def RunCommand(self):
"""Command entry point for the rm command."""
# self.recursion_requested is initialized in command.py (so it can be
# checked in parent class for all commands).
self.continue_on_error = self.parallel_operations
self.read_args_from_stdin = False
self.all_versions = False
if self.sub_opts:
for o, unused_a in self.sub_opts:
if o == '-a':
self.all_versions = True
elif o == '-f':
self.continue_on_error = True
elif o == '-I':
self.read_args_from_stdin = True
elif o == '-r' or o == '-R':
self.recursion_requested = True
self.all_versions = True
if self.read_args_from_stdin:
if self.args:
raise CommandException(
'No arguments allowed with the -I flag.')
url_strs = StdinIterator()
else:
if not self.args:
raise CommandException(
'The rm command (without -I) expects at '
'least one URL.')
url_strs = self.args
# Tracks number of object deletes that failed.
self.op_failure_count = 0
# Tracks if any buckets were missing.
self.bucket_not_found_count = 0
# Tracks buckets that are slated for recursive deletion.
bucket_urls_to_delete = []
self.bucket_strings_to_delete = []
if self.recursion_requested:
bucket_fields = ['id']
for url_str in url_strs:
url = StorageUrlFromString(url_str)
if url.IsBucket() or url.IsProvider():
for blr in self.WildcardIterator(url_str).IterBuckets(
bucket_fields=bucket_fields):
bucket_urls_to_delete.append(blr.storage_url)
self.bucket_strings_to_delete.append(url_str)
self.preconditions = PreconditionsFromHeaders(self.headers or {})
try:
# Expand wildcards, dirs, buckets, and bucket subdirs in URLs.
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
url_strs,
self.recursion_requested,
project_id=self.project_id,
all_versions=self.all_versions,
continue_on_error=self.continue_on_error
or self.parallel_operations)
seek_ahead_iterator = None
# Cannot seek ahead with stdin args, since we can only iterate them
# once without buffering in memory.
if not self.read_args_from_stdin:
seek_ahead_iterator = SeekAheadNameExpansionIterator(
self.command_name,
self.debug,
self.GetSeekAheadGsutilApi(),
url_strs,
self.recursion_requested,
all_versions=self.all_versions,
project_id=self.project_id)
# Perform remove requests in parallel (-m) mode, if requested, using
# configured number of parallel processes and threads. Otherwise,
# perform requests with sequential function calls in current process.
self.Apply(
_RemoveFuncWrapper,
name_expansion_iterator,
_RemoveExceptionHandler,
fail_on_error=(not self.continue_on_error),
shared_attrs=['op_failure_count', 'bucket_not_found_count'],
seek_ahead_iterator=seek_ahead_iterator)
# Assuming the bucket has versioning enabled, url's that don't map to
# objects should throw an error even with all_versions, since the prior
# round of deletes only sends objects to a history table.
# This assumption that rm -a is only called for versioned buckets should be
# corrected, but the fix is non-trivial.
except CommandException as e:
# Don't raise if there are buckets to delete -- it's valid to say:
# gsutil rm -r gs://some_bucket
# if the bucket is empty.
if _ExceptionMatchesBucketToDelete(self.bucket_strings_to_delete,
e):
DecrementFailureCount()
else:
raise
except ServiceException, e:
if not self.continue_on_error:
raise
def _SetIam(self):
"""Set IAM policy for given wildcards on the command line."""
self.continue_on_error = False
self.recursion_requested = False
self.all_versions = False
force_etag = False
etag = ''
if self.sub_opts:
for o, arg in self.sub_opts:
if o in ['-r', '-R']:
self.recursion_requested = True
elif o == '-f':
self.continue_on_error = True
elif o == '-a':
self.all_versions = True
elif o == '-e':
etag = str(arg)
force_etag = True
else:
self.RaiseInvalidArgumentException()
file_url = self.args[0]
patterns = self.args[1:]
# Load the IAM policy file and raise error if the file is invalid JSON or
# does not exist.
try:
with open(file_url, 'r') as fp:
policy = json.loads(fp.read())
except IOError:
raise ArgumentException(
'Specified IAM policy file "%s" does not exist.' % file_url)
except ValueError as e:
self.logger.debug('Invalid IAM policy file, ValueError:\n%s', e)
raise ArgumentException('Invalid IAM policy file "%s".' % file_url)
bindings = policy.get('bindings', [])
if not force_etag:
etag = policy.get('etag', '')
policy_json = json.dumps({
'bindings': bindings,
'etag': etag,
'version': IAM_POLICY_VERSION
})
try:
policy = protojson.decode_message(apitools_messages.Policy,
policy_json)
except DecodeError:
raise ArgumentException(
'Invalid IAM policy file "%s" or etag "%s".' %
(file_url, etag))
