Storage list_blobs iterator doesn't return prefixes

[andtrott]

Title: list_blobs() iterator.prefixes returns empty when API response contains prefixes

Issue Description

When using list_blobs() with a prefix and delimiter to list "directories" in a GCS bucket, the .prefixes property of the returned iterator is empty even though the raw API response shows that prefixes exist.

Reproduction Steps

Here's a complete example using the public bucket gcp-public-data-landsat:

from google.cloud import storage
import logging

# Set up logging to see API responses
logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger(__name__)

# Initialize client and bucket
client = storage.Client()
bucket = client.bucket('gcp-public-data-landsat')

prefix = "LC08/" # This prefix exists in the landsat bucket
delimiter = "/"

# First show that the raw API response contains prefixes
logger.debug("\nChecking raw API response:")
page = bucket.list_blobs(prefix=prefix, delimiter=delimiter)._get_next_page_response()
logger.debug(f"Raw API response: {page}") # Shows prefixes exist

# Now demonstrate that iterator.prefixes is empty
logger.debug("\nTrying to iterate prefixes:")
iterator = bucket.list_blobs(prefix=prefix, delimiter=delimiter)
prefix_count = 0
for prefix in iterator.prefixes:
 prefix_count += 1
 logger.debug(f"Found prefix: {prefix}")
logger.debug(f"Total prefixes found: {prefix_count}") # Shows 0

# Also demonstrate list() returns empty
prefixes = list(iterator.prefixes)
logger.debug(f"List of prefixes: {prefixes}") # Shows []

Expected Behavior

The iterator's .prefixes property should provide access to the prefixes returned in the API response
Both iterating through iterator.prefixes and calling list(iterator.prefixes) should return the prefixes

Actual Behavior

The raw API response shows prefixes exist: {'kind': 'storage#objects', 'prefixes': [...]}
However, iterator.prefixes appears empty when iterated
list(iterator.prefixes) returns an empty list

Workaround

Currently using direct access to the API response as a workaround:
pythonCopypage = bucket.list_blobs(prefix=prefix, delimiter=delimiter)._get_next_page_response()
prefixes = page.get('prefixes', [])

Environment

google-cloud-storage version: 2.18.2
Python version: Python 3.11.6
OS: Darwin 23.6.0

Related Issues

#436
#483

This issue was co-authored by Claude.

[cojenco]

Hi andtrott@ thanks for the question. The iterator has to be called first in order to access the list of prefixes. Try adding a list(iterator)to consume the iterator first

iterator = bucket.list_blobs(prefix=prefix, delimiter=delimiter)
list(iterator)
...
prefix_count = 0
for prefix in iterator.prefixes:

See the list_files_with_prefix example with detailed explanation

See related issues
#1208 (comment)
#863 (comment)

[andtrott]

Thanks for the response! Stand by the expected behavior, but that's a good workaround - thanks!