Michael-- Unless I misunderstand, it sounds like you are writing your own CAS client. Have you considered using any of the existing clients, which exist for a large number of platforms and languages? I believe they all handle GET persistence for you, and POST persistence is on at least a couple client roadmaps.
That being said: 1) GET persistence is easily achieved by keeping the current query attached to the URL when encoding the service parameter in the redirect to the CAS server. 2) POST persistence is harder, since redirects are performed as GETs. You would have to cache the original POST, redirect, recognize the return (GET), and convert the GET request into a POST, with the original POST'd content. HTH, -Matt On Fri, Sep 12, 2008 at 12:47 PM, Michael Dalby <[EMAIL PROTECTED]> wrote: > I have a question about getting CGI variables and using CAS. I'm not > sure if there are different implementations of CAS, but the way we are > currently using it requires a different service ticket for every http > request to my CGI. When a user requests the page, I check for a "ticket" > variable in the URL, if they don't have it I 302 redirect them to our > CAS server who redirects them back to my page with the "ticket" variable > appended to the URL. This means that if a user attempts to GET or POST > variables to my page, I lose them because the redirection to and from > CAS. I'm hoping that I'm just doing something wrong here, because the > alternative involves a lot of hacks to get this working. Thanks much for > the help! > _______________________________________________ > Yale CAS mailing list > [email protected] > http://tp.its.yale.edu/mailman/listinfo/cas > -- [EMAIL PROTECTED] Key ID:D6EEC5B5 _______________________________________________ Yale CAS mailing list [email protected] http://tp.its.yale.edu/mailman/listinfo/cas
