1 # python-simplemediawiki - Extremely low-level wrapper to the MediaWiki API
2 # Copyright (C) 2010 Red Hat, Inc.
4 # This library is free software; you can redistribute it and/or modify it under
5 # the terms of the GNU Lesser General Public License as published by the Free
6 # Software Foundation; either version 2.1 of the License, or (at your option)
9 # This library is distributed in the hope that it will be useful, but WITHOUT
10 # ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
11 # FOR A PARTICULAR PURPOSE. See the GNU Lesser General Public License for more
14 # You should have received a copy of the GNU General Public License along with
15 # this program. If not, see <http://www.gnu.org/licenses/>.
18 simplemediawiki is an extremely low-level wrapper to the MediaWiki API. It
19 automatically handles cookies and gzip compression so that you can make basic
20 calls to the API in the easiest way possible. It also provides a few functions
21 to make day-to-day API access easier.
23 To use this module, instantiate a MediaWiki object, passing it the URL of
24 api.php for the wiki you want to work with. Calls go through MediaWiki.call().
25 A generic login wrapper as well as functions to determine limits and get a list
26 of namespaces are provided for your convenience.
28 >>> from simplemediawiki import MediaWiki
29 >>> wiki = MediaWiki('http://en.wikipedia.org/w/api.php')
30 >>> wiki.call({'action': 'query', 'prop': 'revisions', 'titles': 'Main Page'})
31 {u'query': {u'pages': {...}}}
36 from iso8601 import iso8601
38 from StringIO import StringIO
45 Class to represent a MediaWiki installation with an enabled API.
47 api_url: URL to api.php (usually similar to http://example.com/w/api.php)
51 _psuedo_namespaces = None
53 def __init__(self, api_url, cookie_file=None):
54 self._api_url = api_url
56 self._cj = cookielib.MozillaCookieJar(cookie_file)
63 self._cj = cookielib.CookieJar()
64 self._opener = urllib2.build_opener(urllib2.HTTPCookieProcessor(self._cj))
66 def _fetch_http(self, url, params):
67 request = urllib2.Request(url, urllib.urlencode(params))
68 request.add_header('Accept-encoding', 'gzip')
69 response = self._opener.open(request)
70 if isinstance(self._cj, cookielib.MozillaCookieJar):
72 if response.headers.get('Content-Encoding') == 'gzip':
73 compressed = StringIO(response.read())
74 gzipper = gzip.GzipFile(fileobj=compressed)
77 data = response.read()
80 def call(self, params):
82 Make a call to the wiki. Returns a dictionary that represents the JSON
85 params['format'] = 'json'
86 return json.loads(self._fetch_http(self._api_url, params))
88 def normalize_api_url(self):
90 This function checks the given URL for a correct API endpoint and
91 returns that URL, while also helpfully setting this object's API URL to
92 it. If it can't magically conjure an API endpoint, it returns False.
94 data, data_json = self._normalize_api_url_tester(self._api_url)
98 # if there's an index.php in the URL, we might find the API
99 if 'index.php' in self._api_url:
100 test_api_url = self._api_url.split('index.php')[0] + 'api.php'
102 test_data, test_data_json = \
103 self._normalize_api_url_tester(test_api_url)
104 print (test_data, test_data_json)
106 self._api_url = test_api_url
110 def _normalize_api_url_tester(self, api_url):
111 data = self._fetch_http(api_url, {'action': 'query',
116 data_json = json.loads(data)
117 # may as well set the version
119 version_string = data_json['query']['general']['generator']
120 self._mediawiki_version = version_string.split(' ', 1)[1]
123 return (data, data_json)
127 def login(self, user, passwd, token=None):
129 Convenience function for logging into the wiki. It should never be
130 necessary to provide a token argument; it is part of the login process
131 since MediaWiki 1.15.3 (see MediaWiki bug 23076).
133 data = {'action': 'login',
135 'lgpassword': passwd}
137 data['lgtoken'] = token
138 result = self.call(data)
139 if result['login']['result'] == 'Success':
141 elif result['login']['result'] == 'NeedToken' and not token:
142 return self.login(user, passwd, result['login']['token'])
146 def limits(self, low, high):
148 Convenience function for determining appropriate limits in the API. If
149 the logged in user has the "apihighlimits" right, it will return the
150 high argument; otherwise it will return the low argument.
152 if self._high_limits == None:
153 result = self.call({'action': 'query',
156 self._high_limits = 'apihighlimits' in \
157 result['query']['userinfo']['rights']
158 if self._high_limits:
163 def namespaces(self, psuedo=True):
165 Fetches a list of namespaces for this wiki.
167 if self._namespaces == None:
168 result = self.call({'action': 'query',
170 'siprop': 'namespaces'})
171 self._namespaces = {}
172 self._psuedo_namespaces = {}
173 for nsid in result['query']['namespaces']:
175 self._namespaces[int(nsid)] = \
176 result['query']['namespaces'][nsid]['*']
178 self._psuedo_namespaces[int(nsid)] = \
179 result['query']['namespaces'][nsid]['*']
182 retval.update(self._namespaces)
183 retval.update(self._psuedo_namespaces)
186 return self._namespaces
189 def parse_date(date):
191 Converts dates provided by the MediaWiki API into datetime.datetime
194 return iso8601.parse_date(date)
197 __author__ = 'Ian Weller <ian@ianweller.org>'
198 __version__ = '1.0.1'