X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/529a648458ceefc02d77abf470a2005a099d9907..36360d1f2987cea89ce217a2519c01da7456d533:/sdk/python/arvados/commands/run.py diff --git a/sdk/python/arvados/commands/run.py b/sdk/python/arvados/commands/run.py index 7a298b4c17..860fc8f805 100644 --- a/sdk/python/arvados/commands/run.py +++ b/sdk/python/arvados/commands/run.py @@ -11,22 +11,38 @@ import put import time import subprocess import logging +import sys import arvados.commands._util as arv_cmd +from arvados._version import __version__ + logger = logging.getLogger('arvados.arv-run') logger.setLevel(logging.INFO) arvrun_parser = argparse.ArgumentParser(parents=[arv_cmd.retry_opt]) -arvrun_parser.add_argument('--dry-run', action="store_true", help="Print out the pipeline that would be submitted and exit") -arvrun_parser.add_argument('--local', action="store_true", help="Run locally using arv-run-pipeline-instance") -arvrun_parser.add_argument('--docker-image', type=str, help="Docker image to use, otherwise use instance default.") -arvrun_parser.add_argument('--ignore-rcode', action="store_true", help="Commands that return non-zero return codes should not be considered failed.") -arvrun_parser.add_argument('--no-reuse', action="store_true", help="Do not reuse past jobs.") -arvrun_parser.add_argument('--no-wait', action="store_true", help="Do not wait and display logs after submitting command, just exit.") -arvrun_parser.add_argument('--project-uuid', type=str, help="Parent project of the pipeline") -arvrun_parser.add_argument('--git-dir', type=str, default="", help="Git repository passed to arv-crunch-job when using --local") -arvrun_parser.add_argument('--repository', type=str, default="arvados", help="repository field of component, default 'arvados'") -arvrun_parser.add_argument('--script-version', type=str, default="master", help="script_version field of component, default 'master'") +arvrun_parser.add_argument('--dry-run', action="store_true", + help="Print out the pipeline that would be submitted and exit") +arvrun_parser.add_argument('--local', action="store_true", + help="Run locally using arv-run-pipeline-instance") +arvrun_parser.add_argument('--docker-image', type=str, + help="Docker image to use, otherwise use instance default.") +arvrun_parser.add_argument('--ignore-rcode', action="store_true", + help="Commands that return non-zero return codes should not be considered failed.") +arvrun_parser.add_argument('--no-reuse', action="store_true", + help="Do not reuse past jobs.") +arvrun_parser.add_argument('--no-wait', action="store_true", + help="Do not wait and display logs after submitting command, just exit.") +arvrun_parser.add_argument('--project-uuid', type=str, + help="Parent project of the pipeline") +arvrun_parser.add_argument('--git-dir', type=str, default="", + help="Git repository passed to arv-crunch-job when using --local") +arvrun_parser.add_argument('--repository', type=str, default="arvados", + help="repository field of component, default 'arvados'") +arvrun_parser.add_argument('--script-version', type=str, default="master", + help="script_version field of component, default 'master'") +arvrun_parser.add_argument('--version', action='version', + version="%s %s" % (sys.argv[0], __version__), + help='Print version and exit.') arvrun_parser.add_argument('args', nargs=argparse.REMAINDER) class ArvFile(object): @@ -110,7 +126,7 @@ def statfile(prefix, fn, fnPattern="$(file %s/%s)", dirPattern="$(dir %s/%s/)"): def uploadfiles(files, api, dry_run=False, num_retries=0, project=None, fnPattern="$(file %s/%s)", name=None): # Find the smallest path prefix that includes all the files that need to be uploaded. # This starts at the root and iteratively removes common parent directory prefixes - # until all file pathes no longer have a common parent. + # until all file paths no longer have a common parent. n = True pathprefix = "/" while n: @@ -155,9 +171,13 @@ def uploadfiles(files, api, dry_run=False, num_retries=0, project=None, fnPatter collection.start_new_stream(stream) collection.write_file(f.fn, sp[1]) - exists = api.collections().list(filters=[["owner_uuid", "=", project], - ["portable_data_hash", "=", collection.portable_data_hash()], - ["name", "=", name]]).execute(num_retries=num_retries) + filters=[["portable_data_hash", "=", collection.portable_data_hash()], + ["name", "like", name+"%"]] + if project: + filters.append(["owner_uuid", "=", project]) + + exists = api.collections().list(filters=filters).execute(num_retries=num_retries) + if exists["items"]: item = exists["items"][0] logger.info("Using collection %s", item["uuid"]) @@ -171,6 +191,7 @@ def uploadfiles(files, api, dry_run=False, num_retries=0, project=None, fnPatter pdh = item["portable_data_hash"] for c in files: + c.keepref = "%s/%s" % (pdh, c.fn) c.fn = fnPattern % (pdh, c.fn) os.chdir(orgdir)