# TODO:
# --md5sum - display md5 of each file as read from disk
-import apiclient.errors
import argparse
import arvados
import base64
import socket
import sys
import tempfile
+from apiclient import errors as apiclient_errors
import arvados.commands._util as arv_cmd
upload_opts = argparse.ArgumentParser(add_help=False)
upload_opts.add_argument('paths', metavar='path', type=str, nargs='*',
- help="""
+ help="""
Local file or directory. Default: read from standard input.
""")
-upload_opts.add_argument('--max-manifest-depth', type=int, metavar='N',
+_group = upload_opts.add_mutually_exclusive_group()
+
+_group.add_argument('--max-manifest-depth', type=int, metavar='N',
default=-1, help="""
Maximum depth of directory tree to represent in the manifest
structure. A directory structure deeper than this will be represented
stream per filesystem directory that contains files.
""")
+_group.add_argument('--normalize', action='store_true',
+ help="""
+Normalize the manifest by re-ordering files and streams after writing
+data.
+""")
+
_group = upload_opts.add_mutually_exclusive_group()
_group.add_argument('--as-stream', action='store_true', dest='stream',
- help="""
+ help="""
Synonym for --stream.
""")
_group.add_argument('--stream', action='store_true',
- help="""
+ help="""
Store the file content and display the resulting manifest on
stdout. Do not write the manifest to Keep or save a Collection object
in Arvados.
""")
_group.add_argument('--as-manifest', action='store_true', dest='manifest',
- help="""
+ help="""
Synonym for --manifest.
""")
_group.add_argument('--in-manifest', action='store_true', dest='manifest',
- help="""
+ help="""
Synonym for --manifest.
""")
_group.add_argument('--manifest', action='store_true',
- help="""
+ help="""
Store the file data and resulting manifest in Keep, save a Collection
object in Arvados, and display the manifest locator (Collection uuid)
on stdout. This is the default behavior.
""")
_group.add_argument('--as-raw', action='store_true', dest='raw',
- help="""
+ help="""
Synonym for --raw.
""")
_group.add_argument('--raw', action='store_true',
- help="""
+ help="""
Store the file content and display the data block locators on stdout,
separated by commas, with a trailing newline. Do not store a
manifest.
""")
upload_opts.add_argument('--use-filename', type=str, default=None,
- dest='filename', help="""
+ dest='filename', help="""
Synonym for --filename.
""")
upload_opts.add_argument('--filename', type=str, default=None,
- help="""
+ help="""
Use the given filename in the manifest, instead of the name of the
local file. This is useful when "-" or "/dev/stdin" is given as an
input file. It can be used only if there is exactly one path given and
""")
upload_opts.add_argument('--portable-data-hash', action='store_true',
- help="""
+ help="""
Print the portable data hash instead of the Arvados UUID for the collection
created by the upload.
""")
_group = run_opts.add_mutually_exclusive_group()
_group.add_argument('--progress', action='store_true',
- help="""
+ help="""
Display human-readable progress on stderr (bytes and, if possible,
percentage of total data size). This is the default behavior when
stderr is a tty.
""")
_group.add_argument('--no-progress', action='store_true',
- help="""
+ help="""
Do not display human-readable progress on stderr, even if stderr is a
tty.
""")
_group.add_argument('--batch-progress', action='store_true',
- help="""
+ help="""
Display machine-readable progress on stderr (bytes and, if known,
total data size).
""")
_group = run_opts.add_mutually_exclusive_group()
_group.add_argument('--resume', action='store_true', default=True,
- help="""
+ help="""
Continue interrupted uploads from cached state (default).
""")
_group.add_argument('--no-resume', action='store_false', dest='resume',
- help="""
+ help="""
Do not continue interrupted uploads from cached state.
""")
try:
project_uuid = desired_project_uuid(api_client, args.project_uuid,
args.retries)
- except (apiclient.errors.Error, ValueError) as error:
+ except (apiclient_errors.Error, ValueError) as error:
print >>stderr, error
sys.exit(1)
if args.stream:
output = writer.manifest_text()
+ if args.normalize:
+ output = CollectionReader(output).manifest_text(normalize=True)
elif args.raw:
output = ','.join(writer.data_locators())
else:
try:
+ manifest_text = writer.manifest_text()
+ if args.normalize:
+ manifest_text = CollectionReader(manifest_text).manifest_text(normalize=True)
# Register the resulting collection in Arvados.
collection = api_client.collections().create(
body={
'owner_uuid': project_uuid,
'name': collection_name,
- 'manifest_text': writer.manifest_text()
+ 'manifest_text': manifest_text
},
ensure_unique_name=True
).execute(num_retries=args.retries)
else:
output = collection['uuid']
- except apiclient.errors.Error as error:
+ except apiclient_errors.Error as error:
print >>stderr, (
"arv-put: Error creating Collection on project: {}.".format(
error))