shunping commented on code in PR #32428:
URL: https://github.com/apache/beam/pull/32428#discussion_r1763469183
##########
sdks/python/apache_beam/options/pipeline_options.py:
##########
@@ -939,6 +939,18 @@ def _add_argparse_args(cls, parser):
help=
'Create metrics reporting the approximate number of bytes written per '
'bucket.')
+ parser.add_argument(
+ '--no_gcsio_throttling_counter',
+ default=False,
+ action='store_true',
+ help='Throttling counter in GcsIO is enabled by default. Set '
+ '--no_gcsio_throttling_counter to avoid it.')
+ parser.add_argument(
+ '--enable_gcsio_blob_generation',
Review Comment:
> the difference of bucket.blob and bucket.get_blob is that the latter will
send an API request which has cost. I remember we had concern about excessive
API request before - https://github.com/apache/beam/pull/30205
Thats' right. I had that in the commit message
https://github.com/apache/beam/pull/32428/commits/f82230d3b52f3bc2b8bbbd92d5fe33d3aa1daf0c,
> Possibly add a disclaimer that this will be at the cost of sending API
request verify blob existence?
Good call. I think I can make that clear in the help message of this option.
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
