def login(self, username, password, domain=None):
"""Login via MediaWiki API
@param username: username
@type username: unicode
@param password: password
@type password: unicode
@param domain: optional domain
@type domain: unicode
@returns: True if login succeeded, False otherwise
@rtype: bool
"""
args = {
'action': 'login',
'lgname': username.encode('utf-8'),
'lgpassword': password.encode('utf-8'),
'format': 'json',
}
if domain is not None:
args['lgdomain'] = domain.encode('utf-8')
result = utils.fetch_url('%sapi%s' % (self.base_url, self.script_extension),
post_data=args,
ignore_errors=False,
opener=self.opener,
)
result = json.loads(result)
if 'login' in result and result['login'].get('result') == 'Success':
return True
return False
def getDiskPath(self, name, size=None):
"""Return filename for image with given name and size
@param name: image name (without namespace, i.e. without 'Image:')
@type name: unicode
@param size: if given, the image is converted to the given maximum width
@type size: int or NoneType
@returns: filename of image or None if image could not be found
@rtype: basestring
"""
assert isinstance(name, unicode), 'name must be of type unicode'
url = self.getURL(name, size=size)
if url is None:
return None
ext = url.rsplit('.')[-1]
if size is not None:
ext = '%dpx.%s' % (size, ext)
else:
ext = '.%s' % ext
filename = os.path.join(self.tmpdir, utils.fsescape(name + ext))
if utils.fetch_url(url, ignore_errors=True, output_filename=filename):
return filename
else:
return None
def getDiskPath(self, name, size=None):
"""Return filename for image with given name and size
@param name: image name (without namespace, i.e. without 'Image:')
@type name: unicode
@param size: if given, the image is converted to the given maximum width
@type size: int or NoneType
@returns: filename of image or None if image could not be found
@rtype: basestring
"""
assert isinstance(name, unicode), 'name must be of type unicode'
url = self.getURL(name, size=size)
if url is None:
return None
ext = url.rsplit('.')[-1]
if size is not None:
ext = '%dpx.%s' % (size, ext)
else:
ext = '.%s' % ext
filename = os.path.join(self.tmpdir, utils.fsescape(name + ext))
if utils.fetch_url(url, ignore_errors=True, output_filename=filename):
return filename
else:
return None
def login(self, username, password, domain=None):
"""Login via MediaWiki API
@param username: username
@type username: unicode
@param password: password
@type password: unicode
@param domain: optional domain
@type domain: unicode
@returns: True if login succeeded, False otherwise
@rtype: bool
"""
args = {
'action': 'login',
'lgname': username.encode('utf-8'),
'lgpassword': password.encode('utf-8'),
'format': 'json',
}
if domain is not None:
args['lgdomain'] = domain.encode('utf-8')
result = utils.fetch_url('%sapi%s' % (self.base_url, self.script_extension),
post_data=args,
ignore_errors=False,
opener=self.opener,
)
result = json.loads(result)
if 'login' in result and result['login'].get('result') == 'Success':
return True
return False
def addLicense(mbook):
license_text = utils.fetch_url(
LICENSE_URL,
ignore_errors=False,
expected_content_type='text/x-wiki',
)
license_text = unicode(license_text, 'utf-8')
license = {'mw_rights_text': license_text,
'name': 'GNU Free Documentation License',
}
mbook['licenses'] = [license]
def query(self, ignore_errors=True, num_tries=2, **kwargs):
args = {
'action': 'query',
'format': 'json',
}
args.update(**kwargs)
for k, v in args.items():
if isinstance(v, unicode):
args[k] = v.encode('utf-8')
q = urllib.urlencode(args)
q = q.replace('%3A', ':') # fix for wrong quoting of url for images
q = q.replace('%7C', '|') # fix for wrong quoting of API queries (relevant for redirects)
for i in range(num_tries):
try:
s = time.time()
data = utils.fetch_url('%sapi%s?%s' % (self.base_url, self.script_extension, q),
ignore_errors=ignore_errors,
opener=self.opener,
)
elapsed = time.time() - s
if elapsed > self.long_request:
log.warn('Long request: HTTP request took %f s' % elapsed)
if data is not None:
break
except:
if i == num_tries - 1:
raise
log.warn('Fetching failed. Trying again.')
time.sleep(0.5)
if ignore_errors and data is None:
log.error('Got no data from api%s' % self.script_extension)
return None
try:
data = unicode(data, 'utf-8')
if data and data[0] == u'\ufeff': # strip off BOM
# Note that a BOM is actually *not allowed* at the beginning of a JSON string
# see http://www.ietf.org/rfc/rfc4627.txt, section "3. Encoding"
data = data[1:]
return json.loads(data)['query']
except KeyError:
log.error('Response from api%s did not contain a query result' % self.script_extension)
return None
except Exception, e:
log.error('Got exception: %r' % e)
if ignore_errors:
return None
raise RuntimeError('api%s query failed. Are you sure you specified the correct base URL?' % self.script_extension)
def query(self, ignore_errors=True, num_tries=2, **kwargs):
args = {
'action': 'query',
'format': 'json',
}
args.update(**kwargs)
for k, v in args.items():
if isinstance(v, unicode):
args[k] = v.encode('utf-8')
q = urllib.urlencode(args)
q = q.replace('%3A', ':') # fix for wrong quoting of url for images
q = q.replace('%7C', '|') # fix for wrong quoting of API queries (relevant for redirects)
for i in range(num_tries):
try:
s = time.time()
data = utils.fetch_url('%sapi%s?%s' % (self.base_url, self.script_extension, q),
ignore_errors=ignore_errors,
opener=self.opener,
)
elapsed = time.time() - s
if elapsed > self.long_request:
log.warn('Long request: HTTP request took %f s' % elapsed)
if data is not None:
break
except:
if i == num_tries - 1:
raise
log.warn('Fetching failed. Trying again.')
