def wipe_main(args_parser):
"""
wipe mainline
The mainline for the 'wipe' action.
Args:
args_parser: the argument parser to use for argument processing
Returns:
the exit code
"""
args_parser.add_argument('--danger', action='store_true')
known_args = sys.argv[1:]
args, unknown_args = args_parser.parse_known_args(known_args)
if unknown_args:
logger.warn('unknown arguments: {}'.format(' '.join(unknown_args)))
work_dir = args.work_dir if args.work_dir else os.getcwd()
# protection warning
if not args.danger:
print('')
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print("""
A request has been made to attempt to wipe the pages from a configured
Confluence instance. This is a helper utility call to assist a user in cleaning
out a space since removing a bulk set of data may not be trivial for a user.
Note that this action is not reversible with this tool and may require
assistance from an administrator from a Confluence instance to recover pages.
Only use this action if you know what you are doing.
To use this action, the argument '--danger' must be set.
""")
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print('')
return 1
# check configuration and prepare publisher
publisher = None
with TemporaryDirectory() as tmp_dir:
with docutils_namespace():
app = Sphinx(
work_dir, # document sources
work_dir, # directory with configuration
tmp_dir, # output for generated documents
tmp_dir, # output for doctree files
'confluence') # builder to execute
aggressive_search = app.config.confluence_adv_aggressive_search
server_url = app.config.confluence_server_url
space_name = app.config.confluence_space_name
# initialize the publisher (if permitted)
if app.config.confluence_publish:
process_ask_configs(app.config)
publisher = ConfluencePublisher()
publisher.init(app.config)
if not publisher:
print('(error) publishing not configured in sphinx configuration')
return 1
# reminder warning
print('')
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print("""
A request has been made to attempt to wipe the pages from a configured
Confluence instance. This action is not reversible with this tool and may
require assistance from an administrator from a Confluence instance to recover
pages. Only use this action if you know what you are doing.
""")
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print('')
if not ask_question('Are you sure you want to continue?'):
return 0
print('')
# user has confirmed; start an attempt to wipe
publisher.connect()
if aggressive_search:
legacy_pages = publisher.getDescendantsCompat(None)
else:
legacy_pages = publisher.getDescendants(None)
if not legacy_pages:
print('No pages are published on this space. Exiting...')
return 0
print(' URL:', server_url)
print(' Space:', space_name)
logger.note(' Pages: All Pages')
print(' Total pages:', len(legacy_pages))
print('')
if not ask_question('Are you sure you want to REMOVE these pages?'):
return 0
print('')
logger.info('Removing pages...', nonl=True)
for page_id in legacy_pages:
publisher.removePage(page_id)
logger.info('.', nonl=True)
logger.info(' done\n')
return 0
def report_main(args_parser):
"""
report mainline
The mainline for the 'report' action.
Args:
args_parser: the argument parser to use for argument processing
Returns:
the exit code
"""
args_parser.add_argument('--full-config', '-C', action='store_true')
args_parser.add_argument('--no-sanitize', action='store_true')
args_parser.add_argument('--offline', action='store_true')
known_args = sys.argv[1:]
args, unknown_args = args_parser.parse_known_args(known_args)
if unknown_args:
logger.warn('unknown arguments: {}'.format(' '.join(unknown_args)))
rv = 0
offline = args.offline
work_dir = args.work_dir if args.work_dir else os.getcwd()
# setup sphinx engine to extract configuration
config = {}
configuration_load_issue = None
confluence_instance_info = None
publisher = ConfluencePublisher()
try:
with temp_dir() as tmp_dir:
with docutils_namespace():
print('fetching configuration information...')
