def getIndex(config={}):
""" Returns Index.php for a wiki, if available """
index = ''
if config['wikiengine'] == 'mediawiki':
import mediawiki
index = mediawiki.mwGetIndex(config=config)
return index
def getIndex(config={}):
""" Returns Index.php for a wiki, if available """
index = ''
if config['wikiengine'] == 'mediawiki':
import mediawiki
index = mediawiki.mwGetIndex(config=config)
return index
def getParameters(params=[]):
""" Import parameters into variable """
if not params:
params = sys.argv
config = {}
parser = argparse.ArgumentParser(
description='Tools for downloading and preserving wikis.')
# General params
parser.add_argument('-v',
'--version',
action='version',
version=getVersion())
parser.add_argument('--cookies',
metavar="cookies.txt",
help="Path to a cookies.txt file.")
parser.add_argument('--delay',
metavar=5,
default=0,
type=float,
help="Adds a delay (in seconds).")
parser.add_argument('--retries',
metavar=5,
default=5,
help="Maximum number of retries.")
parser.add_argument('--path', help='Path to store wiki dump at.')
parser.add_argument(
'--resume',
action='store_true',
help='Resumes previous incomplete dump (requires --path).')
parser.add_argument('--force', action='store_true', help='')
parser.add_argument('--user',
help='Username if authentication is required.')
parser.add_argument('--pass',
dest='password',
help='Password if authentication is required.')
# URL params
# This script should work with any general URL, finding out
# API, index.php or whatever by itself when necessary
groupWiki = parser.add_argument_group()
groupWiki.add_argument('wiki',
default='',
nargs='?',
help="URL to wiki (e.g. http://wiki.domain.org).")
# URL params for MediaWiki
groupWiki.add_argument(
'--mwapi',
help="URL to MediaWiki API (e.g. http://wiki.domain.org/w/api.php).")
groupWiki.add_argument(
'--mwindex',
help=
"URL to MediaWiki index.php (e.g. http://wiki.domain.org/w/index.php)."
)
# Download params
groupDownload = parser.add_argument_group(
'Data to download', 'What info download from the wiki')
groupDownload.add_argument(
'--pages',
action='store_true',
help=
"Generates a dump of pages (--pages --curonly for current revisions only)."
)
groupDownload.add_argument('--curonly',
action='store_true',
help='Store only the current version of pages.')
groupDownload.add_argument('--images',
action='store_true',
help="Generates an image dump.")
groupDownload.add_argument(
'--namespaces',
metavar="1,2,3",
help='Comma-separated value of namespaces to include (all by default).'
)
groupDownload.add_argument(
'--exnamespaces',
metavar="1,2,3",
help='Comma-separated value of namespaces to exclude.')
# Meta info params
groupMeta = parser.add_argument_group(
'Meta info', 'What meta info to retrieve from the wiki')
groupMeta.add_argument('--get-api',
action='store_true',
help="Returns wiki API when available.")
groupMeta.add_argument('--get-index',
action='store_true',
help="Returns wiki Index.php when available.")
groupMeta.add_argument('--get-page-titles',
action='store_true',
help="Returns wiki page titles.")
groupMeta.add_argument('--get-image-names',
action='store_true',
help="Returns wiki image names.")
groupMeta.add_argument('--get-namespaces',
action='store_true',
help="Returns wiki namespaces.")
groupMeta.add_argument('--get-wiki-engine',
action='store_true',
help="Returns wiki engine.")
