def test_error_response(mock):
mock.return_value = TestResponse(status_code=500, data={'errors': {'detail': 'Internal server error'}})
with assert_raises(SanError):
san.get(
"prices/santiment_usd",
from_date="2018-06-01",
to_date="2018-06-05",
interval="1d"
)
def crypto_p(crypto, start, end):
df = san.get('projects/all')
ii = df.index[df['name'] == crypto]
li = str(df['slug'][ii]).split()
df1 = san.get("ohlcv/" + li[1],
from_date=start,
to_date=end,
interval="1d")
df1 = df1.assign(returns=df1['closePriceUsd'] -
df1['openPriceUsd']) #Returns Calculation per day
return (df1)
def test_top_transfers():
result = san.get('top_transfers/santiment',
from_date=params['from_date'],
to_date=params['to_date'],
interval='1d')
assert len(result) != 0
def get_san_metric(start,
end,
metric,
asset,
interval,
iterate_over_days=120,
convert_index=True):
''' Iterates over days and gets the metric data. '''
start = datetime.datetime.strptime(start, '%Y-%m-%d').date()
end = datetime.datetime.strptime(end, '%Y-%m-%d').date()
df = pd.DataFrame(None)
for i in range(int(np.ceil((end - start).days / iterate_over_days))):
start_date = str(start +
datetime.timedelta(days=iterate_over_days * i))
end_date = str(
min(
end, start + datetime.timedelta(days=iterate_over_days *
(i + 1) - 1)))
df_batch = san.get(f'{metric}/{asset}',
from_date=start_date,
to_date=end_date,
interval=interval)
df = df.append(df_batch)
if convert_index:
df.index = df.index.astype('datetime64[ns]')
return df
def test_get_without_transform(mock):
api_call_result = {
'query_0': [{
'balance': 212664.33000000002,
'datetime': '2019-05-23T00:00:00Z'
}, {
'balance': 212664.33000000002,
'datetime': '2019-05-24T00:00:00Z'
}, {
'balance': 212664.33000000002,
'datetime': '2019-05-25T00:00:00Z'
}]
}
# The value is passed by refference, that's why deepcopy is used for
# expected
mock.return_value = TestResponse(status_code=200,
data=deepcopy(api_call_result))
res = san.get("historical_balance/santiment",
address="0x1f3df0b8390bb8e9e322972c5e75583e87608ec2",
from_date="2019-05-23",
to_date="2019-05-26",
interval="1d")
expected_df = convert_to_datetime_idx_df(api_call_result['query_0'])
pdt.assert_frame_equal(res, expected_df, check_dtype=False)
def add_project(self, project):
self.asset_list.append(project)
self.portfolio[project] = san.get(
"ohlcv/" + project,
from_date=self.start_date,
to_date=self.end_date).closePriceUsd.pct_change()
self.portfolio = self.portfolio.replace([np.inf, -np.inf], 0)
def test_prices(mock):
api_call_result = {
'query_0': [{
'priceUsd': '1.234634930555555',
'priceBtc': '0.0001649780416666666',
'datetime': '2018-06-01T00:00:00Z'
}, {
'priceUsd': '1.2551352777777771',
'priceBtc': '0.00016521851041666669',
'datetime': '2018-06-02T00:00:00Z'
}, {
'priceUsd': '1.251881943462897',
'priceBtc': '0.000162902558303887',
'datetime': '2018-06-03T00:00:00Z'
}, {
'priceUsd': '1.2135782638888888',
'priceBtc': '0.0001600935277777778',
'datetime': '2018-06-04T00:00:00Z'
}]
}
mock.return_value = TestResponse(status_code=200,
data=deepcopy(api_call_result))
res = san.get("prices/santiment_usd",
from_date="2018-06-01",
to_date="2018-06-05",
interval="1d")
df = convert_to_datetime_idx_df(api_call_result['query_0'])
pdt.assert_frame_equal(res, df, check_dtype=False)
def test_news():
result = san.get('news/' + params['project_slug'],
from_date='2019-04-18',
to_date='2019-07-11',
size=5)
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def test_eth_spent_over_time():
result = san.get('eth_spent_over_time/' + params['project_slug'],
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'])
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def test_token_top_transactions():
result = san.get('token_top_transactions/' + params['project_slug'],
from_date='2019-06-18',
to_date='2019-07-11',
limit=5)
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def _test_ordinary_function(query, slug=params['project_slug']):
result = san.get(query + '/' + slug,
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'])
assert result.empty == False
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
def test_historical_balance():
result = san.get('historical_balance/' + params['project_slug'],
address=params['address'],
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'])
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def test_top_holders_percent_of_total_supply():
result = san.get('top_holders_percent_of_total_supply/' +
params['project_slug'],
number_of_holders=10,
from_date=params['from_date'],
to_date=params['to_date'])
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def metrics(self, metric):
metric_data = pd.DataFrame()
for asset in self.asset_list:
metric_data[asset] = san.get(metric + "/" + asset,
from_date=self.start_date,
to_date=self.end_date).iloc[:, 0]
self.metrics[metric] = metric_data
return metric_data
def loadSantimentMetric(metric, asset, start, end, interval, args):
data = san.get(
(metric + '/' + asset.slug)
, from_date = start.isoformat()
, to_date = end.isoformat()
, interval = intervalISOFormat(interval)
)
cacheSantimentMetric(metric, asset, start, interval, data[metric_dic[metric]])
def test_ohlcv():
