Hi Nick,

thank you for your detailed report.  I think I have fixed the problem,
can you svn up and let me know if it works better?  If not, I'll hack
on it more so that things get back to a good state tonight.  Sorry for
the glitch.

Thanks,
Hans

On Thu, Jul 31, 2008 at 17:43, Nick Allen <rflug05@gmail.com> wrote:
> Hello Hans,
> The specific problem I ran into was a recursive lock error thrown when
> setting a session value in the same request that starts a session. I'm using
> rev 3591. here's an example:
>  (defun test-dispatcher (hunchentoot:*request*)
>    (lambda ()
>    ;; set the session value FOO to the value of
>    ;; the HTTP get parameter "set-foo", if it
>    ;; exists
>    (let ((set-foo (hunchentoot:get-parameter "set-foo")))
>      (when set-foo
>        (setf (hunchentoot:session-value 'foo) set-foo)))
>    ;; then return a page that shows the session
>    ;; value FOO
>    (format nil "The foo is ~A" (hunchentoot:session-value 'foo))))
>
> ;; start the server
> (pushnew 'test-dispatcher hunchentoot:*dispatch-table*)
> (defvar *test-server* (hunchentoot:start-server :port 2001))
> ;; now go to:
> ;;
> ;;    http://localhost:2001?set-foo=bar
> ;;
> ;; it should look like:
> ;;
> ;;    The Foo is bar
> ;;
> ;; but it throws a recursive lock error
> On Thu, Jul 31, 2008 at 7:09 AM, Hans Hübner <hans@huebner.org> wrote:
>>
>> Hi Nick,
>>
>> thank you for reporting a problem - Can you provide me with details on
>> how the removal of WITH-RECURSIVE-LOCK breaks Hunchentoot for
>> Lispworks?  I have removed it because I believed that there is no code
>> path that could be to recursive locking, but I could have been missing
>> something.
>>
>> In general, the Subversion repository is less stable than Edi's
>> releases and we are planning to make a proper release, yet I can't
>> make any promises.
>>
>> Thanks,
>> Hans
>>
>> On Thu, Jul 31, 2008 at 03:58, Nick Allen <rflug05@gmail.com> wrote:
>> > Hello!
>> > Is the "ediware" location of the hunchentoot source the recommended
>> > version
>> > for deployment? or should we be using the tarball linked off the
>> > documentation or the "hans" version?  I'm wondering because
>> > WITH-RECURSIVE-LOCK-HELD seems to have disappeared out of
>> > "lispworks.lisp",
>> > which seems to have introducing some bugs, which seems to me to suggests
>> > to
>> > me that the "ediware" repo is either out of date or too bleeding edge...
>> > take care
>> > Nick
>> > _______________________________________________
>> > tbnl-devel site list
>> > tbnl-devel@common-lisp.net
>> > http://common-lisp.net/mailman/listinfo/tbnl-devel
>> >
>> _______________________________________________
>> tbnl-devel site list
>> tbnl-devel@common-lisp.net
>> http://common-lisp.net/mailman/listinfo/tbnl-devel
>
>
> _______________________________________________
> tbnl-devel site list
> tbnl-devel@common-lisp.net
> http://common-lisp.net/mailman/listinfo/tbnl-devel
>
_______________________________________________
tbnl-devel site list
tbnl-devel@common-lisp.net
http://common-lisp.net/mailman/listinfo/tbnl-devel