in utils/fix-audio-encoding/convert_from_gcs.py [0:0]
def _GetGcsUris(bucket, project_id, impersonated_service_account):
"""Returns a list of GCS uris for files in a bucket.
Args:
bucket: The GCS bucket.
project_id: The project ID (not number) to use.
impersonated_service_account: The service account to impersonate.
Returns:
The GCS uris.
"""
uris, files = [], []
storage_client = storage.Client(
project=project_id,
credentials=_GetClientCredentials(impersonated_service_account),
)
blobs = storage_client.list_blobs(bucket)
for blob in blobs:
# Blobs ending in slashes are actually directory paths.
if not blob.name.endswith('/'):
uris.append(_GetGcsUri(bucket, blob.name))
files.append(blob.name)
return uris, files