self.everything_set_okay = True
# This list of wildcard strings will be handled by NameExpansionIterator.
threaded_wildcards = []
for pattern in patterns:
surl = StorageUrlFromString(pattern)
if surl.IsBucket():
if self.recursion_requested:
surl.object_name = '*'
threaded_wildcards.append(surl.url_string)
else:
self.SetIamHelper(surl, policy)
else:
threaded_wildcards.append(surl.url_string)
# N.B.: If threaded_wildcards contains a non-existent bucket
# (e.g. ["gs://non-existent", "gs://existent"]), NameExpansionIterator
# will raise an exception in iter.next. This halts all iteration, even
# when -f is set. This behavior is also evident in acl set. This behavior
# also appears for any exception that will be raised when iterating over
# wildcard expansions (access denied if bucket cannot be listed, etc.).
if threaded_wildcards:
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
threaded_wildcards,
self.recursion_requested,
all_versions=self.all_versions,
continue_on_error=self.continue_on_error
or self.parallel_operations,
bucket_listing_fields=['name'])
seek_ahead_iterator = SeekAheadNameExpansionIterator(
self.command_name,
self.debug,
self.GetSeekAheadGsutilApi(),
threaded_wildcards,
self.recursion_requested,
all_versions=self.all_versions)
policy_it = itertools.repeat(protojson.encode_message(policy))
self.Apply(_SetIamWrapper,
zip(policy_it, name_expansion_iterator),
_SetIamExceptionHandler,
fail_on_error=not self.continue_on_error,
seek_ahead_iterator=seek_ahead_iterator)
self.everything_set_okay &= not GetFailureCount() > 0
# TODO: Add an error counter for files and objects.
if not self.everything_set_okay:
raise CommandException('Some IAM policies could not be set.')
def RunCommand(self):
"""Command entry point for the du command."""
self.line_ending = '\n'
self.all_versions = False
self.produce_total = False
self.human_readable = False
self.summary_only = False
self.exclude_patterns = []
if self.sub_opts:
for o, a in self.sub_opts:
if o == '-0':
self.line_ending = '\0'
elif o == '-a':
self.all_versions = True
elif o == '-c':
self.produce_total = True
elif o == '-e':
self.exclude_patterns.append(a)
elif o == '-h':
self.human_readable = True
elif o == '-s':
self.summary_only = True
elif o == '-X':
if a == '-':
f = sys.stdin
else:
f = open(a, 'r')
try:
for line in f:
line = line.strip()
if line:
self.exclude_patterns.append(line)
finally:
f.close()
if not self.args:
# Default to listing all gs buckets.
self.args = ['gs://']
total_bytes = 0
got_nomatch_errors = False
def _PrintObjectLong(blr):
return self._PrintInfoAboutBucketListingRef(blr)
def _PrintNothing(unused_blr=None):
pass
def _PrintDirectory(num_bytes, name):
if not self.summary_only:
self._PrintSummaryLine(num_bytes, name)
for url_arg in self.args:
top_level_storage_url = StorageUrlFromString(url_arg)
if top_level_storage_url.IsFileUrl():
raise CommandException('Only cloud URLs are supported for %s' %
self.command_name)
bucket_listing_fields = ['size']
ls_helper = LsHelper(self.WildcardIterator,
self.logger,
print_object_func=_PrintObjectLong,
print_dir_func=_PrintNothing,
print_dir_header_func=_PrintNothing,
print_dir_summary_func=_PrintDirectory,
print_newline_func=_PrintNothing,
all_versions=self.all_versions,
should_recurse=True,
exclude_patterns=self.exclude_patterns,
fields=bucket_listing_fields)
# ls_helper expands to objects and prefixes, so perform a top-level
# expansion first.
if top_level_storage_url.IsProvider():
