def test_lookup_symbol_fuzzy(self):
as_of = pd.Timestamp('2013-01-01', tz='UTC')
frame = pd.DataFrame.from_records([{
'sid': i,
'file_name': 'TEST@%d' % i,
'company_name': "company%d" % i,
'start_date_nano': as_of.value,
'end_date_nano': as_of.value,
'exchange': uuid.uuid4().hex,
} for i in range(3)])
finder = AssetFinder(frame, fuzzy_char='@')
asset_0, asset_1, asset_2 = (finder.retrieve_asset(i)
for i in range(3))
for i in range(2): # we do it twice to test for caching bugs
self.assertIsNone(finder.lookup_symbol('test', as_of))
self.assertEqual(asset_1, finder.lookup_symbol('test@1', as_of))
# Adding an unnecessary fuzzy shouldn't matter.
self.assertEqual(asset_1,
finder.lookup_symbol('test@1', as_of, fuzzy=True))
# Shouldn't find this with no fuzzy_str passed.
self.assertIsNone(finder.lookup_symbol('test1', as_of))
# Should find exact match.
self.assertEqual(
asset_1,
finder.lookup_symbol('test1', as_of, fuzzy=True),
)
def test_lookup_symbol_delimited(self):
as_of = pd.Timestamp('2013-01-01', tz='UTC')
frame = pd.DataFrame.from_records([{
'sid': i,
'symbol': 'TEST.%d' % i,
'company_name': "company%d" % i,
'start_date': as_of.value,
'end_date': as_of.value,
'exchange': uuid.uuid4().hex
} for i in range(3)])
self.env.write_data(equities_df=frame)
finder = AssetFinder(self.env.engine)
asset_0, asset_1, asset_2 = (finder.retrieve_asset(i)
for i in range(3))
# we do it twice to catch caching bugs
for i in range(2):
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol('test', as_of)
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol('test1', as_of)
# '@' is not a supported delimiter
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol('test@1', as_of)
# Adding an unnecessary fuzzy shouldn't matter.
for fuzzy_char in ['-', '/', '_', '.']:
self.assertEqual(
asset_1, finder.lookup_symbol('test%s1' % fuzzy_char,
as_of))
def test_lookup_symbol_delimited(self):
as_of = pd.Timestamp("2013-01-01", tz="UTC")
frame = pd.DataFrame.from_records(
[
{
"sid": i,
"symbol": "TEST.%d" % i,
"company_name": "company%d" % i,
"start_date": as_of.value,
"end_date": as_of.value,
"exchange": uuid.uuid4().hex,
}
for i in range(3)
]
)
self.env.write_data(equities_df=frame)
finder = AssetFinder(self.env.engine)
asset_0, asset_1, asset_2 = (finder.retrieve_asset(i) for i in range(3))
# we do it twice to catch caching bugs
for i in range(2):
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol("TEST", as_of)
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol("TEST1", as_of)
# '@' is not a supported delimiter
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol("TEST@1", as_of)
# Adding an unnecessary fuzzy shouldn't matter.
for fuzzy_char in ["-", "/", "_", "."]:
self.assertEqual(asset_1, finder.lookup_symbol("TEST%s1" % fuzzy_char, as_of))
def test_consume_asset_as_identifier(self):
# Build some end dates
eq_end = pd.Timestamp("2012-01-01", tz="UTC")
fut_end = pd.Timestamp("2008-01-01", tz="UTC")
# Build some simple Assets
equity_asset = Equity(1, symbol="TESTEQ", end_date=eq_end)
future_asset = Future(200, symbol="TESTFUT", end_date=fut_end)
# Consume the Assets
self.env.write_data(equities_identifiers=[equity_asset], futures_identifiers=[future_asset])
finder = AssetFinder(self.env.engine)
# Test equality with newly built Assets
self.assertEqual(equity_asset, finder.retrieve_asset(1))
self.assertEqual(future_asset, finder.retrieve_asset(200))
self.assertEqual(eq_end, finder.retrieve_asset(1).end_date)
self.assertEqual(fut_end, finder.retrieve_asset(200).end_date)
def test_consume_asset_as_identifier(self):
# Build some end dates
eq_end = pd.Timestamp('2012-01-01', tz='UTC')
