Geoffrey Talvola wrote:

Aha. Exceptions can definitely cause sleep() to be skipped.

If an exception happens during respond() or during awake() then sleep() does
not get called. This smells like a bug to me.


I always thought that this was funny. We wrap up the EndResponse(s) that we raise with a .cleanup() method (which is usually called from sleep).

SO, we're still cleaning up, even when EndResponse is sent. But this is basically working around what we think is a design flaw in the whole endresponse idea.


Looking at this more closely, there ought to be a try/finally in there so
that sleep() always gets called if awake() succeeds, something like:


I've thought the same thing, and have been meaning to whine about it to the list. If awake runs, sleep should run. And you may be right; maybe it should ALWAYS run.


------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click _______________________________________________ Webware-discuss mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/webware-discuss

Reply via email to