# Provider URL: use bucket wildcard to iterate over all buckets.
top_level_iter = self.WildcardIterator(
'%s://*' % top_level_storage_url.scheme).IterBuckets(
bucket_fields=['id'])
elif top_level_storage_url.IsBucket():
top_level_iter = self.WildcardIterator(
'%s://%s' %
(top_level_storage_url.scheme,
top_level_storage_url.bucket_name)).IterBuckets(
bucket_fields=['id'])
else:
top_level_iter = [BucketListingObject(top_level_storage_url)]
for blr in top_level_iter:
storage_url = blr.storage_url
if storage_url.IsBucket() and self.summary_only:
storage_url = StorageUrlFromString(
storage_url.CreatePrefixUrl(wildcard_suffix='**'))
_, exp_objs, exp_bytes = ls_helper.ExpandUrlAndPrint(
storage_url)
if (storage_url.IsObject() and exp_objs == 0
and ContainsWildcard(url_arg)
and not self.exclude_patterns):
got_nomatch_errors = True
total_bytes += exp_bytes
if self.summary_only:
self._PrintSummaryLine(exp_bytes,
blr.url_string.rstrip('/'))
if self.produce_total:
self._PrintSummaryLine(total_bytes, 'total')
if got_nomatch_errors:
raise CommandException('One or more URLs matched no objects.')
return 0
def RunCommand(self):
"""Command entry point for the rm command."""
# self.recursion_requested is initialized in command.py (so it can be
# checked in parent class for all commands).
self.continue_on_error = False
self.all_versions = False
if self.sub_opts:
for o, unused_a in self.sub_opts:
if o == '-a':
self.all_versions = True
elif o == '-f':
self.continue_on_error = True
elif o == '-r' or o == '-R':
self.recursion_requested = True
self.all_versions = True
bucket_urls_to_delete = []
bucket_strings_to_delete = []
if self.recursion_requested:
bucket_fields = ['id']
for url_str in self.args:
url = StorageUrlFromString(url_str)
if url.IsBucket() or url.IsProvider():
for blr in self.WildcardIterator(url_str).IterBuckets(
bucket_fields=bucket_fields):
bucket_urls_to_delete.append(blr.storage_url)
bucket_strings_to_delete.append(url_str)
# Used to track if any files failed to be removed.
self.everything_removed_okay = True
try:
# Expand wildcards, dirs, buckets, and bucket subdirs in URLs.
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
self.args,
self.recursion_requested,
project_id=self.project_id,
all_versions=self.all_versions,
continue_on_error=self.continue_on_error
or self.parallel_operations)
# Perform remove requests in parallel (-m) mode, if requested, using
# configured number of parallel processes and threads. Otherwise,
# perform requests with sequential function calls in current process.
self.Apply(_RemoveFuncWrapper,
name_expansion_iterator,
_RemoveExceptionHandler,
fail_on_error=(not self.continue_on_error))
# Assuming the bucket has versioning enabled, url's that don't map to
# objects should throw an error even with all_versions, since the prior
# round of deletes only sends objects to a history table.
# This assumption that rm -a is only called for versioned buckets should be
# corrected, but the fix is non-trivial.
except CommandException as e:
# Don't raise if there are buckets to delete -- it's valid to say:
# gsutil rm -r gs://some_bucket
# if the bucket is empty.
if not bucket_urls_to_delete and not self.continue_on_error:
raise
# Reset the failure count if we failed due to an empty bucket that we're
# going to delete.
msg = 'No URLs matched: '
if msg in str(e):
parts = str(e).split(msg)
if len(parts) == 2 and parts[1] in bucket_strings_to_delete:
ResetFailureCount()
except ServiceException, e:
if not self.continue_on_error:
raise
def _PatchIam(self):
self.continue_on_error = False
self.recursion_requested = False
patch_bindings_tuples = []
if self.sub_opts:
for o, a in self.sub_opts:
if o in ['-r', '-R']:
self.recursion_requested = True
elif o == '-f':
self.continue_on_error = True
elif o == '-d':
patch_bindings_tuples.append(BindingStringToTuple(
False, a))
patterns = []
# N.B.: self.sub_opts stops taking in options at the first non-flagged
# token. The rest of the tokens are sent to self.args. Thus, in order to
# handle input of the form "-d <binding> <binding> <url>", we will have to
# parse self.args for a mix of both bindings and CloudUrls. We are not
# expecting to come across the -r, -f flags here.
it = iter(self.args)
for token in it:
if STORAGE_URI_REGEX.match(token):
patterns.append(token)
break
if token == '-d':
patch_bindings_tuples.append(
BindingStringToTuple(False, next(it)))
else:
patch_bindings_tuples.append(BindingStringToTuple(True, token))
if not patch_bindings_tuples:
raise CommandException('Must specify at least one binding.')
