[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kfm-devel
Subject:    Re: [PATCH] Bug #28908: login not possible (cookies ?)
From:       Dawit Alemayehu <adawit () kde ! org>
Date:       2001-08-26 3:05:24
[Download RAW message or body]

On Saturday 25 August 2001 11:27, Martin Antenreiter wrote:
> Hi!
>
> Konqi doesn't send "domain" cookies back to the server
> (http://online.tu-graz.ac.at). The cookies look like (You can see it if
> you use the link "Anmelden"):
>
>   Set-Cookie: TUG_LOGIN_ID=87789E213DBBB9633CD43FB1486E3C
>               ;domain=tu-graz.ac.at ;path=/tug_online/ ;secure
>
> I think the kcookiejar save this cookie under the name "tu-graz.ac.at".
> Before a GET call kcookiejar search the cookie via ".tu-graz.ac.at" and
> can't find it.  Therefore you get the message "Your browser does not
> support cookies...". The little patch corrects this and shouldn't break
> anything. Hopefully!
>
>  But there is another problem with the page (maybe I must blame the page
> author). However Netscape, IExplorer & Opera handle the site correctly.
> It's a problem how Konqi handles session cookies.
>  Description: After you went to "Anmelden" you get a child-window open,
> get a cookie and a redirect. If you send the cookie back you will see in
> the child-window a login-form. After successful login you get another
> cookie and then the javascript closes my child-window. Now comes the call
> "KCookieJar::eatSessionCookies(int winId)" and eats my login-cookies. My
> Konqi comes in troubles! After that the main-window makes a reload
> (JavaScript). Konqi doesn't and can't send my login-cookies anymore. The
> consequence of this is, that I got the same page as an anonymous user.
> Arg! I thought about it but couldn't find a smart solution for the
> problem. Any suggestions?

Hi again :),

This is my doing as well in my attempt to implement session based cookie handling!
It looks like I have no choice but to do reference counting and remove the cookie
only when all windows refering to it have disappeared.  Hmm... I have to think about
this one a bit...

Regards,
Dawit A.

[prev in list] [next in list] [prev in thread] [next in thread]