fut_end = pd.Timestamp('2008-01-01', tz='UTC')
# Build some simple Assets
equity_asset = Equity(1, symbol="TESTEQ", end_date=eq_end)
future_asset = Future(200, symbol="TESTFUT", end_date=fut_end)
# Consume the Assets
finder = AssetFinder()
finder.consume_identifiers([equity_asset, future_asset])
# Test equality with newly built Assets
self.assertEqual(equity_asset, finder.retrieve_asset(1))
self.assertEqual(future_asset, finder.retrieve_asset(200))
self.assertEqual(eq_end, finder.retrieve_asset(1).end_date)
self.assertEqual(fut_end, finder.retrieve_asset(200).end_date)
def test_consume_asset_as_identifier(self):
# Build some end dates
eq_end = pd.Timestamp('2012-01-01', tz='UTC')
fut_end = pd.Timestamp('2008-01-01', tz='UTC')
# Build some simple Assets
equity_asset = Equity(1, symbol="TESTEQ", end_date=eq_end)
future_asset = Future(200, symbol="TESTFUT", end_date=fut_end)
# Consume the Assets
finder = AssetFinder()
finder.consume_identifiers([equity_asset, future_asset])
# Test equality with newly built Assets
self.assertEqual(equity_asset, finder.retrieve_asset(1))
self.assertEqual(future_asset, finder.retrieve_asset(200))
self.assertEqual(eq_end, finder.retrieve_asset(1).end_date)
self.assertEqual(fut_end, finder.retrieve_asset(200).end_date)
def test_consume_metadata(self):
# Test dict consumption
dict_to_consume = {0: {'symbol': 'PLAY'}, 1: {'symbol': 'MSFT'}}
self.env.write_data(equities_data=dict_to_consume)
finder = AssetFinder(self.env.engine)
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
# Test dataframe consumption
df = pd.DataFrame(columns=['asset_name', 'exchange'], index=[0, 1])
df['asset_name'][0] = "Dave'N'Busters"
df['exchange'][0] = "NASDAQ"
df['asset_name'][1] = "Microsoft"
df['exchange'][1] = "NYSE"
self.env = TradingEnvironment()
self.env.write_data(equities_df=df)
finder = AssetFinder(self.env.engine)
self.assertEqual('NASDAQ', finder.retrieve_asset(0).exchange)
self.assertEqual('Microsoft', finder.retrieve_asset(1).asset_name)
def test_consume_metadata(self):
# Test dict consumption
dict_to_consume = {0: {"symbol": "PLAY"}, 1: {"symbol": "MSFT"}}
self.env.write_data(equities_data=dict_to_consume)
finder = AssetFinder(self.env.engine)
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual("PLAY", equity.symbol)
# Test dataframe consumption
df = pd.DataFrame(columns=["asset_name", "exchange"], index=[0, 1])
df["asset_name"][0] = "Dave'N'Busters"
df["exchange"][0] = "NASDAQ"
df["asset_name"][1] = "Microsoft"
df["exchange"][1] = "NYSE"
self.env = TradingEnvironment(load=noop_load)
self.env.write_data(equities_df=df)
finder = AssetFinder(self.env.engine)
self.assertEqual("NASDAQ", finder.retrieve_asset(0).exchange)
self.assertEqual("Microsoft", finder.retrieve_asset(1).asset_name)
def test_insert_metadata(self):
data = {0: {"start_date": "2014-01-01", "end_date": "2015-01-01", "symbol": "PLAY", "foo_data": "FOO"}}
self.env.write_data(equities_data=data)
finder = AssetFinder(self.env.engine)
# Test proper insertion
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual("PLAY", equity.symbol)
self.assertEqual(pd.Timestamp("2015-01-01", tz="UTC"), equity.end_date)
# Test invalid field
with self.assertRaises(AttributeError):
equity.foo_data
def test_consume_metadata(self):
# Test dict consumption
dict_to_consume = {0: {'symbol': 'PLAY'},
1: {'symbol': 'MSFT'}}
self.env.write_data(equities_data=dict_to_consume)
finder = AssetFinder(self.env.engine)
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
# Test dataframe consumption
df = pd.DataFrame(columns=['asset_name', 'exchange'], index=[0, 1])