args = parser.parse_args()
#sys.stderr.write(args)
# Not wiki? Exit
if not args.wiki:
sys.stderr.write('ERROR: Provide a URL to a wiki\n')
parser.print_help()
sys.exit(1)
# Don't mix download params and meta info params
if (args.pages or args.images) and \
(args.get_api or args.get_index or args.get_page_titles or args.get_image_names or args.get_namespaces or args.get_wiki_engine):
sys.stderr.write(
'ERROR: Don\'t mix download params and meta info params\n')
parser.print_help()
sys.exit(1)
# No download params and no meta info params? Exit
if (not args.pages and not args.images) and \
(not args.get_api and not args.get_index and not args.get_page_titles and not args.get_image_names and not args.get_namespaces and not args.get_wiki_engine):
sys.stderr.write(
'ERROR: Use at least one download param or meta info param\n')
parser.print_help()
sys.exit(1)
# Load cookies
cj = cookielib.MozillaCookieJar()
if args.cookies:
cj.load(args.cookies)
sys.stderr.write('Using cookies from %s\n' % args.cookies)
# check user and pass (one requires both)
if (args.user and not args.password) or (args.password and not args.user):
sys.stderr.write(
'ERROR: Both --user and --pass are required for authentication.\n')
parser.print_help()
sys.exit(1)
session = None
if args.user and args.password:
import requests
session = requests.Session()
session.cookies = cj
session.headers.update({'User-Agent': getUserAgent()})
session.auth = (args.user, args.password)
#session.mount(args.mw_api.split('/api.php')[0], HTTPAdapter(max_retries=max_ret)) Mediawiki-centric, be careful
# check URLs
for url in [args.mwapi, args.mwindex, args.wiki]:
if url and (not url.startswith('http://')
and not url.startswith('https://')):
sys.stderr.write(url)
sys.stderr.write(
'ERROR: URLs must start with http:// or https://\n')
parser.print_help()
sys.exit(1)
# Meta info params
metainfo = '' # only one allowed, so we don't mix output
if args.get_api:
metainfo = 'get_api'
elif args.get_index:
metainfo = 'get_index'
elif args.get_page_titles:
metainfo = 'get_page_titles'
elif args.get_image_names:
metainfo = 'get_image_names'
elif args.get_namespaces:
metainfo = 'get_namespaces'
elif args.get_wiki_engine:
metainfo = 'get_wiki_engine'
namespaces = ['all']
exnamespaces = []
# Process namespace inclusions
if args.namespaces:
# fix, why - ? and... --namespaces= all with a space works?
if re.search(r'[^\d, \-]',
args.namespaces) and args.namespaces.lower() != 'all':
sys.stderr.write(
"Invalid namespace values.\nValid format is integer(s) separated by commas\n"
)
sys.exit()
else:
ns = re.sub(' ', '', args.namespaces)
if ns.lower() == 'all':
namespaces = ['all']
else:
namespaces = [int(i) for i in ns.split(',')]
# Process namespace exclusions
if args.exnamespaces:
if re.search(r'[^\d, \-]', args.exnamespaces):
sys.stderr.write(
"Invalid namespace values.\nValid format is integer(s) separated by commas\n"
)
sys.exit(1)
else:
ns = re.sub(' ', '', args.exnamespaces)
if ns.lower() == 'all':
sys.stderr.write('You cannot exclude all namespaces.\n')
sys.exit(1)
else:
exnamespaces = [int(i) for i in ns.split(',')]
# --curonly requires --xml
if args.curonly and not args.pages:
sys.stderr.write("--curonly requires --pages\n")
parser.print_help()
sys.exit(1)
config = {
'cookies': args.cookies or '',
'curonly': args.curonly,
'date': datetime.datetime.now().strftime('%Y%m%d'),
'delay': args.delay,
'exnamespaces': exnamespaces,
'images': args.images,
'logs': False,
'metainfo': metainfo,
'namespaces': namespaces,
'pages': args.pages,
'path': args.path and os.path.normpath(args.path) or '',
'retries': int(args.retries),
'wiki': args.wiki,
'wikicanonical': '',
'wikiengine': getWikiEngine(args.wiki),
'other': {
'configfilename': 'config.txt',
'filenamelimit': 100, # do not change
'force': args.force,
'resume': args.resume,
'session': session,
}
}
# Get ready special variables (API for MediWiki, etc)
if config['wikiengine'] == 'mediawiki':
import mediawiki
config['mwexport'] = 'Special:Export'
if not args.mwapi:
config['mwapi'] = mediawiki.mwGetAPI(config=config)
if not config['mwapi']:
sys.stderr.write('ERROR: Provide a URL to API\n')
sys.exit(1)
else:
data = {
'action': 'query',
'meta': 'siteinfo',
'siprop': 'namespaces',
'format': 'json'
}
r = getURL(config['mwapi'], data=data)
config['mwexport'] = getJSON(r)['query']['namespaces']['-1']['*'] \
+ ':Export'
if not args.mwindex:
config['mwindex'] = mediawiki.mwGetIndex(config=config)
if not config['mwindex']:
sys.stderr.write('ERROR: Provide a URL to Index.php\n')
sys.exit(1)
elif wikiengine == 'wikispaces':
import wikispaces
# use wikicanonical for base url for Wikispaces?
# calculating path, if not defined by user with --path=
if not config['path']:
config['path'] = './%s-%s-wikidump' % (domain2prefix(config=config),
config['date'])
return config
def getParameters(params=[]):
""" Import parameters into variable """
if not params:
params = sys.argv
config = {}
parser = argparse.ArgumentParser(description='Tools for downloading and preserving wikis.')