time.sleep(0.5)
if ignore_errors and data is None:
log.error('Got no data from api%s' % self.script_extension)
return None
try:
data = unicode(data, 'utf-8')
if data and data[0] == u'\ufeff': # strip off BOM
# Note that a BOM is actually *not allowed* at the beginning of a JSON string
# see http://www.ietf.org/rfc/rfc4627.txt, section "3. Encoding"
data = data[1:]
return json.loads(data)['query']
except KeyError:
log.error('Response from api%s did not contain a query result' % self.script_extension)
return None
except Exception, e:
log.error('Got exception: %r' % e)
if ignore_errors:
return None
raise RuntimeError('api%s query failed. Are you sure you specified the correct base URL?' % self.script_extension)
def get_licenses(self):
"""Return list of licenses
@returns: list of dicts with license info
@rtype: [dict]
"""
if 'licenses' not in self.metabook:
return []
licenses = []
for license in self.metabook['licenses']:
wikitext = ''
if license.get('mw_license_url'):
wikitext = utils.fetch_url(
license['mw_license_url'],
ignore_errors=True,
expected_content_type='text/x-wiki',
)
if wikitext:
try:
wikitext = unicode(wikitext, 'utf-8')
except UnicodeError:
wikitext = None
else:
wikitext = ''
if license.get('mw_rights_text'):
wikitext = license['mw_rights_text']
if license.get('mw_rights_page'):
wikitext += '\n\n[[%s]]' % license['mw_rights_page']
if license.get('mw_rights_url'):
wikitext += '\n\n' + license['mw_rights_url']
if not wikitext:
continue
licenses.append({
'title': license.get('name', u'License'),
'wikitext': wikitext,
})
return licenses
def get_licenses(self):
"""Return list of licenses
@returns: list of dicts with license info
@rtype: [dict]
"""
if 'licenses' not in self.metabook:
return []
licenses = []
for license in self.metabook['licenses']:
wikitext = ''
if license.get('mw_license_url'):
wikitext = utils.fetch_url(
license['mw_license_url'],
ignore_errors=True,
expected_content_type='text/x-wiki',
)
if wikitext:
try:
wikitext = unicode(wikitext, 'utf-8')
except UnicodeError:
wikitext = None
else:
wikitext = ''
if license.get('mw_rights_text'):
wikitext = license['mw_rights_text']
if license.get('mw_rights_page'):
wikitext += '\n\n[[%s]]' % license['mw_rights_page']
if license.get('mw_rights_url'):
wikitext += '\n\n' + license['mw_rights_url']
if not wikitext:
continue
licenses.append({
'title': license.get('name', u'License'),
'wikitext': wikitext,
})
return licenses
def get_licenses(metabook):
"""Return list of licenses
@returns: list of dicts with license info
@rtype: [dict]
"""
import re
from mwlib import utils
retval = []
for l in metabook.licenses:
wikitext = ''
if l.get('mw_license_url'):
url = l['mw_license_url']
if re.match(r'^.*/index\.php.*action=raw',
url) and 'templates=expand' not in url:
url += '&templates=expand'
wikitext = utils.fetch_url(
url,
ignore_errors=True,
expected_content_type='text/x-wiki',
)
if wikitext:
try:
wikitext = unicode(wikitext, 'utf-8')
except UnicodeError:
wikitext = None
else:
wikitext = ''
if l.get('mw_rights_text'):
wikitext = l['mw_rights_text']
if l.get('mw_rights_page'):
wikitext += '\n\n[[%s]]' % l['mw_rights_page']
if l.get('mw_rights_url'):
wikitext += '\n\n' + l['mw_rights_url']
if not wikitext:
continue
retval.append(
license(title=l.get('name', u'License'), wikitext=wikitext))
return retval
def get_licenses(metabook):
"""Return list of licenses
@returns: list of dicts with license info
@rtype: [dict]
"""
import re
from mwlib import utils
retval = []
for l in metabook.licenses:
wikitext = ''
if l.get('mw_license_url'):
url = l['mw_license_url']
if re.match(r'^.*/index\.php.*action=raw', url) and 'templates=expand' not in url:
url += '&templates=expand'
wikitext = utils.fetch_url(url,
ignore_errors=True,
expected_content_type='text/x-wiki',
)
if wikitext:
try:
wikitext = unicode(wikitext, 'utf-8')
except UnicodeError:
wikitext = None
else:
wikitext = ''
if l.get('mw_rights_text'):
wikitext = l['mw_rights_text']
if l.get('mw_rights_page'):
wikitext += '\n\n[[%s]]' % l['mw_rights_page']
if l.get('mw_rights_url'):
wikitext += '\n\n' + l['mw_rights_url']
if not wikitext:
continue
retval.append(license(title=l.get('name', u'License'),
wikitext=wikitext))
return retval