df['asset_name'][0] = "Dave'N'Busters"
df['exchange'][0] = "NASDAQ"
df['asset_name'][1] = "Microsoft"
df['exchange'][1] = "NYSE"
self.env = TradingEnvironment(load=noop_load)
self.env.write_data(equities_df=df)
finder = AssetFinder(self.env.engine)
self.assertEqual('NASDAQ', finder.retrieve_asset(0).exchange)
self.assertEqual('Microsoft', finder.retrieve_asset(1).asset_name)
def test_insert_metadata(self):
data = {0: {'start_date': '2014-01-01',
'end_date': '2015-01-01',
'symbol': "PLAY",
'foo_data': "FOO"}}
self.env.write_data(equities_data=data)
finder = AssetFinder(self.env.engine)
# Test proper insertion
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
self.assertEqual(pd.Timestamp('2015-01-01', tz='UTC'),
equity.end_date)
# Test invalid field
with self.assertRaises(AttributeError):
equity.foo_data
def test_insert_metadata(self):
finder = AssetFinder()
finder.insert_metadata(0,
asset_type='equity',
start_date='2014-01-01',
end_date='2015-01-01',
symbol="PLAY",
foo_data="FOO",)
# Test proper insertion
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
self.assertEqual(pd.Timestamp('2015-01-01', tz='UTC'),
equity.end_date)
# Test invalid field
self.assertFalse('foo_data' in finder.metadata_cache[0])
def test_insert_metadata(self):
finder = AssetFinder()
finder.insert_metadata(
0,
asset_type='equity',
start_date='2014-01-01',
end_date='2015-01-01',
symbol="PLAY",
foo_data="FOO",
)
# Test proper insertion
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
self.assertEqual(pd.Timestamp('2015-01-01', tz='UTC'), equity.end_date)
# Test invalid field
self.assertFalse('foo_data' in finder.metadata_cache[0])
def test_lookup_symbol_fuzzy(self):
as_of = pd.Timestamp('2013-01-01', tz='UTC')
frame = pd.DataFrame.from_records(
[
{
'sid': i,
'file_name': 'TEST@%d' % i,
'company_name': "company%d" % i,
'start_date_nano': as_of.value,
'end_date_nano': as_of.value,
'exchange': uuid.uuid4().hex,
}
for i in range(3)
]
)
finder = AssetFinder(frame)
asset_0, asset_1, asset_2 = (
finder.retrieve_asset(i) for i in range(3)
)
for i in range(2): # we do it twice to test for caching bugs
self.assertIsNone(finder.lookup_symbol('test', as_of))
self.assertEqual(
asset_1,
finder.lookup_symbol('test@1', as_of)
)
# Adding an unnecessary fuzzy shouldn't matter.
self.assertEqual(
asset_1,
finder.lookup_symbol('test@1', as_of, fuzzy='@')
)
# Shouldn't find this with no fuzzy_str passed.
self.assertIsNone(finder.lookup_symbol('test1', as_of))
# Shouldn't find this with an incorrect fuzzy_str.
self.assertIsNone(finder.lookup_symbol('test1', as_of, fuzzy='*'))
# Should find it with the correct fuzzy_str.
self.assertEqual(
asset_1,
finder.lookup_symbol('test1', as_of, fuzzy='@'),
)
def test_lookup_symbol_fuzzy(self):
as_of = pd.Timestamp('2013-01-01', tz='UTC')
frame = pd.DataFrame.from_records(
[
{
'sid': i,
'symbol': 'TEST@%d' % i,
'company_name': "company%d" % i,
'start_date': as_of.value,
'end_date': as_of.value,
'exchange': uuid.uuid4().hex,
'fuzzy': 'TEST%d' % i
}
for i in range(3)
]
)
self.env.write_data(equities_df=frame)
finder = AssetFinder(self.env.engine, fuzzy_char='@')
asset_0, asset_1, asset_2 = (
finder.retrieve_asset(i) for i in range(3)
)
for i in range(2): # we do it twice to test for caching bugs
self.assertIsNone(finder.lookup_symbol('test', as_of))
self.assertEqual(
asset_1,
finder.lookup_symbol('test@1', as_of)
)
# Adding an unnecessary fuzzy shouldn't matter.
self.assertEqual(
asset_1,
finder.lookup_symbol('test@1', as_of, fuzzy=True)
)
# Shouldn't find this with no fuzzy_str passed.
self.assertIsNone(finder.lookup_symbol('test1', as_of))
