Chris Gahan wrote:
> I think I've found a bug in the session handling. I found it by
> invalidate()ing a session and having webkit close
> unexpectedly. The next
> time I loaded the page, I got this:
> 
>   File "c:\Python23\Webware\WebKit\SessionFileStore.py", line 55, in
>     __getitem__ raise KeyError, key
> KeyError: '20040112215608-a98df35317154b63e8e2e6e201e31b7f'
> 
> The SessionFileStore doesn't have this session in it anymore (for
> whatever reason), and since Webkit can't find this session, it will
> always return
> this page until I manually delete my cookie.
> 
> Not very desirable behaviour. :)
> 
> What should happen, obivously, is that an empty session
> should be re-created
> when it can't find one, or that it should tell the browser to
> delete its
> cookie.
> 
> Should I fix this?

Yes, an invalid session ID should never cause a traceback.  My understanding
is that if the IgnoreInvalidSession config setting is set to 0 then you are
suppsed to get a canned error page; and if it's set to 1 then you are
supposed to get a brand-new empty session.  This used to work some time in
the past, but maybe it's broken in CVS?  I have to confess that I'm still
using Webware 0.8.1 myself.

If you could submit a patch, that would be appreciated.  I'll try to make
some time to review the patch and check it in.

- Geoff



-------------------------------------------------------
This SF.net email is sponsored by: Perforce Software.
Perforce is the Fast Software Configuration Management System offering
advanced branching capabilities and atomic changes on 50+ platforms.
Free Eval! http://www.perforce.com/perforce/loadprog.html
_______________________________________________
Webware-discuss mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/webware-discuss

Reply via email to