builder = ConfluenceReportBuilder.name
app = Sphinx(
work_dir, # document sources
work_dir, # directory with configuration
tmp_dir, # output for built documents
tmp_dir, # output for doctree files
builder, # builder to execute
status=sys.stdout, # sphinx status output
warning=sys.stderr) # sphinx warning output
if app.config.confluence_publish:
try:
process_ask_configs(app.config)
except ConfluenceConfigurationError:
offline = True
# extract configuration information
for k, v in app.config.values.items():
raw = getattr(app.config, k)
if raw is None:
continue
if callable(raw):
value = '(callable)'
else:
value = raw
prefixes = (
'confluence_',
'singleconfluence_',
)
if not args.full_config and not k.startswith(prefixes):
continue
# always extract some known builder configurations
if args.full_config and k.startswith(IGNORE_BUILDER_CONFS):
continue
config[k] = value
# initialize the publisher (if needed later)
publisher.init(app.config)
except Exception:
sys.stdout.flush()
tb_msg = traceback.format_exc()
logger.error(tb_msg)
if os.path.isfile(os.path.join(work_dir, 'conf.py')):
configuration_load_issue = 'unable to load configuration'
configuration_load_issue += '\n\n' + tb_msg.strip()
else:
configuration_load_issue = 'no documentation/missing configuration'
rv = 1
# attempt to fetch confluence instance version
confluence_publish = config.get('confluence_publish')
confluence_server_url = config.get('confluence_server_url')
if not offline and confluence_publish and confluence_server_url:
base_url = ConfluenceUtil.normalize_base_url(confluence_server_url)
info = ''
session = None
try:
print('connecting to confluence instance...')
sys.stdout.flush()
publisher.connect()
info += ' connected: yes\n'
session = publisher.rest_client.session
except Exception:
sys.stdout.flush()
logger.error(traceback.format_exc())
info += ' connected: no\n'
rv = 1
if session:
try:
# fetch
print('fetching confluence instance information...')
manifest_url = base_url + MANIFEST_PATH
rsp = session.get(manifest_url)
if rsp.status_code == 200:
info += ' fetched: yes\n'
# extract
print('decoding information...')
rsp.encoding = 'utf-8'
raw_data = rsp.text
info += ' decoded: yes\n'
# parse
print('parsing information...')
xml_data = ElementTree.fromstring(raw_data)
info += ' parsed: yes\n'
root = ElementTree.ElementTree(xml_data)
for o in root.findall('typeId'):
info += ' type: ' + o.text + '\n'
for o in root.findall('version'):
info += ' version: ' + o.text + '\n'
for o in root.findall('buildNumber'):
info += ' build: ' + o.text + '\n'
else:
logger.error('bad response from server ({})'.format(
rsp.status_code))
info += ' fetched: error ({})\n'.format(rsp.status_code)
rv = 1
except Exception:
sys.stdout.flush()
logger.error(traceback.format_exc())
info += 'failure to determine confluence data\n'
rv = 1
confluence_instance_info = info
def sensitive_config(key):
if key in config:
if config[key]:
config[key] = '(set)'
else:
config[key] = '(set; empty)'
# always sanitize out sensitive information
sensitive_config('confluence_client_cert_pass')
sensitive_config('confluence_publish_headers')
sensitive_config('confluence_publish_token')
sensitive_config('confluence_server_pass')
# optional sanitization
if not args.no_sanitize:
sensitive_config('author')
sensitive_config('confluence_client_cert')
sensitive_config('confluence_global_labels')
sensitive_config('confluence_jira_servers')
sensitive_config('confluence_mentions')
sensitive_config('confluence_parent_page')
sensitive_config('confluence_parent_page_id_check')
sensitive_config('confluence_proxy')
sensitive_config('confluence_publish_root')
sensitive_config('confluence_server_auth')
sensitive_config('confluence_server_cookies')
sensitive_config('confluence_server_user')
sensitive_config('project')
# remove confluence instance (attempt to keep scheme)
if 'confluence_server_url' in config:
value = config['confluence_server_url']
parsed = urlparse(value)
if parsed.scheme:
value = parsed.scheme + '://<removed>'
else:
value = '(set; no scheme)'
if parsed.netloc and parsed.netloc.endswith('atlassian.net'):
value += ' (cloud)'
config['confluence_server_url'] = value
# remove space key, but track casing
space_cfgs = [
'confluence_space_key',
'confluence_space_name', # deprecated
]
for space_cfg in space_cfgs:
if space_cfg not in config:
continue
value = config[space_cfg]
if value.startswith('~'):
value = '(set; user)'
elif value.isupper():
value = '(set; upper)'
elif value.islower():
value = '(set; lower)'
else:
value = '(set; mixed)'
config[space_cfg] = value
print('')
print('Confluence builder report has been generated.')
print('Please copy the following text for the GitHub issue:')
print('')
logger.note('------------[ cut here ]------------')
print('```')
print('(system)')
print(' platform:', single_line_version(platform.platform()))
print(' python:', single_line_version(sys.version))
print(' sphinx:', single_line_version(sphinx_version))
print(' requests:', single_line_version(requests_version))
print(' builder:', single_line_version(scb_version))
print('')
print('(configuration)')
if config:
for k, v in OrderedDict(sorted(config.items())).items():
print('{}: {}'.format(k, v))
else:
print('~default configuration~')
if configuration_load_issue:
print('')
print('(error loading configuration)')
print(configuration_load_issue)
if confluence_instance_info:
print('')
print('(confluence instance)')
print(confluence_instance_info.rstrip())
print('```')
logger.note('------------[ cut here ]------------')
return rv
def wipe_main(args_parser):
"""
wipe mainline
The mainline for the 'wipe' action.
