def export_tsv(self, dest_path):
"""
Export the current query result to a tsv file.
:param str dest_path: file to write query results to
:raise NotImplementedError:
Example
.. code-block:: python
:linenos:
>>> top = client.topn(
datasource='twitterstream',
granularity='all',
intervals='2013-10-04/pt1h',
aggregations={"count": doublesum("count")},
dimension='user_name',
filter = Dimension('user_lang') == 'en',
metric='count',
threshold=2
)
>>> top.export_tsv('top.tsv')
>>> !cat top.tsv
>>> count user_name timestamp
7.0 user_1 2013-10-04T00:00:00.000Z
6.0 user_2 2013-10-04T00:00:00.000Z
"""
if six.PY3:
f = open(dest_path, "w", newline="", encoding="utf-8")
else:
f = open(dest_path, "wb")
w = UnicodeWriter(f)
if self.query_type == "timeseries":
header = list(self.result[0]["result"].keys())
header.append("timestamp")
elif self.query_type == "topN":
header = list(self.result[0]["result"][0].keys())
header.append("timestamp")
elif self.query_type == "groupBy":
header = list(self.result[0]["event"].keys())
header.append("timestamp")
header.append("version")
else:
raise NotImplementedError(
"TSV export not implemented for query type: {0}".format(
self.query_type))
w.writerow(header)
if self.result:
if self.query_type == "topN" or self.query_type == "timeseries":
for item in self.result:
timestamp = item["timestamp"]
result = item["result"]
if type(result) is list: # topN
for line in result:
w.writerow(list(line.values()) + [timestamp])
else: # timeseries
w.writerow(list(result.values()) + [timestamp])
elif self.query_type == "groupBy":
for item in self.result:
timestamp = item["timestamp"]
version = item["version"]
w.writerow(
list(item["event"].values()) + [timestamp] + [version])
f.close()
def export_tsv(self, dest_path):
"""
Export the current query result to a tsv file.
:param str dest_path: file to write query results to
:raise NotImplementedError:
Example
.. code-block:: python
:linenos:
>>> top = client.topn(
datasource='twitterstream',
granularity='all',
intervals='2013-10-04/pt1h',
aggregations={"count": doublesum("count")},
dimension='user_name',
filter = Dimension('user_lang') == 'en',
metric='count',
threshold=2
)
>>> top.export_tsv('top.tsv')
>>> !cat top.tsv
>>> count user_name timestamp
7.0 user_1 2013-10-04T00:00:00.000Z
6.0 user_2 2013-10-04T00:00:00.000Z
"""
if six.PY3:
f = open(dest_path, 'w', newline='', encoding='utf-8')
else:
f = open(dest_path, 'wb')
w = UnicodeWriter(f)
if self.query_type == "timeseries":
header = list(self.result[0]['result'].keys())
header.append('timestamp')
elif self.query_type == 'topN':
header = list(self.result[0]['result'][0].keys())
header.append('timestamp')
elif self.query_type == "groupBy":
header = list(self.result[0]['event'].keys())
header.append('timestamp')
header.append('version')
else:
raise NotImplementedError('TSV export not implemented for query type: {0}'.format(self.query_type))
w.writerow(header)
if self.result:
if self.query_type == "topN" or self.query_type == "timeseries":
for item in self.result:
timestamp = item['timestamp']
result = item['result']
if type(result) is list: # topN
for line in result:
w.writerow(list(line.values()) + [timestamp])
else: # timeseries
w.writerow(list(result.values()) + [timestamp])
elif self.query_type == "groupBy":
for item in self.result:
timestamp = item['timestamp']
version = item['version']
w.writerow(
list(item['event'].values()) + [timestamp] + [version])
f.close()