# All following arguments are urls.
for token in it:
patterns.append(token)
self.everything_set_okay = True
self.tried_ch_on_resource_with_conditions = False
threaded_wildcards = []
for pattern in patterns:
surl = StorageUrlFromString(pattern)
try:
if surl.IsBucket():
if self.recursion_requested:
surl.object = '*'
threaded_wildcards.append(surl.url_string)
else:
self.PatchIamHelper(surl, patch_bindings_tuples)
else:
threaded_wildcards.append(surl.url_string)
except AttributeError:
error_msg = 'Invalid Cloud URL "%s".' % surl.object_name
if set(surl.object_name).issubset(set('-Rrf')):
error_msg += (
' This resource handle looks like a flag, which must appear '
'before all bindings. See "gsutil help iam ch" for more details.'
)
raise CommandException(error_msg)
if threaded_wildcards:
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
threaded_wildcards,
self.recursion_requested,
all_versions=self.all_versions,
continue_on_error=self.continue_on_error
or self.parallel_operations,
bucket_listing_fields=['name'])
seek_ahead_iterator = SeekAheadNameExpansionIterator(
self.command_name,
self.debug,
self.GetSeekAheadGsutilApi(),
threaded_wildcards,
self.recursion_requested,
all_versions=self.all_versions)
serialized_bindings_tuples_it = itertools.repeat(
[SerializeBindingsTuple(t) for t in patch_bindings_tuples])
self.Apply(_PatchIamWrapper,
zip(serialized_bindings_tuples_it,
name_expansion_iterator),
_PatchIamExceptionHandler,
fail_on_error=not self.continue_on_error,
seek_ahead_iterator=seek_ahead_iterator)
self.everything_set_okay &= not GetFailureCount() > 0
# TODO: Add an error counter for files and objects.
if not self.everything_set_okay:
msg = 'Some IAM policies could not be patched.'
if self.tried_ch_on_resource_with_conditions:
msg += '\n'
msg += '\n'.join(
textwrap.wrap(
'Some resources had conditions present in their IAM policy '
'bindings, which is not supported by "iam ch". %s' %
(IAM_CH_CONDITIONS_WORKAROUND_MSG)))
raise CommandException(msg)
def RunCommand(self):
"""Command entry point for the ls command."""
got_nomatch_errors = False
got_bucket_nomatch_errors = False
listing_style = ListingStyle.SHORT
get_bucket_info = False
self.recursion_requested = False
self.all_versions = False
self.include_etag = False
self.human_readable = False
self.list_subdir_contents = True
if self.sub_opts:
for o, a in self.sub_opts:
if o == '-a':
self.all_versions = True
elif o == '-e':
self.include_etag = True
elif o == '-b':
get_bucket_info = True
elif o == '-h':
self.human_readable = True
elif o == '-l':
listing_style = ListingStyle.LONG
elif o == '-L':
listing_style = ListingStyle.LONG_LONG
elif o == '-p':
self.project_id = a
elif o == '-r' or o == '-R':
self.recursion_requested = True
elif o == '-d':
self.list_subdir_contents = False
if not self.args:
# default to listing all gs buckets
self.args = ['gs://']
total_objs = 0
total_bytes = 0
def MaybePrintBucketHeader(blr):
if len(self.args) > 1:
print '%s:' % blr.url_string.encode(UTF8)
print_bucket_header = MaybePrintBucketHeader
for url_str in self.args:
storage_url = StorageUrlFromString(url_str)
if storage_url.IsFileUrl():
raise CommandException('Only cloud URLs are supported for %s' %
self.command_name)
bucket_fields = None
if (listing_style == ListingStyle.SHORT
or listing_style == ListingStyle.LONG):
bucket_fields = ['id']
elif listing_style == ListingStyle.LONG_LONG:
bucket_fields = [
'location', 'storageClass', 'versioning', 'acl',
'defaultObjectAcl', 'website', 'logging', 'cors',
'lifecycle'
]
if storage_url.IsProvider():
# Provider URL: use bucket wildcard to list buckets.
for blr in self.WildcardIterator(
'%s://*' % storage_url.scheme).IterBuckets(
bucket_fields=bucket_fields):
self._PrintBucketInfo(blr, listing_style)
elif storage_url.IsBucket() and get_bucket_info:
# ls -b bucket listing request: List info about bucket(s).
total_buckets = 0
for blr in self.WildcardIterator(url_str).IterBuckets(
bucket_fields=bucket_fields):
if not ContainsWildcard(url_str) and not blr.root_object:
# Iterator does not make an HTTP call for non-wildcarded
# listings with fields=='id'. Ensure the bucket exists by calling
# GetBucket.
self.gsutil_api.GetBucket(blr.storage_url.bucket_name,
fields=['id'],
provider=storage_url.scheme)
self._PrintBucketInfo(blr, listing_style)
total_buckets += 1
if not ContainsWildcard(url_str) and not total_buckets:
got_bucket_nomatch_errors = True
else:
# URL names a bucket, object, or object subdir ->
# list matching object(s) / subdirs.
def _PrintPrefixLong(blr):
print '%-33s%s' % ('', blr.url_string.encode(UTF8))
if listing_style == ListingStyle.SHORT:
# ls helper by default readies us for a short listing.
ls_helper = LsHelper(
self.WildcardIterator,
self.logger,
all_versions=self.all_versions,
print_bucket_header_func=print_bucket_header,
should_recurse=self.recursion_requested,
list_subdir_contents=self.list_subdir_contents)
elif listing_style == ListingStyle.LONG:
bucket_listing_fields = ['name', 'updated', 'size']
if self.all_versions:
bucket_listing_fields.extend(
['generation', 'metageneration'])
if self.include_etag:
bucket_listing_fields.append('etag')
ls_helper = LsHelper(
self.WildcardIterator,
self.logger,
print_object_func=self._PrintLongListing,
print_dir_func=_PrintPrefixLong,
print_bucket_header_func=print_bucket_header,
all_versions=self.all_versions,
should_recurse=self.recursion_requested,
fields=bucket_listing_fields,
list_subdir_contents=self.list_subdir_contents)
elif listing_style == ListingStyle.LONG_LONG:
# List all fields
bucket_listing_fields = None
ls_helper = LsHelper(
self.WildcardIterator,
self.logger,
print_object_func=PrintFullInfoAboutObject,
print_dir_func=_PrintPrefixLong,
print_bucket_header_func=print_bucket_header,
all_versions=self.all_versions,
should_recurse=self.recursion_requested,
fields=bucket_listing_fields,
list_subdir_contents=self.list_subdir_contents)
else:
raise CommandException('Unknown listing style: %s' %
listing_style)
exp_dirs, exp_objs, exp_bytes = ls_helper.ExpandUrlAndPrint(
storage_url)
if storage_url.IsObject() and exp_objs == 0 and exp_dirs == 0:
got_nomatch_errors = True
total_bytes += exp_bytes
total_objs += exp_objs
if total_objs and listing_style != ListingStyle.SHORT:
print('TOTAL: %d objects, %d bytes (%s)' %
(total_objs, total_bytes,
MakeHumanReadable(float(total_bytes))))
if got_nomatch_errors:
raise CommandException('One or more URLs matched no objects.')
if got_bucket_nomatch_errors:
raise NotFoundException(
'One or more bucket URLs matched no buckets.')
return 0
def RunCommand(self):
"""Command entry point for the rm command."""
# self.recursion_requested is initialized in command.py (so it can be
# checked in parent class for all commands).
self.continue_on_error = self.parallel_operations
self.read_args_from_stdin = False
self.all_versions = False
if self.sub_opts:
for o, unused_a in self.sub_opts:
if o == '-a':
self.all_versions = True
elif o == '-f':
self.continue_on_error = True
elif o == '-I':
self.read_args_from_stdin = True
elif o == '-r' or o == '-R':
self.recursion_requested = True
self.all_versions = True
if self.read_args_from_stdin:
if self.args:
raise CommandException(
'No arguments allowed with the -I flag.')
url_strs = StdinIterator()
else:
if not self.args:
raise CommandException(
'The rm command (without -I) expects at '
'least one URL.')
url_strs = self.args
# Tracks number of object deletes that failed.
self.op_failure_count = 0
# Tracks if any buckets were missing.
self.bucket_not_found_count = 0
# Tracks buckets that are slated for recursive deletion.
bucket_urls_to_delete = []
self.bucket_strings_to_delete = []
if self.recursion_requested:
bucket_fields = ['id']
for url_str in url_strs:
url = StorageUrlFromString(url_str)
if url.IsBucket() or url.IsProvider():
for blr in self.WildcardIterator(url_str).IterBuckets(
bucket_fields=bucket_fields):
bucket_urls_to_delete.append(blr.storage_url)
self.bucket_strings_to_delete.append(url_str)
self.preconditions = PreconditionsFromHeaders(self.headers or {})
try:
# Expand wildcards, dirs, buckets, and bucket subdirs in URLs.
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
url_strs,
self.recursion_requested,
project_id=self.project_id,
all_versions=self.all_versions,
continue_on_error=self.continue_on_error
or self.parallel_operations)
seek_ahead_iterator = None
# Cannot seek ahead with stdin args, since we can only iterate them
# once without buffering in memory.
if not self.read_args_from_stdin:
seek_ahead_iterator = SeekAheadNameExpansionIterator(
self.command_name,
self.debug,
self.GetSeekAheadGsutilApi(),
url_strs,
self.recursion_requested,
all_versions=self.all_versions,
project_id=self.project_id)
# Perform remove requests in parallel (-m) mode, if requested, using
# configured number of parallel processes and threads. Otherwise,
# perform requests with sequential function calls in current process.
self.Apply(
_RemoveFuncWrapper,
name_expansion_iterator,
_RemoveExceptionHandler,
fail_on_error=(not self.continue_on_error),
shared_attrs=['op_failure_count', 'bucket_not_found_count'],
seek_ahead_iterator=seek_ahead_iterator)
# Assuming the bucket has versioning enabled, url's that don't map to
# objects should throw an error even with all_versions, since the prior
# round of deletes only sends objects to a history table.
# This assumption that rm -a is only called for versioned buckets should be
# corrected, but the fix is non-trivial.
except CommandException as e:
# Don't raise if there are buckets to delete -- it's valid to say:
# gsutil rm -r gs://some_bucket
# if the bucket is empty.
if _ExceptionMatchesBucketToDelete(self.bucket_strings_to_delete,
e):
DecrementFailureCount()
else:
raise
except ServiceException as e:
if not self.continue_on_error:
raise
if self.bucket_not_found_count:
raise CommandException(
'Encountered non-existent bucket during listing')
if self.op_failure_count and not self.continue_on_error:
raise CommandException('Some files could not be removed.')
# If this was a gsutil rm -r command covering any bucket subdirs,
# remove any dir_$folder$ objects (which are created by various web UI
# tools to simulate folders).
if self.recursion_requested:
folder_object_wildcards = []
for url_str in url_strs:
url = StorageUrlFromString(url_str)
if url.IsObject():
folder_object_wildcards.append(
url_str.rstrip('*') + '*_$folder$')
if folder_object_wildcards:
self.continue_on_error = True
try:
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
folder_object_wildcards,
self.recursion_requested,
project_id=self.project_id,
all_versions=self.all_versions)
# When we're removing folder objects, always continue on error
self.Apply(_RemoveFuncWrapper,
name_expansion_iterator,
_RemoveFoldersExceptionHandler,
fail_on_error=False)
except CommandException as e:
# Ignore exception from name expansion due to an absent folder file.
if not e.reason.startswith(NO_URLS_MATCHED_PREFIX):
raise
# Now that all data has been deleted, delete any bucket URLs.
for url in bucket_urls_to_delete:
self.logger.info('Removing %s...', url)
@Retry(NotEmptyException, tries=3, timeout_secs=1)
def BucketDeleteWithRetry():
self.gsutil_api.DeleteBucket(url.bucket_name,
provider=url.scheme)
BucketDeleteWithRetry()
if self.op_failure_count:
plural_str = 's' if self.op_failure_count else ''
raise CommandException(
'%d file%s/object%s could not be removed.' %
(self.op_failure_count, plural_str, plural_str))
return 0
def _Create(self):
self.CheckArguments()
# User-specified options
pubsub_topic = None
payload_format = None
custom_attributes = {}
event_types = []
object_name_prefix = None
should_setup_topic = True
if self.sub_opts:
for o, a in self.sub_opts:
if o == '-e':
event_types.append(a)
elif o == '-f':
payload_format = a
elif o == '-m':
if ':' not in a:
raise CommandException(
'Custom attributes specified with -m should be of the form '
'key:value')
key, value = a.split(':')
custom_attributes[key] = value
elif o == '-p':
object_name_prefix = a
elif o == '-s':
should_setup_topic = False
elif o == '-t':
pubsub_topic = a
if payload_format not in PAYLOAD_FORMAT_MAP:
raise CommandException(
"Must provide a payload format with -f of either 'json' or 'none'"
)
payload_format = PAYLOAD_FORMAT_MAP[payload_format]
bucket_arg = self.args[-1]
bucket_url = StorageUrlFromString(bucket_arg)
if not bucket_url.IsCloudUrl() or not bucket_url.IsBucket():
raise CommandException(
"%s %s requires a GCS bucket name, but got '%s'" %
(self.command_name, self.subcommand_name, bucket_arg))
if bucket_url.scheme != 'gs':
raise CommandException(
'The %s command can only be used with gs:// bucket URLs.' %
self.command_name)
bucket_name = bucket_url.bucket_name
self.logger.debug('Creating notification for bucket %s', bucket_url)
# Find the project this bucket belongs to
bucket_metadata = self.gsutil_api.GetBucket(bucket_name,
fields=['projectNumber'],
provider=bucket_url.scheme)