# Should find exact match.
self.assertEqual(
asset_1,
finder.lookup_symbol('test1', as_of, fuzzy=True),
)
def test_insert_metadata(self):
data = {
0: {
'start_date': '2014-01-01',
'end_date': '2015-01-01',
'symbol': "PLAY",
'foo_data': "FOO"
}
}
self.env.write_data(equities_data=data)
finder = AssetFinder(self.env.engine)
# Test proper insertion
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
self.assertEqual(pd.Timestamp('2015-01-01', tz='UTC'), equity.end_date)
# Test invalid field
with self.assertRaises(AttributeError):
equity.foo_data
def test_consume_metadata(self):
# Test dict consumption
finder = AssetFinder()
dict_to_consume = {0: {'symbol': 'PLAY'}, 1: {'symbol': 'MSFT'}}
finder.consume_metadata(dict_to_consume)
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
finder = AssetFinder()
# Test dataframe consumption
df = pd.DataFrame(columns=['asset_name', 'exchange'], index=[0, 1])
df['asset_name'][0] = "Dave'N'Busters"
df['exchange'][0] = "NASDAQ"
df['asset_name'][1] = "Microsoft"
df['exchange'][1] = "NYSE"
finder.consume_metadata(df)
self.assertEqual('NASDAQ', finder.metadata_cache[0]['exchange'])
self.assertEqual('Microsoft', finder.metadata_cache[1]['asset_name'])
def test_consume_metadata(self):
# Test dict consumption
finder = AssetFinder()
dict_to_consume = {0: {'symbol': 'PLAY'},
1: {'symbol': 'MSFT'}}
finder.consume_metadata(dict_to_consume)
equity = finder.retrieve_asset(0)
self.assertIsInstance(equity, Equity)
self.assertEqual('PLAY', equity.symbol)
finder = AssetFinder()
# Test dataframe consumption
df = pd.DataFrame(columns=['asset_name', 'exchange'], index=[0, 1])
df['asset_name'][0] = "Dave'N'Busters"
df['exchange'][0] = "NASDAQ"
df['asset_name'][1] = "Microsoft"
df['exchange'][1] = "NYSE"
finder.consume_metadata(df)
self.assertEqual('NASDAQ', finder.metadata_cache[0]['exchange'])
self.assertEqual('Microsoft', finder.metadata_cache[1]['asset_name'])
def test_lookup_symbol_delimited(self):
as_of = pd.Timestamp('2013-01-01', tz='UTC')
frame = pd.DataFrame.from_records(
[
{
'sid': i,
'symbol': 'TEST.%d' % i,
'company_name': "company%d" % i,
'start_date': as_of.value,
'end_date': as_of.value,
'exchange': uuid.uuid4().hex
}
for i in range(3)
]
)
self.env.write_data(equities_df=frame)
finder = AssetFinder(self.env.engine)
asset_0, asset_1, asset_2 = (
finder.retrieve_asset(i) for i in range(3)
)
# we do it twice to catch caching bugs
for i in range(2):
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol('TEST', as_of)
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol('TEST1', as_of)
# '@' is not a supported delimiter
with self.assertRaises(SymbolNotFound):
finder.lookup_symbol('TEST@1', as_of)
# Adding an unnecessary fuzzy shouldn't matter.
for fuzzy_char in ['-', '/', '_', '.']:
self.assertEqual(
asset_1,
finder.lookup_symbol('TEST%s1' % fuzzy_char, as_of)
)
def build_lookup_generic_cases():
"""
Generate test cases for AssetFinder test_lookup_generic.
"""
unique_start = pd.Timestamp('2013-01-01', tz='UTC')
unique_end = pd.Timestamp('2014-01-01', tz='UTC')
dupe_0_start = pd.Timestamp('2013-01-01', tz='UTC')
dupe_0_end = dupe_0_start + timedelta(days=1)
dupe_1_start = pd.Timestamp('2013-01-03', tz='UTC')
dupe_1_end = dupe_1_start + timedelta(days=1)
frame = pd.DataFrame.from_records([
{
'sid': 0,
'file_name': 'duplicated',
'company_name': 'duplicated_0',
'start_date_nano': dupe_0_start.value,
'end_date_nano': dupe_0_end.value,
'exchange': '',
},
{
'sid': 1,
'file_name': 'duplicated',
'company_name': 'duplicated_1',
'start_date_nano': dupe_1_start.value,
'end_date_nano': dupe_1_end.value,
'exchange': '',
},
{
'sid': 2,
'file_name': 'unique',
'company_name': 'unique',
'start_date_nano': unique_start.value,
'end_date_nano': unique_end.value,
'exchange': '',
},
], )
finder = AssetFinder(metadata=frame)
dupe_0, dupe_1, unique = assets = [
finder.retrieve_asset(i) for i in range(3)
]
