def post(self):
# get args
self.start_cursor = self.request.get('cursor')
self.filtering_event_key = self.request.get('event')
self.filename = self.request.get('filename')
self.csv_header = self.request.get('csv_header')
self.worker_url = self.request.get('worker_url')
self.event = Event.get(self.filtering_event_key) if self.filtering_event_key else None
# get (base) query, skip query to cursor, filter for sites
query = self.get_base_query()
if self.start_cursor:
query.with_cursor(self.start_cursor)
fetched_sites = query.fetch(limit=self.sites_per_task)
sites = self.filter_sites(fetched_sites)
# write part of csv file to GCS
csv_part_gcs_fd = cloudstorage.open(
BUCKET_NAME + '/' + self.filename + '.part.' + self.start_cursor,
'w',
content_type='text/csv'
)
self._write_csv_rows(csv_part_gcs_fd, sites)
csv_part_gcs_fd.close()
# decide what to do next
self.end_cursor = query.cursor()
if self.end_cursor and self.start_cursor != self.end_cursor:
# chain to next task
taskqueue.add(
url=self.worker_url,
params=self.get_continuation_param_dict(),
retry_options=taskqueue.TaskRetryOptions(task_retry_limit=3),
)
else:
# finish file: combine parts and deduplicate lines
logging.info(u"Deduplicating to create %s ..." % self.filename)
sio = StringIO()
path_prefix = BUCKET_NAME + '/' + self.filename + '.part'
for gcs_file_stat in cloudstorage.listbucket(path_prefix):
csv_part_gcs_fd = cloudstorage.open(gcs_file_stat.filename)
for line in csv_part_gcs_fd:
sio.write(line)
csv_part_gcs_fd.close()
sio.seek(0)
deduplicated_lines = set(line for line in sio)
# write csv header and deduplicated lines to new file
csv_complete_gcs_fd = cloudstorage.open(
BUCKET_NAME + '/' + self.filename,
'w',
content_type='text/csv'
)
csv_complete_gcs_fd.write(self.csv_header.encode('utf-8'))
for line in deduplicated_lines:
csv_complete_gcs_fd.write(line)
csv_complete_gcs_fd.close()
def post(self):
# get args
self.start_cursor = self.request.get('cursor')
self.filtering_event_key = self.request.get('event')
self.filename = self.request.get('filename')
self.csv_header = self.request.get('csv_header')
self.worker_url = self.request.get('worker_url')
self.event = Event.get(
self.filtering_event_key) if self.filtering_event_key else None
# get (base) query, skip query to cursor, filter for sites
query = self.get_base_query()
if self.start_cursor:
query.with_cursor(self.start_cursor)
fetched_sites = query.fetch(limit=self.sites_per_task)
sites = self.filter_sites(fetched_sites)
# write part of csv file to GCS
csv_part_gcs_fd = cloudstorage.open(BUCKET_NAME + '/' + self.filename +
'.part.' + self.start_cursor,
'w',
content_type='text/csv')
self._write_csv_rows(csv_part_gcs_fd, sites)
csv_part_gcs_fd.close()
# decide what to do next
self.end_cursor = query.cursor()
if self.end_cursor and self.start_cursor != self.end_cursor:
# chain to next task
taskqueue.add(
url=self.worker_url,
params=self.get_continuation_param_dict(),
retry_options=taskqueue.TaskRetryOptions(task_retry_limit=3),
)
else:
# finish file: combine parts and deduplicate lines
logging.info(u"Deduplicating to create %s ..." % self.filename)
sio = StringIO()
path_prefix = BUCKET_NAME + '/' + self.filename + '.part'
for gcs_file_stat in cloudstorage.listbucket(path_prefix):
csv_part_gcs_fd = cloudstorage.open(gcs_file_stat.filename)
for line in csv_part_gcs_fd:
sio.write(line)
csv_part_gcs_fd.close()
sio.seek(0)
deduplicated_lines = set(line for line in sio)
# write csv header and deduplicated lines to new file
csv_complete_gcs_fd = cloudstorage.open(BUCKET_NAME + '/' +
self.filename,
'w',
content_type='text/csv')
csv_complete_gcs_fd.write(self.csv_header.encode('utf-8'))
for line in deduplicated_lines:
csv_complete_gcs_fd.write(line)
csv_complete_gcs_fd.close()
def _crunch_and_save(cls, event_key):
event = Event.get(event_key)
# crunch
stats_d = crunch_incident_statistics(event)
csv_content = incident_statistics_csv(stats_d)
html_content = incident_statistics_html(stats_d)
# save csv & html
csv_gcs_fd = cloudstorage.open(BUCKET_NAME + '/' +
incident_statistics_csv_filename(event),
'w',
content_type='text/csv')
csv_gcs_fd.write(csv_content.encode('utf-8'))
csv_gcs_fd.close()
html_gcs_fd = cloudstorage.open(
BUCKET_NAME + '/' + incident_statistics_html_filename(event),
'w',
content_type='text/html')
html_gcs_fd.write(html_content.encode('utf-8'))
html_gcs_fd.close()
def post(self):
# get args
self.start_cursor = self.request.get('cursor')
self.filtering_event_key = self.request.get('event')
self.filename = self.request.get('filename')
self.csv_header = self.request.get('csv_header')
self.worker_url = self.request.get('worker_url')
self.event = Event.get(self.filtering_event_key) if self.filtering_event_key else None
# get (base) query, skip query to cursor, filter for sites
query = self.get_base_query()
if self.start_cursor:
query.with_cursor(self.start_cursor)
fetched_sites = query.fetch(limit=self.sites_per_task)
sites = self.filter_sites(fetched_sites)
# try deleting before uploading
try:
logging.info("try to delete bucket")
cloudstorage.delete(BUCKET_NAME + '/' + self.filename)
except Exception, e:
logging.error("Deleting bucket failed: %s" % e)
def _crunch_and_save(cls, event_key):
event = Event.get(event_key)
# crunch
stats_d = crunch_incident_statistics(event)
csv_content = incident_statistics_csv(stats_d)
html_content = incident_statistics_html(stats_d)
# save csv & html
csv_gcs_fd = cloudstorage.open(
BUCKET_NAME + '/' + incident_statistics_csv_filename(event),
'w',
content_type='text/csv'
)
csv_gcs_fd.write(csv_content.encode('utf-8'))
csv_gcs_fd.close()
html_gcs_fd = cloudstorage.open(
BUCKET_NAME + '/' + incident_statistics_html_filename(event),
'w',
content_type='text/html'
)
html_gcs_fd.write(html_content.encode('utf-8'))
html_gcs_fd.close()
def get_base_query(self):
query = Site.all()
if self.filtering_event_key:
query.filter('event', Event.get(self.filtering_event_key))
return query
def get_base_query(self):
query = Site.all()
if self.filtering_event_key:
query.filter('event', Event.get(self.filtering_event_key))
return query
