Encoding fixes to support py2 and py3 non-ascii paths (#58414)

* Encoding fixes to support py2 and py3 non-ascii paths

* Remove unused import

* endswith instead of comparing slice on bytes

* join bytes, convert to to_native after

Co-Authored-By: Toshio Kuratomi <a.badger@gmail.com>

* Fix review comments

* Add missing comma

* Encoding fixes to support py2 and py3 non-ascii paths

* Use ascii encoding on paths added to the archive also
pull/58446/head
Matt Martz 5 years ago committed by Matt Clay
parent 1a9b1d0edd
commit 11e2ac3abf

@ -256,12 +256,13 @@ class CertificateSet(object):
'''
Load lists of PEM certificates from a file or a directory.
'''
if os.path.isdir(path):
for dir, dummy, files in os.walk(path, followlinks=True):
b_path = to_bytes(path, errors='surrogate_or_strict')
if os.path.isdir(b_path):
for dir, dummy, files in os.walk(b_path, followlinks=True):
for file in files:
self._load_file(os.path.join(dir, file))
else:
self._load_file(path)
self._load_file(b_path)
def find_parent(self, cert):
'''

@ -172,7 +172,7 @@ import zipfile
from traceback import format_exc
from ansible.module_utils.basic import AnsibleModule, missing_required_lib
from ansible.module_utils._text import to_native
from ansible.module_utils._text import to_bytes, to_native
from ansible.module_utils.six import PY3
@ -211,12 +211,14 @@ def main():
check_mode = module.check_mode
paths = params['path']
dest = params['dest']
b_dest = None if not dest else to_bytes(dest, errors='surrogate_or_strict')
exclude_paths = params['exclude_path']
remove = params['remove']
expanded_paths = []
expanded_exclude_paths = []
format = params['format']
b_expanded_paths = []
b_expanded_exclude_paths = []
fmt = params['format']
b_fmt = to_bytes(fmt, errors='surrogate_or_strict')
force_archive = params['force_archive']
globby = False
changed = False
@ -224,111 +226,128 @@ def main():
# Simple or archive file compression (inapplicable with 'zip' since it's always an archive)
archive = False
successes = []
b_successes = []
# Fail early
if not HAS_LZMA and format == 'xz':
if not HAS_LZMA and fmt == 'xz':
module.fail_json(msg=missing_required_lib("lzma or backports.lzma", reason="when using xz format"),
exception=LZMA_IMP_ERR)
module.fail_json(msg="lzma or backports.lzma is required when using xz format.")
for path in paths:
path = os.path.expanduser(os.path.expandvars(path))
b_path = os.path.expanduser(
os.path.expandvars(
to_bytes(path, errors='surrogate_or_strict')
)
)
# Expand any glob characters. If found, add the expanded glob to the
# list of expanded_paths, which might be empty.
if ('*' in path or '?' in path):
expanded_paths = expanded_paths + glob.glob(path)
if (b'*' in b_path or b'?' in b_path):
b_expanded_paths.extend(glob.glob(b_path))
globby = True
# If there are no glob characters the path is added to the expanded paths
# whether the path exists or not
else:
expanded_paths.append(path)
b_expanded_paths.append(b_path)
# Only attempt to expand the exclude paths if it exists
if exclude_paths:
for exclude_path in exclude_paths:
exclude_path = os.path.expanduser(os.path.expandvars(exclude_path))
b_exclude_path = os.path.expanduser(
os.path.expandvars(
to_bytes(exclude_path, errors='surrogate_or_strict')
)
)
# Expand any glob characters. If found, add the expanded glob to the
# list of expanded_paths, which might be empty.
if ('*' in exclude_path or '?' in exclude_path):
expanded_exclude_paths = expanded_exclude_paths + glob.glob(exclude_path)
if (b'*' in b_exclude_path or b'?' in b_exclude_path):
b_expanded_exclude_paths.extend(glob.glob(b_exclude_path))
# If there are no glob character the exclude path is added to the expanded
# exclude paths whether the path exists or not.
else:
expanded_exclude_paths.append(exclude_path)
b_expanded_exclude_paths.append(b_exclude_path)
if not expanded_paths:
return module.fail_json(path=', '.join(paths), expanded_paths=', '.join(expanded_paths), msg='Error, no source paths were found')
if not b_expanded_paths:
return module.fail_json(
path=', '.join(paths),
expanded_paths=to_native(b', '.join(b_expanded_paths), errors='surrogate_or_strict'),
