get of a file with "+" in the name fails with [Runtime Failure] for google.cloud

[novodim]

$ s4cmd -c 32 --debug get --endpoint-url https://storage.googleapis.com s3://mib_dump/test/file_+.txt /tmp/abcd
(D)s4cmd:631 read S3 keys from environment
(D)s4cmd:129 >> get_handler(<main.CommandHandler object at 0x7f8c17a52c10>, ['get', 's3://mib_dump/test/file+.txt', '/tmp/abcd'])
(D)s4cmd:129 >> validate(<main.CommandHandler object at 0x7f8c17a52c10>, 'cmd|s3|local', ['get', 's3://mib_dump/test/file+.txt', '/tmp/abcd'])
(D)s4cmd:131 << validate(<main.CommandHandler object at 0x7f8c17a52c10>, 'cmd|s3|local', ['get', 's3://mib_dump/test/file+.txt', '/tmp/abcd']): None
(D)s4cmd:129 >> get_files(<main.S3Handler object at 0x7f8c17a67150>, 's3://mib_dump/test/file+.txt', '/tmp/abcd')
(D)s4cmd:129 >> s3walk(<main.S3Handler object at 0x7f8c17a67150>, 's3://mib_dump/test/file+.txt', True)
(D)s4cmd:129 >> s3walk(<ThreadUtil(Thread-33, started daemon 140237611177728)>, <main.S3URL instance at 0x7f8bdc45b758>, 'test/file+.txt', 'test/file+.txt', [])
(D)s4cmd:402 >> S3APICALL get_paginator('list_objects')
(D)s4cmd:404 << S3APICALL get_paginator('list_objects'): <botocore.client.S3.Paginator.ListObjects object at 0x7f8b7c647590>
(D)s4cmd:129 >> partial_match(<ThreadUtil(Thread-33, started daemon 140237611177728)>, 'test/file%2B_.txt', 'test/file_+.txt')
(D)s4cmd:131 << partial_match(<ThreadUtil(Thread-33, started daemon 140237611177728)>, 'test/file%2B_.txt', 'test/file_+.txt'): False
(D)s4cmd:131 << s3walk(<ThreadUtil(Thread-33, started daemon 140237611177728)>, <main.S3URL instance at 0x7f8bdc45b758>, 'test/file+.txt', 'test/file+.txt', []): None
(D)s4cmd:131 << s3walk(<main.S3Handler object at 0x7f8c17a67150>, 's3://mib_dump/test/file+_.txt', True): []
(E)s4cmd:183 [Runtime Failure] Source doesn't exist.

$ s4cmd --version
s4cmd version 2.1.0
$ python2 --version
Python 2.7.16