Merge branch '18027-unmount-fuse'

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

diff --git a/sdk/python/arvados/commands/get.py b/sdk/python/arvados/commands/get.py
index 48f076b4204b5471636e9554459ff3c656919ba1..eb682976253b66a3c0c6c0ebd0a9dc5ca306d499 100755 (executable)
--- a/sdk/python/arvados/commands/get.py
+++ b/sdk/python/arvados/commands/get.py
@@ -1,4 +1,7 @@
-#!/usr/bin/env python
+#!/usr/bin/env python3
+# Copyright (C) The Arvados Authors. All rights reserved.
+#
+# SPDX-License-Identifier: Apache-2.0

 
 import argparse
 import hashlib
 
 import argparse
 import hashlib


@@ -10,6 +13,7 @@ import logging
 
 import arvados
 import arvados.commands._util as arv_cmd
 
 import arvados
 import arvados.commands._util as arv_cmd

+import arvados.util as util

 
 from arvados._version import __version__
 
 
 from arvados._version import __version__
 


@@ -77,6 +81,10 @@ Overwrite existing files while writing. The default behavior is to
 refuse to write *anything* if any of the output files already
 exist. As a special case, -f is not needed to write to stdout.
 """)
 refuse to write *anything* if any of the output files already
 exist. As a special case, -f is not needed to write to stdout.
 """)

+group.add_argument('-v', action='count', default=0,
+                    help="""
+Once for verbose mode, twice for debug mode.
+""")

 group.add_argument('--skip-existing', action='store_true',
                    help="""
 Skip files that already exist. The default behavior is to refuse to
 group.add_argument('--skip-existing', action='store_true',
                    help="""
 Skip files that already exist. The default behavior is to refuse to


@@ -84,6 +92,11 @@ write *anything* if any files exist that would have to be
 overwritten. This option causes even devices, sockets, and fifos to be
 skipped.
 """)
 overwritten. This option causes even devices, sockets, and fifos to be
 skipped.
 """)

+group.add_argument('--strip-manifest', action='store_true', default=False,
+                   help="""
+When getting a collection manifest, strip its access tokens before writing
+it.
+""")

 
 def parse_arguments(arguments, stdout, stderr):
     args = parser.parse_args(arguments)
 
 def parse_arguments(arguments, stdout, stderr):
     args = parser.parse_args(arguments)


@@ -131,19 +144,19 @@ def main(arguments=None, stdout=sys.stdout, stderr=sys.stderr):
         stdout = stdout.buffer
 
     args = parse_arguments(arguments, stdout, stderr)
         stdout = stdout.buffer
 
     args = parse_arguments(arguments, stdout, stderr)

+    logger.setLevel(logging.WARNING - 10 * args.v)
+
+    request_id = arvados.util.new_request_id()
+    logger.info('X-Request-Id: '+request_id)
+

     if api_client is None:
     if api_client is None:

-        api_client = arvados.api('v1')
+        api_client = arvados.api('v1', request_id=request_id)

 
     r = re.search(r'^(.*?)(/.*)?$', args.locator)
 
     r = re.search(r'^(.*?)(/.*)?$', args.locator)

-    collection = r.group(1)
+    col_loc = r.group(1)

     get_prefix = r.group(2)
     if args.r and not get_prefix:
         get_prefix = os.sep
     get_prefix = r.group(2)
     if args.r and not get_prefix:
         get_prefix = os.sep

-    try:
-        reader = arvados.CollectionReader(collection, num_retries=args.retries)
-    except Exception as error:
-        logger.error("failed to read collection: {}".format(error))
-        return 1

 
     # User asked to download the collection's manifest
     if not get_prefix:
 
     # User asked to download the collection's manifest
     if not get_prefix:


@@ -153,19 +166,36 @@ def main(arguments=None, stdout=sys.stdout, stderr=sys.stderr):
                 open_flags |= os.O_EXCL
             try:
                 if args.destination == "-":
                 open_flags |= os.O_EXCL
             try:
                 if args.destination == "-":

-                    stdout.write(reader.manifest_text().encode())
+                    write_block_or_manifest(
+                        dest=stdout, src=col_loc,
+                        api_client=api_client, args=args)

                 else:
                     out_fd = os.open(args.destination, open_flags)
                     with os.fdopen(out_fd, 'wb') as out_file:
                 else:
                     out_fd = os.open(args.destination, open_flags)
                     with os.fdopen(out_fd, 'wb') as out_file:

-                        out_file.write(reader.manifest_text().encode())
+                        write_block_or_manifest(
+                            dest=out_file, src=col_loc,
+                            api_client=api_client, args=args)

             except (IOError, OSError) as error:
                 logger.error("can't write to '{}': {}".format(args.destination, error))
                 return 1
             except (arvados.errors.ApiError, arvados.errors.KeepReadError) as error:
             except (IOError, OSError) as error:
                 logger.error("can't write to '{}': {}".format(args.destination, error))
                 return 1
             except (arvados.errors.ApiError, arvados.errors.KeepReadError) as error:

-                logger.error("failed to download '{}': {}".format(collection, error))
+                logger.error("failed to download '{}': {}".format(col_loc, error))

                 return 1
                 return 1

+            except arvados.errors.ArgumentError as error:
+                if 'Argument to CollectionReader' in str(error):
+                    logger.error("error reading collection: {}".format(error))
+                    return 1
+                else:
+                    raise

         return 0
 
         return 0
 

+    try:
+        reader = arvados.CollectionReader(
+            col_loc, api_client=api_client, num_retries=args.retries)
+    except Exception as error:
+        logger.error("failed to read collection: {}".format(error))
+        return 1
+

     # Scan the collection. Make an array of (stream, file, local
     # destination filename) tuples, and add up total size to extract.
     todo = []
     # Scan the collection. Make an array of (stream, file, local
     # destination filename) tuples, and add up total size to extract.
     todo = []


@@ -278,3 +308,14 @@ def files_in_collection(c):
         elif isinstance(c[i], arvados.collection.Subcollection):
             for s, f in files_in_collection(c[i]):
                 yield (s, f)
         elif isinstance(c[i], arvados.collection.Subcollection):
             for s, f in files_in_collection(c[i]):
                 yield (s, f)

+
+def write_block_or_manifest(dest, src, api_client, args):
+    if '+A' in src:
+        # block locator
+        kc = arvados.keep.KeepClient(api_client=api_client)
+        dest.write(kc.get(src, num_retries=args.retries))
+    else:
+        # collection UUID or portable data hash
+        reader = arvados.CollectionReader(
+            src, api_client=api_client, num_retries=args.retries)
+        dest.write(reader.manifest_text(strip=args.strip_manifest).encode())