msg='Error, no source paths were found'
)
# Only try to determine if we are working with an archive or not if we haven't set archive to true
if not force_archive:
# If we actually matched multiple files or TRIED to, then
# treat this as a multi-file archive
archive = globby or os.path.isdir(expanded_paths[0]) or len(expanded_paths) > 1
archive = globby or os.path.isdir(b_expanded_paths[0]) or len(b_expanded_paths) > 1
else:
archive = True
# Default created file name (for single-file archives) to
# <file>.<format>
if not dest and not archive:
dest = '%s.%s' % (expanded_paths[0], format)
if not b_dest and not archive:
b_dest = b'%s.%s' % (b_expanded_paths[0], b_fmt)
# Force archives to specify 'dest'
if archive and not dest:
if archive and not b_dest:
module.fail_json(dest=dest, path=', '.join(paths), msg='Error, must specify "dest" when archiving multiple files or trees')
archive_paths = []
missing = []
arcroot = ''
b_sep = to_bytes(os.sep, errors='surrogate_or_strict')
for path in expanded_paths:
b_archive_paths = []
b_missing = []
b_arcroot = b''
for b_path in b_expanded_paths:
# Use the longest common directory name among all the files
# as the archive root path
if arcroot == '':
arcroot = os.path.dirname(path) + os.sep
if b_arcroot == b'':
b_arcroot = os.path.dirname(b_path) + b_sep
else:
for i in range(len(arcroot)):
if path[i] != arcroot[i]:
for i in range(len(b_arcroot)):
if b_path[i] != b_arcroot[i]:
break
if i < len(arcroot):
arcroot = os.path.dirname(arcroot[0:i + 1])
if i < len(b_arcroot):
b_arcroot = os.path.dirname(b_arcroot[0:i + 1])
arcroot += os.sep
b_arcroot += b_sep
# Don't allow archives to be created anywhere within paths to be removed
if remove and os.path.isdir(path):
path_dir = path
if path[-1] != '/':
path_dir += '/'
if dest.startswith(path_dir):
module.fail_json(path=', '.join(paths), msg='Error, created archive can not be contained in source paths when remove=True')
if os.path.lexists(path) and path not in expanded_exclude_paths:
archive_paths.append(path)
if remove and os.path.isdir(b_path):
b_path_dir = b_path
if not b_path.endswith(b'/'):
b_path_dir += b'/'
if b_dest.startswith(b_path_dir):
module.fail_json(
path=', '.join(paths),
msg='Error, created archive can not be contained in source paths when remove=True'
)
if os.path.lexists(b_path) and b_path not in b_expanded_exclude_paths:
b_archive_paths.append(b_path)
else:
missing.append(path)
b_missing.append(b_path)
# No source files were found but the named archive exists: are we 'compress' or 'archive' now?
if len(missing) == len(expanded_paths) and dest and os.path.exists(dest):
if len(b_missing) == len(b_expanded_paths) and b_dest and os.path.exists(b_dest):
# Just check the filename to know if it's an archive or simple compressed file
if re.search(r'(\.tar|\.tar\.gz|\.tgz|\.tbz2|\.tar\.bz2|\.tar\.xz|\.zip)$', os.path.basename(dest), re.IGNORECASE):
if re.search(br'(\.tar|\.tar\.gz|\.tgz|\.tbz2|\.tar\.bz2|\.tar\.xz|\.zip)$', os.path.basename(b_dest), re.IGNORECASE):
state = 'archive'
else:
state = 'compress'
# Multiple files, or globbiness
elif archive:
if not archive_paths:
if not b_archive_paths:
# No source files were found, but the archive is there.
if os.path.lexists(dest):
if os.path.lexists(b_dest):
state = 'archive'
elif missing:
elif b_missing:
# SOME source files were found, but not all of them
state = 'incomplete'
@ -336,8 +355,8 @@ def main():
size = 0
errors = []
if os.path.lexists(dest):
size = os.path.getsize(dest)
if os.path.lexists(b_dest):
size = os.path.getsize(b_dest)
if state != 'archive':
if check_mode:
@ -346,76 +365,88 @@ def main():
else:
try:
# Slightly more difficult (and less efficient!) compression using zipfile module
if format == 'zip':
arcfile = zipfile.ZipFile(dest, 'w', zipfile.ZIP_DEFLATED, True)
if fmt == 'zip':
arcfile = zipfile.ZipFile(
to_native(b_dest, errors='surrogate_or_strict', encoding='ascii'),
'w',
zipfile.ZIP_DEFLATED,
True
)
# Easier compression using tarfile module
elif format == 'gz' or format == 'bz2':
arcfile = tarfile.open(dest, 'w|' + format)
elif fmt == 'gz' or fmt == 'bz2':
arcfile = tarfile.open(to_native(b_dest, errors='surrogate_or_strict', encoding='ascii'), 'w|' + fmt)
# python3 tarfile module allows xz format but for python2 we have to create the tarfile
# in memory and then compress it with lzma.
elif format == 'xz':
elif fmt == 'xz':
arcfileIO = io.BytesIO()
arcfile = tarfile.open(fileobj=arcfileIO, mode='w')
# Or plain tar archiving
elif format == 'tar':
arcfile = tarfile.open(dest, 'w')
elif fmt == 'tar':
arcfile = tarfile.open(to_native(b_dest, errors='surrogate_or_strict', encoding='ascii'), 'w')
match_root = re.compile('^%s' % re.escape(arcroot))
for path in archive_paths:
if os.path.isdir(path):
b_match_root = re.compile(br'^%s' % re.escape(b_arcroot))
for b_path in b_archive_paths:
if os.path.isdir(b_path):
# Recurse into directories
for dirpath, dirnames, filenames in os.walk(path, topdown=True):
if not dirpath.endswith(os.sep):
dirpath += os.sep
for b_dirpath, b_dirnames, b_filenames in os.walk(b_path, topdown=True):
if not b_dirpath.endswith(b_sep):
b_dirpath += b_sep
for dirname in dirnames:
fullpath = dirpath + dirname
arcname = match_root.sub('', fullpath)
for b_dirname in b_dirnames:
b_fullpath = b_dirpath + b_dirname
n_fullpath = to_native(b_fullpath, errors='surrogate_or_strict', encoding='ascii')
n_arcname = to_native(b_match_root.sub(b'', b_fullpath), errors='surrogate_or_strict')
try:
if format == 'zip':
arcfile.write(fullpath, arcname)
if fmt == 'zip':
arcfile.write(n_fullpath, n_arcname)
else:
arcfile.add(fullpath, arcname, recursive=False)
arcfile.add(n_fullpath, n_arcname, recursive=False)
except Exception as e:
errors.append('%s: %s' % (fullpath, to_native(e)))
errors.append('%s: %s' % (n_fullpath, to_native(e)))
for filename in filenames:
fullpath = dirpath + filename
arcname = match_root.sub('', fullpath)
for b_filename in b_filenames:
b_fullpath = b_dirpath + b_filename
n_fullpath = to_native(b_fullpath, errors='surrogate_or_strict', encoding='ascii')
n_arcname = to_native(b_match_root.sub(b'', b_fullpath), errors='surrogate_or_strict')
if not filecmp.cmp(fullpath, dest):
if not filecmp.cmp(b_fullpath, b_dest):
try:
if format == 'zip':
arcfile.write(fullpath, arcname)
if fmt == 'zip':
arcfile.write(n_fullpath, n_arcname)
else:
arcfile.add(fullpath, arcname, recursive=False)
arcfile.add(n_fullpath, n_arcname, recursive=False)
successes.append(fullpath)
b_successes.append(b_fullpath)
except Exception as e:
errors.append('Adding %s: %s' % (path, to_native(e)))
errors.append('Adding %s: %s' % (to_native(b_path), to_native(e)))
else:
if format == 'zip':
arcfile.write(path, match_root.sub('', path))
path = to_native(b_path, errors='surrogate_or_strict', encoding='ascii')
arcname = to_native(b_match_root.sub(b'', b_path), errors='surrogate_or_strict')
if fmt == 'zip':
arcfile.write(path, arcname)
else:
arcfile.add(path, match_root.sub('', path), recursive=False)
arcfile.add(path, arcname, recursive=False)
successes.append(path)
b_successes.append(b_path)
except Exception as e:
module.fail_json(msg='Error when writing %s archive at %s: %s' % (format == 'zip' and 'zip' or ('tar.' + format), dest, to_native(e)),
exception=format_exc())
expanded_fmt = 'zip' if fmt == 'zip' else ('tar.' + fmt)
module.fail_json(
msg='Error when writing %s archive at %s: %s' % (expanded_fmt, dest, to_native(e)),
exception=format_exc()
)
if arcfile:
arcfile.close()
state = 'archive'
if format == 'xz':
with lzma.open(dest, 'wb') as f:
if fmt == 'xz':
with lzma.open(b_dest, 'wb') as f:
f.write(arcfileIO.getvalue())
arcfileIO.close()
@ -423,76 +454,89 @@ def main():
module.fail_json(msg='Errors when writing archive at %s: %s' % (dest, '; '.join(errors)))
if state in ['archive', 'incomplete'] and remove:
for path in successes:
for b_path in b_successes:
try:
if os.path.isdir(path):
shutil.rmtree(path)
if os.path.isdir(b_path):
shutil.rmtree(b_path)
elif not check_mode:
os.remove(path)
os.remove(b_path)
except OSError as e:
errors.append(path)
errors.append(to_native(b_path))
if errors:
module.fail_json(dest=dest, msg='Error deleting some source files: ', files=errors)
