Merge branch '4552-collection-unique-name' closes #4552

[arvados.git] / sdk / python / arvados / commands / put.py

diff --git a/sdk/python/arvados/commands/put.py b/sdk/python/arvados/commands/put.py
index 4a926c701c9f45648dc6337fc294391986e03379..d070a8b9c79037b8ea22de911861a6544f773cb1 100644 (file)
--- a/sdk/python/arvados/commands/put.py
+++ b/sdk/python/arvados/commands/put.py
@@ -27,11 +27,13 @@ api_client = None
 upload_opts = argparse.ArgumentParser(add_help=False)
 
 upload_opts.add_argument('paths', metavar='path', type=str, nargs='*',
 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.
 """)
 
 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
                     default=-1, help="""
 Maximum depth of directory tree to represent in the manifest
 structure. A directory structure deeper than this will be represented


@@ -40,56 +42,62 @@ a single stream. Default: -1 (unlimited), i.e., exactly one manifest
 stream per filesystem directory that contains files.
 """)
 
 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',
 _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',
 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',
 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',
 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',
 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',
 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',
 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,
 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,
 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
 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


@@ -97,7 +105,7 @@ it is not a directory. Implies --manifest.
 """)
 
 upload_opts.add_argument('--portable-data-hash', action='store_true',
 """)
 
 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.
 """)
 Print the portable data hash instead of the Arvados UUID for the collection
 created by the upload.
 """)


@@ -115,31 +123,31 @@ Save the collection with the specified name.
 
 _group = run_opts.add_mutually_exclusive_group()
 _group.add_argument('--progress', action='store_true',
 
 _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',
 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',
 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,
 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',
 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.
 """)
 
 Do not continue interrupted uploads from cached state.
 """)
 


@@ -447,16 +455,21 @@ def main(arguments=None, stdout=sys.stdout, stderr=sys.stderr):
 
     if args.stream:
         output = writer.manifest_text()
 
     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:
     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,
             # 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)
                     },
                 ensure_unique_name=True
                 ).execute(num_retries=args.retries)