# General params
parser.add_argument(
'-v', '--version', action='version', version=getVersion())
parser.add_argument(
'--cookies', metavar="cookies.txt", help="Path to a cookies.txt file.")
parser.add_argument(
'--delay',
metavar=5,
default=0,
type=float,
help="Adds a delay (in seconds).")
parser.add_argument(
'--retries',
metavar=5,
default=5,
help="Maximum number of retries.")
parser.add_argument('--path', help='Path to store wiki dump at.')
parser.add_argument(
'--resume',
action='store_true',
help='Resumes previous incomplete dump (requires --path).')
parser.add_argument('--force', action='store_true', help='')
parser.add_argument(
'--user', help='Username if authentication is required.')
parser.add_argument(
'--pass',
dest='password',
help='Password if authentication is required.')
# URL params
# This script should work with any general URL, finding out
# API, index.php or whatever by itself when necessary
groupWiki = parser.add_argument_group()
groupWiki.add_argument(
'wiki',
default='',
nargs='?',
help="URL to wiki (e.g. http://wiki.domain.org).")
# URL params for MediaWiki
groupWiki.add_argument(
'--mwapi',
help="URL to MediaWiki API (e.g. http://wiki.domain.org/w/api.php).")
groupWiki.add_argument(
'--mwindex',
help="URL to MediaWiki index.php (e.g. http://wiki.domain.org/w/index.php).")
# Download params
groupDownload = parser.add_argument_group(
'Data to download',
'What info download from the wiki')
groupDownload.add_argument(
'--pages',
action='store_true',
help="Generates a dump of pages (--pages --curonly for current revisions only).")
groupDownload.add_argument('--curonly', action='store_true',
help='Store only the current version of pages.')
groupDownload.add_argument(
'--images', action='store_true', help="Generates an image dump.")
groupDownload.add_argument(
'--namespaces',
metavar="1,2,3",
help='Comma-separated value of namespaces to include (all by default).')
groupDownload.add_argument(
'--exnamespaces',
metavar="1,2,3",
help='Comma-separated value of namespaces to exclude.')
# Meta info params
groupMeta = parser.add_argument_group(
'Meta info',
'What meta info to retrieve from the wiki')
groupMeta.add_argument(
'--get-api',
action='store_true',
help="Returns wiki API when available.")
groupMeta.add_argument(
'--get-index',
action='store_true',
help="Returns wiki Index.php when available.")
groupMeta.add_argument(
'--get-page-titles',
action='store_true',
help="Returns wiki page titles.")
groupMeta.add_argument(
'--get-image-names',
action='store_true',
help="Returns wiki image names.")
groupMeta.add_argument(
'--get-namespaces',
action='store_true',
help="Returns wiki namespaces.")
groupMeta.add_argument(
'--get-wiki-engine',
action='store_true',
help="Returns wiki engine.")
