On Tuesday 23 November 2004 23:06, Hans Meine wrote:
> from reading the comment about thread-safety, I would propose one of the
> attached patches (the _shorter one doing the acquire() before the len()
> call) just to be sure.
GRRRRRRRRRRRR! I don't get it - everytime when I send a patch, I manage to
introduce some spurious tabs in the last moment, just before doing the final
"cvs diff". :-( Inspite of my desperate attempts to double-check everything
in my xemacs with whitespace-mode.. :(
(The same happend with my last nvram-wakeup patch IIRC.)
Here is the shorter version again..
--
Ciao, / / .o.
/--/ ..o
/ / ANS ooo
Index: rc.py
===================================================================
RCS file: /cvsroot/freevo/freevo/src/Attic/rc.py,v
retrieving revision 1.36.2.1
diff -u -3 -p -d -r1.36.2.1 rc.py
--- rc.py 20 Oct 2004 18:31:46 -0000 1.36.2.1
+++ rc.py 23 Nov 2004 22:12:43 -0000
@@ -495,12 +495,14 @@ class EventHandler:
# search for events in the queue
- if len(self.queue):
- self.lock.acquire()
- ret = self.queue[0]
- del self.queue[0]
+ self.lock.acquire()
+ try:
+ if len(self.queue):
+ ret = self.queue[0]
+ del self.queue[0]
+ return ret
+ finally:
self.lock.release()
- return ret
# search all input objects for new events
for i in self.inputs:
