[PEAK] Problems with events.twisted_support

[Phillip J. Eby]

At 03:52 PM 2/5/04 -0800, John Landahl wrote:
>On Thu, 05 Feb 2004 18:40:16 -0500, Phillip J. Eby <pje at telecommunity.com> 
>wrote:
>
>>Okay, I'll try to take a look at it this weekend.  I'm buried with "real 
>>work" issues this week.  If you can distill this down to a simple failing 
>>unit test, it would be a *big* help.
>
>Attached is a small program which demonstrates the problem.

Aha!  twisted_support.EventLoop.runUntil() is what's broken.  It shouldn't 
call reactor.run() if 'exit' is non-empty at that point (i.e., if the event 
fired immediately upon adding the callback).

No time to fix it right now, but at least you can work around.  Thanks for 
the report and I will add the test and fix the other problems this weekend 
if I can.