dupe_0_start = dupe_0.start_date
dupe_1_start = dupe_1.start_date
cases = [
##
# Scalars
# Asset object
(finder, assets[0], None, assets[0]),
(finder, assets[1], None, assets[1]),
(finder, assets[2], None, assets[2]),
# int
(finder, 0, None, assets[0]),
(finder, 1, None, assets[1]),
(finder, 2, None, assets[2]),
# Duplicated symbol with resolution date
(finder, 'duplicated', dupe_0_start, dupe_0),
(finder, 'duplicated', dupe_1_start, dupe_1),
# Unique symbol, with or without resolution date.
(finder, 'unique', unique_start, unique),
(finder, 'unique', None, unique),
##
# Iterables
# Iterables of Asset objects.
(finder, assets, None, assets),
(finder, iter(assets), None, assets),
# Iterables of ints
(finder, (0, 1), None, assets[:-1]),
(finder, iter((0, 1)), None, assets[:-1]),
# Iterables of symbols.
(finder, ('duplicated', 'unique'), dupe_0_start, [dupe_0, unique]),
(finder, ('duplicated', 'unique'), dupe_1_start, [dupe_1, unique]),
# Mixed types
(finder, ('duplicated', 2, 'unique', 1, dupe_1), dupe_0_start,
[dupe_0, assets[2], unique, assets[1], dupe_1]),
]
return cases
def build_lookup_generic_cases():
"""
Generate test cases for AssetFinder test_lookup_generic.
"""
unique_start = pd.Timestamp('2013-01-01', tz='UTC')
unique_end = pd.Timestamp('2014-01-01', tz='UTC')
dupe_0_start = pd.Timestamp('2013-01-01', tz='UTC')
dupe_0_end = dupe_0_start + timedelta(days=1)
dupe_1_start = pd.Timestamp('2013-01-03', tz='UTC')
dupe_1_end = dupe_1_start + timedelta(days=1)
frame = pd.DataFrame.from_records(
[
{
'sid': 0,
'file_name': 'duplicated',
'company_name': 'duplicated_0',
'start_date_nano': dupe_0_start.value,
'end_date_nano': dupe_0_end.value,
'exchange': '',
},
{
'sid': 1,
'file_name': 'duplicated',
'company_name': 'duplicated_1',
'start_date_nano': dupe_1_start.value,
'end_date_nano': dupe_1_end.value,
'exchange': '',
},
{
'sid': 2,
'file_name': 'unique',
'company_name': 'unique',
'start_date_nano': unique_start.value,
'end_date_nano': unique_end.value,
'exchange': '',
},
],
)
finder = AssetFinder(metadata=frame)
dupe_0, dupe_1, unique = assets = [
finder.retrieve_asset(i)
for i in range(3)
]
dupe_0_start = dupe_0.start_date
dupe_1_start = dupe_1.start_date
cases = [
##
# Scalars
# Asset object
(finder, assets[0], None, assets[0]),
(finder, assets[1], None, assets[1]),
(finder, assets[2], None, assets[2]),
# int
(finder, 0, None, assets[0]),
(finder, 1, None, assets[1]),
(finder, 2, None, assets[2]),
# Duplicated symbol with resolution date
(finder, 'duplicated', dupe_0_start, dupe_0),
(finder, 'duplicated', dupe_1_start, dupe_1),
# Unique symbol, with or without resolution date.
(finder, 'unique', unique_start, unique),
(finder, 'unique', None, unique),
##
# Iterables
# Iterables of Asset objects.
(finder, assets, None, assets),
(finder, iter(assets), None, assets),
# Iterables of ints
(finder, (0, 1), None, assets[:-1]),
(finder, iter((0, 1)), None, assets[:-1]),
# Iterables of symbols.
(finder, ('duplicated', 'unique'), dupe_0_start, [dupe_0, unique]),
(finder, ('duplicated', 'unique'), dupe_1_start, [dupe_1, unique]),
# Mixed types
(finder,
('duplicated', 2, 'unique', 1, dupe_1),
dupe_0_start,
[dupe_0, assets[2], unique, assets[1], dupe_1]),
]
return cases
