MediaWiki自动登陆和更新页面

Mediawiki系统开放了API，向wiki系统的api.php发送http请求，可实现很多功能。

1.登陆：原理可查看mediawiki api文档

# Login

userName = '***'

password = '***'

 

headers = dict()

headers['Content-Type'] = 'application/x-www-form-urlencoded'

headers['Connection'] = 'Keep-Alive'

 

params = dict()

params['action'] = 'login'

params['format'] = 'xml'

params['lgname'] = userName

params['lgpassword'] = password

 

response, content = http.request(LOGIN_URL, method='POST', body=urllib.urlencode(params), headers=headers)

 

cookie = response.get('set-cookie')

xmldoc = xml.dom.minidom.parseString(content)

token = xmldoc.getElementsByTagName('login')[0].getAttribute('token')

 

headers['Cookie'] = cookie

params['lgtoken'] = token

 

response, content = http.request(LOGIN_URL, method='POST', body=urllib.urlencode(params), headers=headers)

xmldoc = xml.dom.minidom.parseString(content)

node = xmldoc.getElementsByTagName('login')[0]

status = node.getAttribute('result')

if status != 'Success':

    print "Login Not success..."

    exit()

print "Success Login..."

 

token = node.getAttribute('lgtoken')

userId = node.getAttribute('lguserid')

userName = urllib.quote(userName)

cookiePrefix = node.getAttribute('cookieprefix')

 

# Construct request headers

cookies = list()

cookies.append(cookie)

cookies.append(cookiePrefix + 'UserName=' + userName)

cookies.append(cookiePrefix + 'UserId=' + userId)

cookies.append(cookiePrefix + 'Token=' + token)

 

headers['Cookie'] = '; '.join(cookies)

 

然后使用这个包含登录成功后返回的Cookies的Header请求其他页面，就被认为是已登录的

 

2.编辑：原理也可参见mediawiki api文档，就是先获取edittoken，然后发送请求编辑

page = chineseEncode('首页')

#response, content = http.request('http://****/mediawiki/api.php?format=xml&action=query&titles=%s&prop=revisions&rvprop=content' % (page), method='GET')

response, content = http.request('http://****/mediawiki/api.php?format=xml&action=query&titles=%s&prop=info|revisions&intoken=edit' % (page), method='GET', headers=headers)

 

match = re.search(r'edittoken="(\w+\+\\)"', content)

if match:

    editToken = match.group(1)

print "editToken is %s" % (editToken)

 

params = {}

params['format'] = 'xml'

params['action'] = 'edit'

params['title'] = 'newPage'

params['summary'] = 'This is a test summary.'

params['text'] = 'This is my edit text content.'

params['token'] = editToken

response, content = http.request('http://****/mediawiki/api.php', method='POST', body=urllib.urlencode(params), headers=headers)

// 查看content内容，发现里面包含success，编辑成功，查看页面，确实内容改变了

print content