ohlcv_df = san.get("{}/{}".format('ohlcv', params['project_slug']),
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'])
assert len(ohlcv_df.index) >= 1
assert 'DatetimeIndex' in str(type(ohlcv_df.index))
assert ohlcv_df.empty == False
def test_emerging_trends():
get_metric_df = san.get('emerging_trends',
from_date=params['from_date'],
to_date=params['to_date'],
interval='1d',
size=5)
assert len(get_metric_df) >= 1
assert 'DatetimeIndex' in str(type(get_metric_df.index))
assert get_metric_df.empty == False
def test_get_metric():
get_metric_df = san.get('daily_active_addresses/' + params['project_slug'],
from_date=params['from_date'],
to_date=params['to_date'],
interval='2d',
aggregation='AVG')
assert len(get_metric_df) >= 1
assert 'DatetimeIndex' in str(type(get_metric_df.index))
assert get_metric_df.empty == False
def loadPrice(metric, asset, start, end, interval, args):
data = san.get(
('prices/' + asset.slug)
, from_date = start.isoformat()
, to_date = end.isoformat()
, interval = intervalISOFormat(interval)
)
# Cache all the data
fillMissingData(start, end, interval, data)
cacheSantimentMetric(metric, asset, start, interval, data[UNIT])
def test_top_social_gainers_losers():
tsgl_df = san.get('top_social_gainers_losers',
from_date=params['from_date'],
to_date=params['to_date'],
size=5,
time_window='2d',
status='ALL')
assert len(tsgl_df) >= 1
assert 'DatetimeIndex' in str(type(tsgl_df.index))
assert tsgl_df.empty == False
def test_eth_top_transactions():
transaction_types = ['ALL', 'IN', 'OUT']
for item in transaction_types:
result = san.get('eth_top_transactions/' + params['project_slug'],
from_date='2019-06-11',
to_date='2019-07-11',
limit=5,
transaction_type=item)
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def test_price_volume_difference():
currencies = ['USD', 'BTC']
for item in currencies:
result = san.get('price_volume_difference/' + params['project_slug'],
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'],
currency=item)
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def test_social_dominance():
sources = ['ALL', 'REDDIT', 'TELEGRAM']
for item in sources:
result = san.get('social_dominance/' + params['project_slug'],
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'],
source=item)
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def loadSocialVolume(metric, asset, start, end, interval, args):
assert len(args) > 0
data = san.get(
('social_volume/' + asset.slug)
, from_date = start.isoformat()
, to_date = end.isoformat()
, interval = intervalISOFormat(interval)
, social_volume_type = SOCIAL_VOLUME_TYPES[args[0]]
)
cacheSantimentMetric(metric , asset, start, interval, data['mentionsCount'])
def __init__(self,
start_date="2017-01-01",
end_date=datetime.datetime.now().strftime("%Y-%m-%d"),
asset_list=[]):
""" Takes in list of project slugs"""
self.start_date = start_date
self.end_date = end_date
self.asset_list = asset_list
self.portfolio = pd.DataFrame()
self.benchmark = san.get("ohlcv/bitcoin",
from_date=start_date,
to_date=end_date).closePriceUsd.pct_change()
for portfolio_asset in asset_list:
self.portfolio[portfolio_asset] = san.get(
"ohlcv/" + portfolio_asset,
from_date=start_date,
to_date=end_date).closePriceUsd.pct_change()
self.portfolio = self.portfolio.replace([np.inf, -np.inf], 0)
self.metrics = dict()
def get_metric(self, metric, interval="1d"):
"""
Retrieve a Santiment Metric
- returns a DataFrame with 1 column
"""
data = san.get(metric + '/' + self.project,
from_date=self.start_date,
to_date=self.end_date,
interval=interval)
data = data.rename(columns={data.columns[0]: metric})
return data
def filter_projects_by_marketcap(number):
projects = san.get('projects/all')
slugs = projects['slug'].values
caps = []
for i in range(len(slugs) // BATCH_SIZE):
slugs_sub = slugs[BATCH_SIZE * i:BATCH_SIZE * (i + 1)]
caps += get_marketcap_batch(slugs_sub)
logging.info("Batch %s executed", i)
results = zip(slugs, caps)
return [
x[0]
for x in sorted(results, key=lambda k: k[1], reverse=True)[:number]
]
def test_social_volume():
social_volume_types = [
'TELEGRAM_CHATS_OVERVIEW', 'TELEGRAM_DISCUSSION_OVERVIEW'
]
for item in social_volume_types:
result = san.get('social_volume/' + params['project_slug'],
from_date=params['from_date'],
to_date=params['to_date'],
interval=params['interval'],
social_volume_type=item)
assert len(result.index) >= 1
assert 'DatetimeIndex' in str(type(result.index))
assert result.empty == False
def loadSocialChartData(metric, asset, start, end, interval, args):
assert len(args) > 1
data = san.get(
('topic_search/chart_data')
, source = SOCIAL_SOURCE_TYPES[args[0]]
, search_text = args[1]
, from_date = start.isoformat()
, to_date = end.isoformat()
, interval = intervalISOFormat(interval)
)
cacheSantimentMetric(metric, asset, start, interval, data['chartData'])
def getSocialMessages(metric, asset, time, args):
assert len(args) > 1
start = getSanStartTime(time)
end = getSanEndTime(time)
interval = getSanInterval()
data = san.get(
('topic_search/messages')
, source = SOCIAL_SOURCE_TYPES[args[0]]
, search_text = args[1]
, from_date = start.isoformat()
, to_date = end.isoformat()
, interval = intervalISOFormat(interval)
)
return data['messages'].values