args = parser.parse_args()
#sys.stderr.write(args)
# Not wiki? Exit
if not args.wiki:
sys.stderr.write('ERROR: Provide a URL to a wiki\n')
parser.print_help()
sys.exit(1)
# Don't mix download params and meta info params
if (args.pages or args.images) and \
(args.get_api or args.get_index or args.get_page_titles or args.get_image_names or args.get_namespaces or args.get_wiki_engine):
sys.stderr.write('ERROR: Don\'t mix download params and meta info params\n')
parser.print_help()
sys.exit(1)
# No download params and no meta info params? Exit
if (not args.pages and not args.images) and \
(not args.get_api and not args.get_index and not args.get_page_titles and not args.get_image_names and not args.get_namespaces and not args.get_wiki_engine):
sys.stderr.write('ERROR: Use at least one download param or meta info param\n')
parser.print_help()
sys.exit(1)
# Load cookies
cj = cookielib.MozillaCookieJar()
if args.cookies:
cj.load(args.cookies)
sys.stderr.write('Using cookies from %s\n' % args.cookies)
# check user and pass (one requires both)
if (args.user and not args.password) or (args.password and not args.user):
sys.stderr.write('ERROR: Both --user and --pass are required for authentication.\n')
parser.print_help()
sys.exit(1)
session = None
if args.user and args.password:
import requests
session = requests.Session()
session.cookies = cj
session.headers.update({'User-Agent': getUserAgent()})
session.auth = (args.user, args.password)
#session.mount(args.mw_api.split('/api.php')[0], HTTPAdapter(max_retries=max_ret)) Mediawiki-centric, be careful
# check URLs
for url in [args.mwapi, args.mwindex, args.wiki]:
if url and (not url.startswith('http://') and not url.startswith('https://')):
sys.stderr.write(url)
sys.stderr.write('ERROR: URLs must start with http:// or https://\n')
parser.print_help()
sys.exit(1)
# Meta info params
metainfo = '' # only one allowed, so we don't mix output
if args.get_api:
metainfo = 'get_api'
elif args.get_index:
metainfo = 'get_index'
elif args.get_page_titles:
metainfo = 'get_page_titles'
elif args.get_image_names:
metainfo = 'get_image_names'
elif args.get_namespaces:
metainfo = 'get_namespaces'
elif args.get_wiki_engine:
metainfo = 'get_wiki_engine'
namespaces = ['all']
exnamespaces = []
# Process namespace inclusions
if args.namespaces:
# fix, why - ? and... --namespaces= all with a space works?
if re.search(
r'[^\d, \-]',
args.namespaces) and args.namespaces.lower() != 'all':
sys.stderr.write("Invalid namespace values.\nValid format is integer(s) separated by commas\n")
sys.exit()
else:
ns = re.sub(' ', '', args.namespaces)
if ns.lower() == 'all':
namespaces = ['all']
else:
namespaces = [int(i) for i in ns.split(',')]
# Process namespace exclusions
if args.exnamespaces:
if re.search(r'[^\d, \-]', args.exnamespaces):
sys.stderr.write("Invalid namespace values.\nValid format is integer(s) separated by commas\n")
sys.exit(1)
else:
ns = re.sub(' ', '', args.exnamespaces)
if ns.lower() == 'all':
sys.stderr.write('You cannot exclude all namespaces.\n')
sys.exit(1)
else:
exnamespaces = [int(i) for i in ns.split(',')]
# --curonly requires --xml
if args.curonly and not args.pages:
sys.stderr.write("--curonly requires --pages\n")
parser.print_help()
sys.exit(1)
config = {
'cookies': args.cookies or '',
'curonly': args.curonly,
'date': datetime.datetime.now().strftime('%Y%m%d'),
'delay': args.delay,
'exnamespaces': exnamespaces,
'images': args.images,
'logs': False,
'metainfo': metainfo,
'namespaces': namespaces,
'pages': args.pages,
'path': args.path and os.path.normpath(args.path) or '',
'retries': int(args.retries),
'wiki': args.wiki,
'wikicanonical': '',
'wikiengine': getWikiEngine(args.wiki),
'other': {
'configfilename': 'config.txt',
'filenamelimit': 100, # do not change
'force': args.force,
'resume': args.resume,
'session': session,
}
}
# Get ready special variables (API for MediWiki, etc)
if config['wikiengine'] == 'mediawiki':
import mediawiki
config['mwexport'] = 'Special:Export'
if not args.mwapi:
config['mwapi'] = mediawiki.mwGetAPI(config=config)
if not config['mwapi']:
sys.stderr.write('ERROR: Provide a URL to API\n')
sys.exit(1)
else:
data={
'action': 'query',
'meta': 'siteinfo',
'siprop': 'namespaces',
'format': 'json'}
r = getURL(config['mwapi'], data=data)
config['mwexport'] = getJSON(r)['query']['namespaces']['-1']['*'] \
+ ':Export'
if not args.mwindex:
config['mwindex'] = mediawiki.mwGetIndex(config=config)
if not config['mwindex']:
sys.stderr.write('ERROR: Provide a URL to Index.php\n')
sys.exit(1)
elif wikiengine == 'wikispaces':
import wikispaces
# use wikicanonical for base url for Wikispaces?
# calculating path, if not defined by user with --path=
if not config['path']:
config['path'] = './%s-%s-wikidump' % (domain2prefix(config=config), config['date'])
return config