Below is what my response.info() gives when I attempt login.. There is no ‘set-cookie’ header at all in the response…  L

Date: Mon, 28 Apr 2008 14:30:47 GMT

Content-Length: 6286

Content-Type: text/html; charset=utf-8

Expires: Mon, 28 Apr 2008 14:30:47 GMT

Cache-Control: private, s-maxage=0, max-age=0, must-revalidate

Connection: close

Server: Apache

X-Powered-By: PHP/5.2.5

Content-Language: ml

Vary: Accept-Encoding,Cookie

X-Vary-Options: Accept-Encoding;list-contains=gzip,Cookie;string-contains=mlwiki

Token;string-contains=mlwikiLoggedOut;string-contains=mlwiki_session;string-cont

ains=centralauth_Token;string-contains=centralauth_Session;string-contains=centr

alauth_LoggedOut

Content-Encoding: gzip

X-Cache: MISS from sq33.wikimedia.org

X-Cache-Lookup: MISS from sq33.wikimedia.org:3128

X-Cache: MISS from sq24.wikimedia.org

X-Cache-Lookup: MISS from sq24.wikimedia.org:80

Via: 1.0 sq33.wikimedia.org:3128 (squid/2.6.STABLE18), 1.0 sq24.wikimedia.org:80

 (squid/2.6.STABLE18), 1.0 proxyqh01 (NetCache NetApp/6.0.5), 1.1 proxyin11 (Net

Cache NetApp/6.0.5)

Below is the postData function that I have tweaked

    def postData(self, address, data,

                 contentType='application/x-www-form-urlencoded',

                 sysop=False, useCookie=True, compress=True):

        req=urllib2.Request('http://' + self.hostname() + address)

        req.add_unredirected_header('Content-Length', str(len(data)))

        req.add_header('Content-type', contentType)

        req.add_header('User-agent', useragent)

        if useCookie and self.cookies(sysop = sysop):

            req.add_header('Cookie', self.cookies(sysop = sysop))

        if compress:

            req.add_header('Accept-encoding', 'gzip')

        req.add_data(data)

        cookies = urllib2.HTTPCookieProcessor()

        opener = urllib2.build_opener(cookies)

        response = opener.open(req)

        data = response.read()     

        if compress:

            print response.info()

            headers = str(response.info()).split('\n')

            for i in headers:

                  if i.startswith('Content-Encoding: gzip'):

                        data = decompress_gzip(data)

        data = data.decode(self.encoding())

        response.close()

        self._getUserData(data, sysop = sysop)

        return response, data

Any help would be much appreciated !

Regards,

Jacob

---

From: Jacob Jose [mailto:jacob@ti.com]
Sent: Sunday, April 27, 2008 7:54 PM
To: 'pywikipedia-l@lists.wikimedia.org'
Subject: Login from behind Proxy

Hi all,

I operate my bot – DragonBot – in wikipedias. Until a week or so back, my bot could login to wikipedias and wiktionaries from behind a proxy – by tweaking the wikipedia.py post function . Now I observe that it can’t login.

1. My source code hasn’t changed.
2. I tried with the 24-Apr Nightly codebase as well.

Has something been changed in the wikipedia server configuration? Say, disabling proxy caching or something like that at the server….

Regards,

Jacob