# Rudimentary check: If size changed then file changed. Not perfect, but easy.
if not check_mode and os.path.getsize(dest) != size:
if not check_mode and os.path.getsize(b_dest) != size:
changed = True
if successes and state != 'incomplete':
if b_successes and state != 'incomplete':
state = 'archive'
# Simple, single-file compression
else:
path = expanded_paths[0]
b_path = b_expanded_paths[0]
# No source or compressed file
if not (os.path.exists(path) or os.path.lexists(dest)):
if not (os.path.exists(b_path) or os.path.lexists(b_dest)):
state = 'absent'
# if it already exists and the source file isn't there, consider this done
elif not os.path.lexists(path) and os.path.lexists(dest):
elif not os.path.lexists(b_path) and os.path.lexists(b_dest):
state = 'compress'
else:
if module.check_mode:
if not os.path.exists(dest):
if not os.path.exists(b_dest):
changed = True
else:
size = 0
f_in = f_out = arcfile = None
if os.path.lexists(dest):
size = os.path.getsize(dest)
if os.path.lexists(b_dest):
size = os.path.getsize(b_dest)
try:
if format == 'zip':
arcfile = zipfile.ZipFile(dest, 'w', zipfile.ZIP_DEFLATED, True)
arcfile.write(path, path[len(arcroot):])
if fmt == 'zip':
arcfile = zipfile.ZipFile(
to_native(b_dest, errors='surrogate_or_strict', encoding='ascii'),
'w',
zipfile.ZIP_DEFLATED,
True
)
arcfile.write(
to_native(b_path, errors='surrogate_or_strict', encoding='ascii'),
to_native(b_path[len(b_arcroot):], errors='surrogate_or_strict')
)
arcfile.close()
state = 'archive' # because all zip files are archives
elif format == 'tar':
arcfile = tarfile.open(dest, 'w')
arcfile.add(path)
elif fmt == 'tar':
arcfile = tarfile.open(to_native(b_dest, errors='surrogate_or_strict', encoding='ascii'), 'w')
arcfile.add(to_native(b_path, errors='surrogate_or_strict', encoding='ascii'))
arcfile.close()
else:
f_in = open(path, 'rb')
if format == 'gz':
f_out = gzip.open(dest, 'wb')
elif format == 'bz2':
f_out = bz2.BZ2File(dest, 'wb')
elif format == 'xz':
f_out = lzma.LZMAFile(dest, 'wb')
f_in = open(b_path, 'rb')
n_dest = to_native(b_dest, errors='surrogate_or_strict', encoding='ascii')
if fmt == 'gz':
f_out = gzip.open(n_dest, 'wb')
elif fmt == 'bz2':
f_out = bz2.BZ2File(n_dest, 'wb')
elif fmt == 'xz':
f_out = lzma.LZMAFile(n_dest, 'wb')
else:
raise OSError("Invalid format")
shutil.copyfileobj(f_in, f_out)
successes.append(path)
b_successes.append(b_path)
except OSError as e:
module.fail_json(path=path, dest=dest, msg='Unable to write to compressed file: %s' % to_native(e), exception=format_exc())
module.fail_json(
path=to_native(b_path),
dest=dest,
msg='Unable to write to compressed file: %s' % to_native(e), exception=format_exc()
)
if arcfile:
arcfile.close()
@ -502,32 +546,37 @@ def main():
f_out.close()
# Rudimentary check: If size changed then file changed. Not perfect, but easy.
if os.path.getsize(dest) != size:
if os.path.getsize(b_dest) != size:
changed = True
state = 'compress'
if remove and not check_mode:
try:
os.remove(path)
os.remove(b_path)
except OSError as e:
module.fail_json(path=path, msg='Unable to remove source file: %s' % to_native(e), exception=format_exc())
module.fail_json(
path=to_native(b_path),
msg='Unable to remove source file: %s' % to_native(e), exception=format_exc()
)
params['path'] = dest
params['path'] = b_dest
file_args = module.load_file_common_arguments(params)
if not check_mode:
changed = module.set_fs_attributes_if_different(file_args, changed)
module.exit_json(archived=successes,
dest=dest,
changed=changed,
state=state,
arcroot=arcroot,
missing=missing,
expanded_paths=expanded_paths,
expanded_exclude_paths=expanded_exclude_paths)
module.exit_json(
archived=[to_native(p, errors='surrogate_or_strict') for p in b_successes],
dest=dest,
changed=changed,
state=state,
arcroot=to_native(b_arcroot, errors='surrogate_or_strict'),
missing=[to_native(p, errors='surrogate_or_strict') for p in b_missing],
expanded_paths=[to_native(p, errors='surrogate_or_strict') for p in b_expanded_paths],
expanded_exclude_paths=[to_native(p, errors='surrogate_or_strict') for p in b_expanded_exclude_paths],
)
if __name__ == '__main__':

Loading…
Cancel
Save