def __next__(self):
if self.max_items is not None:
if self.count >= self.max_items:
raise StopIteration
# For filered lists, we might have to do several requests
# to get the next element due to miser mode.
# See: https://github.com/mwclient/mwclient/issues/194
while True:
try:
item = six.next(self._iter)
if item is not None:
break
except StopIteration:
if self.last:
raise
self.load_chunk()
self.count += 1
if 'timestamp' in item:
item['timestamp'] = parse_timestamp(item['timestamp'])
if isinstance(self, GeneratorList):
return item
if isinstance(self.return_values, tuple):
return tuple((item[i] for i in self.return_values))
if self.return_values is not None:
return item[self.return_values]
return item
def __init__(self, site, name, info=None, extra_properties=None):
if type(name) is type(self):
self.__dict__.update(name.__dict__)
return
self.site = site
self.name = name
self._textcache = {}
if not info:
if extra_properties:
prop = 'info|' + '|'.join(six.iterkeys(extra_properties))
extra_props = []
for extra_prop in six.itervalues(extra_properties):
extra_props.extend(extra_prop)
else:
prop = 'info'
extra_props = ()
if isinstance(name, int):
info = self.site.get('query',
prop=prop,
pageids=name,
inprop='protection',
*extra_props)
else:
info = self.site.get('query',
prop=prop,
titles=name,
inprop='protection',
*extra_props)
info = six.next(six.itervalues(info['query']['pages']))
self._info = info
if 'invalid' in info:
raise mwklient.errors.InvalidPageTitle(info.get('invalidreason'))
self.namespace = info.get('ns', 0)
self.name = info.get('title', u'')
if self.namespace:
self.page_title = strip_namespace(self.name)
else:
self.page_title = self.name
self.touched = parse_timestamp(info.get('touched'))
self.revision = info.get('lastrevid', 0)
self.exists = 'missing' not in info
self.length = info.get('length')
self.protection = {
i['type']: (i['level'], i['expiry'])
for i in info.get('protection', ()) if i
}
self.redirect = 'redirect' in info
self.pageid = info.get('pageid', None)
self.contentmodel = info.get('contentmodel', None)
self.pagelanguage = info.get('pagelanguage', None)
self.restrictiontypes = info.get('restrictiontypes', None)
self.last_rev_time = None
self.edit_time = None
def revisions(self, revids, prop='ids|timestamp|flags|comment|user'):
"""Get data about a list of revisions.
See also the `Page.revisions()` method.
API doc: https://www.mediawiki.org/wiki/API:Revisions
Example: Get revision text for two revisions:
>>> for revision in site.revisions([689697696, 689816909],
prop='content'):
... print revision['*']
Args:
revids (list): A list of (max 50) revisions.
prop (str): Which properties to get for each revision.
Returns:
A list of revisions
"""
kwargs = {
'prop': 'revisions',
'rvprop': prop,
'revids': '|'.join(map(text_type, revids))
}
revisions = []
pages = self.get('query', **kwargs).get('query', {}).get('pages',
{}).values()
for page in pages:
for revision in page.get('revisions', ()):
revision['pageid'] = page.get('pageid')
revision['pagetitle'] = page.get('title')
revision['timestamp'] = parse_timestamp(revision['timestamp'])
revisions.append(revision)
return revisions
def _edit(self, summary, minor, bot, section, **kwargs):
self._check_edit()
data = {}
if minor:
data['minor'] = '1'
if not minor:
data['notminor'] = '1'
if self.last_rev_time:
data['basetimestamp'] = time.strftime('%Y%m%d%H%M%S',
self.last_rev_time)
if self.edit_time:
data['starttimestamp'] = time.strftime('%Y%m%d%H%M%S',
self.edit_time)
if bot:
data['bot'] = '1'
if section:
data['section'] = section
data.update(kwargs)
if self.site.force_login:
data['assert'] = 'user'
def do_edit():
result = self.site.post(
'edit',
title=self.name,
summary=summary,
token=self.__get_token__('edit'),
**data)
if result['edit'].get('result').lower() == 'failure':
raise mwklient.errors.EditError(self, result['edit'])
return result
try:
result = do_edit()
except mwklient.errors.APIError as err:
if err.code == 'badtoken':
# Retry, but only once to avoid an infinite loop
self.__get_token__('edit', force=True)
try:
result = do_edit()
except mwklient.errors.APIError as err:
self.handle_edit_error(err, summary)
else:
self.handle_edit_error(err, summary)
# 'newtimestamp' is not included if no change was made
if 'newtimestamp' in result['edit'].keys():
self.last_rev_time = parse_timestamp(
result['edit'].get('newtimestamp'))
# Workaround for https://phabricator.wikimedia.org/T211233
for cookie in self.site.connection.cookies:
if 'PostEditRevision' in cookie.name:
self.site.connection.cookies.clear(
cookie.domain, cookie.path, cookie.name)
# clear the page text cache
self._textcache = {}
return result['edit']
def test_parse_nonempty_timestamp(self):
nice_ts = time.struct_time((2015, 1, 2, 20, 18, 36, 4, 2, -1))
self.assertEqual(nice_ts, parse_timestamp('2015-01-02T20:18:36Z'))
def test_parse_none_timestamp(self):
empty_ts = time.struct_time((0, 0, 0, 0, 0, 0, 0, 0, 0))
self.assertEqual(empty_ts, parse_timestamp(None))
def test_parse_empty_timestamp(self):
empty_ts = time.struct_time((0, 0, 0, 0, 0, 0, 0, 0, 0))
self.assertEqual(empty_ts, parse_timestamp('0000-00-00T00:00:00Z'))
def test_parse_missing_timestamp(self):
assert time.struct_time(
(0, 0, 0, 0, 0, 0, 0, 0, 0)) == parse_timestamp(None)