bucket_project_number = bucket_metadata.projectNumber
# If not specified, choose a sensible default for the Cloud Pub/Sub topic
# name.
if not pubsub_topic:
pubsub_topic = 'projects/%s/topics/%s' % (PopulateProjectId(None),
bucket_name)
if not pubsub_topic.startswith('projects/'):
# If a user picks a topic ID (mytopic) but doesn't pass the whole name (
# projects/my-project/topics/mytopic ), pick a default project.
pubsub_topic = 'projects/%s/topics/%s' % (PopulateProjectId(None),
pubsub_topic)
self.logger.debug('Using Cloud Pub/Sub topic %s', pubsub_topic)
just_modified_topic_permissions = False
if should_setup_topic:
# Ask GCS for the email address that represents GCS's permission to
# publish to a Cloud Pub/Sub topic from this project.
service_account = self.gsutil_api.GetProjectServiceAccount(
bucket_project_number,
provider=bucket_url.scheme).email_address
self.logger.debug('Service account for project %d: %s',
bucket_project_number, service_account)
just_modified_topic_permissions = self._CreateTopic(
pubsub_topic, service_account)
for attempt_number in range(0, 2):
try:
create_response = self.gsutil_api.CreateNotificationConfig(
bucket_name,
pubsub_topic=pubsub_topic,
payload_format=payload_format,
custom_attributes=custom_attributes,
event_types=event_types if event_types else None,
object_name_prefix=object_name_prefix,
provider=bucket_url.scheme)
break
except PublishPermissionDeniedException:
if attempt_number == 0 and just_modified_topic_permissions:
# If we have just set the IAM policy, it may take up to 10 seconds to
# take effect.
self.logger.info(
'Retrying create notification in 10 seconds '
'(new permissions may take up to 10 seconds to take effect.)'
)
time.sleep(10)
else:
raise
notification_name = 'projects/_/buckets/%s/notificationConfigs/%s' % (
bucket_name, create_response.id)
self.logger.info('Created notification config %s', notification_name)
return 0
def RunCommand(self):
"""Command entry point for the mb command."""
bucket_policy_only = None
location = None
storage_class = None
seconds = None
public_access_prevention = None
if self.sub_opts:
for o, a in self.sub_opts:
if o == '-l':
location = a
elif o == '-p':
# Project IDs are sent as header values when using gs and s3 XML APIs.
InsistAscii(a, 'Invalid non-ASCII character found in project ID')
self.project_id = a
elif o == '-c' or o == '-s':
storage_class = NormalizeStorageClass(a)
elif o == '--retention':
seconds = RetentionInSeconds(a)
elif o == '-b':
if self.gsutil_api.GetApiSelector('gs') != ApiSelector.JSON:
raise CommandException('The -b <on|off> option '
'can only be used with the JSON API')
InsistOnOrOff(a, 'Only on and off values allowed for -b option')
bucket_policy_only = (a == 'on')
elif o == '--pap':
public_access_prevention = a
bucket_metadata = apitools_messages.Bucket(location=location,
storageClass=storage_class)
if bucket_policy_only or public_access_prevention:
bucket_metadata.iamConfiguration = IamConfigurationValue()
iam_config = bucket_metadata.iamConfiguration
if bucket_policy_only:
iam_config.bucketPolicyOnly = BucketPolicyOnlyValue()
iam_config.bucketPolicyOnly.enabled = bucket_policy_only
if public_access_prevention:
iam_config.publicAccessPrevention = public_access_prevention
for bucket_url_str in self.args:
bucket_url = StorageUrlFromString(bucket_url_str)
if seconds is not None:
if bucket_url.scheme != 'gs':
raise CommandException('Retention policy can only be specified for '
'GCS buckets.')