Args:
args_parser: the argument parser to use for argument processing
Returns:
the exit code
"""
args_parser.add_argument('--danger', action='store_true')
args_parser.add_argument('--parent', '-P', action='store_true')
known_args = sys.argv[1:]
args, unknown_args = args_parser.parse_known_args(known_args)
if unknown_args:
logger.warn('unknown arguments: {}'.format(' '.join(unknown_args)))
work_dir = args.work_dir if args.work_dir else os.getcwd()
# protection warning
if not args.danger:
print('')
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print("""
A request has been made to attempt to wipe the pages from a configured
Confluence instance. This is a helper utility call to assist a user in cleaning
out a space since removing a bulk set of data may not be trivial for a user.
Note that this action is not reversible with this tool and may require
assistance from an administrator from a Confluence instance to recover pages.
Only use this action if you know what you are doing.
To use this action, the argument '--danger' must be set.
""")
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
return 1
# check configuration and prepare publisher
dryrun = False
publisher = None
try:
with temp_dir() as tmp_dir:
with docutils_namespace():
app = Sphinx(
work_dir, # document sources
work_dir, # directory with configuration
tmp_dir, # output for built documents
tmp_dir, # output for doctree files
'confluence', # builder to execute
status=sys.stdout, # sphinx status output
warning=sys.stderr) # sphinx warning output
aggressive_search = app.config.confluence_adv_aggressive_search
dryrun = app.config.confluence_publish_dryrun
server_url = app.config.confluence_server_url
space_key = app.config.confluence_space_key
parent_ref = app.config.confluence_parent_page
# initialize the publisher (if permitted)
if app.config.confluence_publish:
process_ask_configs(app.config)
publisher = ConfluencePublisher()
publisher.init(app.config)
except Exception:
sys.stdout.flush()
logger.error(traceback.format_exc())
if os.path.isfile(os.path.join(work_dir, 'conf.py')):
logger.error('unable to load configuration')
else:
logger.error('no documentation/missing configuration')
return 1
if not publisher:
logger.error('publishing not configured in sphinx configuration')
return 1
if args.parent and not parent_ref:
logger.error('parent option provided but no parent page is configured')
return 1
# reminder warning
print('')
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print("""
A request has been made to attempt to wipe the pages from a configured
Confluence instance. This action is not reversible with this tool and may
require assistance from an administrator from a Confluence instance to recover
pages. Only use this action if you know what you are doing.
""")
sys.stdout.flush()
logger.warn('!!! DANGER DANGER DANGER !!!')
print('')
if not ask_question('Are you sure you want to continue?'):
return 0
print('')
# user has confirmed; start an attempt to wipe
publisher.connect()
base_page_id = None
if args.parent:
base_page_id = publisher.get_base_page_id()
if aggressive_search:
legacy_pages = publisher.get_descendants_compat(base_page_id)
else:
legacy_pages = publisher.get_descendants(base_page_id)
print(' URL:', server_url)
print(' Space:', space_key)
if base_page_id:
logger.note(' Pages: Child pages of ' + parent_ref)
else:
logger.note(' Pages: All Pages')
print(' Total pages:', len(legacy_pages))
if dryrun:
print(' Dry run:', 'Enabled (no pages will be removed)')
if not legacy_pages:
print('')
print('No pages detected on this space. Exiting...')
return 0
if args.verbose:
print('-------------------------')
page_names = []
for p in legacy_pages:
page_names.append(publisher._name_cache[p])
sorted(page_names)
print('\n'.join(page_names))
print('-------------------------')
print('')
if not ask_question('Are you sure you want to REMOVE these pages?'):
return 0
print('')
logger.info('Removing pages...', nonl=True)
if dryrun:
logger.info('')
for page_id in legacy_pages:
publisher.remove_page(page_id)
if not dryrun:
logger.info('.', nonl=True)
if not dryrun:
logger.info(__('done'))
return 0