I help run a BBS that's perl/cgi based. One script(threadmachine.cgi) obtains and displays the posts for a given thread with a submit post form at the bottom.
When a new post is submitted, the form action is another script(bbs.cgi) that actually adds the post to the thread data. bbs.cgi's normal (non-error) output is simply "Location: threadmachine.cgi\n\n". When run, the script should detect new posts and refresh the page accordingly. The problem is it doesn't refresh.
What I can figure out to this point is since the browser is at threadmachine.cgi, and the only output of bbs.cgi is the location line telling the browser to go back to where it essentially still is, it doesn't refresh at all. The location line in the browser actually displays bbs.cgi, something it should never really see at all.
However, it sporadically works correctly with no setting changes at all. If the browser is set to cache nothing and always get new documents, it works fine every time.
The only thing I can think of is to make an html page with the meta-refresh headers to redirect back to threadmachine, but I'd like some other way to convince the brower that it really does need to query the script again. Any ideas? ----------------------------------------------------------------------------------
...but I'm just a C man trying to see the light
When a new post is submitted, the form action is another script(bbs.cgi) that actually adds the post to the thread data. bbs.cgi's normal (non-error) output is simply "Location: threadmachine.cgi\n\n". When run, the script should detect new posts and refresh the page accordingly. The problem is it doesn't refresh.
What I can figure out to this point is since the browser is at threadmachine.cgi, and the only output of bbs.cgi is the location line telling the browser to go back to where it essentially still is, it doesn't refresh at all. The location line in the browser actually displays bbs.cgi, something it should never really see at all.
However, it sporadically works correctly with no setting changes at all. If the browser is set to cache nothing and always get new documents, it works fine every time.
The only thing I can think of is to make an html page with the meta-refresh headers to redirect back to threadmachine, but I'd like some other way to convince the brower that it really does need to query the script again. Any ideas? ----------------------------------------------------------------------------------
...but I'm just a C man trying to see the light