retention_policy = (apitools_messages.Bucket.RetentionPolicyValue(
retentionPeriod=seconds))
bucket_metadata.retentionPolicy = retention_policy
if public_access_prevention and self.gsutil_api.GetApiSelector(
bucket_url.scheme) != ApiSelector.JSON:
raise CommandException(
'The --pap option can only be used for GCS Buckets with the JSON API'
)
if not bucket_url.IsBucket():
raise CommandException('The mb command requires a URL that specifies a '
'bucket.\n"%s" is not valid.' % bucket_url)
if (not BUCKET_NAME_RE.match(bucket_url.bucket_name) or
TOO_LONG_DNS_NAME_COMP.search(bucket_url.bucket_name)):
raise InvalidUrlError('Invalid bucket name in URL "%s"' %
bucket_url.bucket_name)
self.logger.info('Creating %s...', bucket_url)
# Pass storage_class param only if this is a GCS bucket. (In S3 the
# storage class is specified on the key object.)
try:
self.gsutil_api.CreateBucket(bucket_url.bucket_name,
project_id=self.project_id,
metadata=bucket_metadata,
provider=bucket_url.scheme)
except BadRequestException as e:
if (e.status == 400 and e.reason == 'DotfulBucketNameNotUnderTld' and
bucket_url.scheme == 'gs'):
bucket_name = bucket_url.bucket_name
final_comp = bucket_name[bucket_name.rfind('.') + 1:]
raise CommandException('\n'.join(
textwrap.wrap(
'Buckets with "." in the name must be valid DNS names. The bucket'
' you are attempting to create (%s) is not a valid DNS name,'
' because the final component (%s) is not currently a valid part'
' of the top-level DNS tree.' % (bucket_name, final_comp))))
else:
raise
return 0
def _PatchIam(self):
self.continue_on_error = False
self.recursion_requested = False
patch_bindings_tuples = []
if self.sub_opts:
for o, a in self.sub_opts:
if o in ['-r', '-R']:
self.recursion_requested = True
elif o == '-f':
self.continue_on_error = True
elif o == '-d':
patch_bindings_tuples.append(BindingStringToTuple(
False, a))
patterns = []
# N.B.: self.sub_opts stops taking in options at the first non-flagged
# token. The rest of the tokens are sent to self.args. Thus, in order to
# handle input of the form "-d <binding> <binding> <url>", we will have to
# parse self.args for a mix of both bindings and CloudUrls. We are not
# expecting to come across the -r, -f flags here.
it = iter(self.args)
for token in it:
if token == '-d':
patch_bindings_tuples.append(
BindingStringToTuple(False, it.next()))
else:
try:
patch_bindings_tuples.append(
BindingStringToTuple(True, token))
# All following arguments are urls.
except (ArgumentException, CommandException):
patterns.append(token)
for token in it:
patterns.append(token)
# We must have some bindings to process, else this is pointless.
if not patch_bindings_tuples:
raise CommandException('Must specify at least one binding.')
self.everything_set_okay = True
threaded_wildcards = []
for pattern in patterns:
surl = StorageUrlFromString(pattern)
try:
if surl.IsBucket():
if self.recursion_requested:
surl.object = '*'
threaded_wildcards.append(surl.url_string)
else:
self.PatchIamHelper(surl, patch_bindings_tuples)
else:
threaded_wildcards.append(surl.url_string)
except AttributeError:
error_msg = 'Invalid Cloud URL "%s".' % surl.object_name
if set(surl.object_name).issubset(set('-Rrf')):
error_msg += (
' This resource handle looks like a flag, which must appear '
'before all bindings. See "gsutil help iam ch" for more details.'
)
raise CommandException(error_msg)
if threaded_wildcards:
name_expansion_iterator = NameExpansionIterator(
self.command_name,
self.debug,
self.logger,
self.gsutil_api,
threaded_wildcards,
self.recursion_requested,
all_versions=self.all_versions,
continue_on_error=self.continue_on_error
or self.parallel_operations,
bucket_listing_fields=['name'])
seek_ahead_iterator = SeekAheadNameExpansionIterator(
self.command_name,
self.debug,
self.GetSeekAheadGsutilApi(),
threaded_wildcards,
self.recursion_requested,
all_versions=self.all_versions)
# N.B.: Python2.6 support means we can't use a partial function here to
# curry the bindings tuples into the wrapper function. We instead pass
# the bindings along by zipping them with each name_expansion_iterator
# result. See http://bugs.python.org/issue5228.
serialized_bindings_tuples_it = itertools.repeat(
[SerializeBindingsTuple(t) for t in patch_bindings_tuples])
self.Apply(_PatchIamWrapper,
itertools.izip(serialized_bindings_tuples_it,
name_expansion_iterator),
_PatchIamExceptionHandler,
fail_on_error=not self.continue_on_error,
seek_ahead_iterator=seek_ahead_iterator)
self.everything_set_okay &= not GetFailureCount() > 0
# TODO: Add an error counter for files and objects.
if not self.everything_set_okay:
raise CommandException('Some IAM policies could not be patched.')
