandas as pd
 from logbook import TestHandler
 from pandas.util.testing import assert_frame_equal
@@ -11,7 +12,6 @@ from catalyst.exchange.utils.exchange_utils import get_candles_df
 from catalyst.exchange.utils.factory import get_exchange
 from catalyst.exchange.utils.test_utils import output_df, \
     select_random_assets
-from catalyst.testing.fixtures import WithLogger, ZiplineTestCase
 
 pd.set_option('display.expand_frame_repr', False)
 pd.set_option('precision', 8)
@@ -19,7 +19,7 @@ pd.set_option('display.width', 1000)
 pd.set_option('display.max_colwidth', 1000)
 
 
-class TestSuiteBundle(WithLogger, ZiplineTestCase):
+class TestSuiteBundle:
     @staticmethod
     def get_data_portal(exchanges):
         open_calendar = get_calendar('OPEN')
@@ -46,7 +46,9 @@ class TestSuiteBundle(WithLogger, ZiplineTestCase):
         assets
         end_dt
         bar_count
-        sample_minutes
+        freq
+        data_frequency
+        data_portal
 
         Returns
         -------
@@ -64,10 +66,6 @@ class TestSuiteBundle(WithLogger, ZiplineTestCase):
                 field='close',
                 data_frequency=data_frequency,
             )
-            print('bundle data:\n{}'.format(
-                data['bundle'].tail(10))
-            )
-
             candles = exchange.get_candles(
                 end_dt=end_dt,
                 freq=freq,
@@ -81,19 +79,31 @@ class TestSuiteBundle(WithLogger, ZiplineTestCase):
                 bar_count=bar_count,
                 end_dt=end_dt,
             )
-            print('exchange data:\n{}'.format(
-                data['exchange'].tail(10))
-            )
             for source in data:
                 df = data[source]
-                path = output_df(df, assets, '{}_{}'.format(freq, source))
-                print('saved {}:\n{}'.format(source, path))
+                path, folder = output_df(
+                    df, assets, '{}_{}'.format(freq, source)
+                )
 
             assert_frame_equal(
                 right=data['bundle'],
                 left=data['exchange'],
-                check_less_precise=True,
+                check_less_precise=1,
             )
+            try:
+                assert_frame_equal(
+                    right=data['bundle'],
+                    left=data['exchange'],
+                    check_less_precise=min([a.decimals for a in assets]),
+                )
+            except Exception as e:
+                print('Some differences were found within a 1 decimal point '
+                      'interval of confidence: {}'.format(e))
+                with open(os.path.join(folder, 'compare.txt'), 'w+') as handle:
+                    handle.write(e.args[0])
+
+            print('saved test results: {}'.format(folder))
+            pass
 
     def test_validate_bundles(self):
         # exchange_population = 3