def resolve_uri(self, data):
uri, storage = get_uri_via_regex(data, prefixes=(url_scheme,))
if not storage:
return
logger.debug("Found matching storage uri in task data value: {uri}".format(uri=uri))
resolved_uri = self.resolve_gs(uri)
return data.replace(uri, resolved_uri)
def can_resolve_url(self, url):
# TODO: later check to the full prefix like "url.startswith(self.path_full)"
# Search of occurrences inside string, e.g. for cases like "gs://bucket/file.pdf" or "<embed src='gs://bucket/file.pdf'/>" # noqa
_, storage = get_uri_via_regex(url, prefixes=(self.url_scheme, ))
if storage == self.url_scheme:
return True
# if not found any occurrences - this Storage can't resolve url
return False
def resolve_uri(self, data):
try:
uri, storage = get_uri_via_regex(data, prefixes=(url_scheme,))
if not storage:
return
resolved_uri = resolve_s3_url(uri, self.get_client(), self.presign)
return data.replace(uri, resolved_uri)
except NoCredentialsError:
logger.warning(f'No AWS credentials specified for {data}')
def resolve_uri(self, uri):
try:
extracted_uri, extracted_storage = get_uri_via_regex(
uri, prefixes=(self.url_scheme, ))
if not extracted_storage:
logger.info(f'No storage info found for URI={uri}')
return
http_url = self.generate_http_url(extracted_uri)
return uri.replace(extracted_uri, http_url)
except Exception as exc:
logger.error(f'Can\'t resolve URI={uri}. Reason: {exc}',
exc_info=True)
