On Saturday 30 Jul 2005 18:28, Guillaume Laurent wrote:
> Looks like a problem with how the size of the peak values array is
> computed :
If you look at lines 446 - 451 (451 is where it crashes):
446 if (position < 0) continue;
447 if (position >= values.size() - channels) break;
448
449 if (channels == 1) {
450
451 h1 = values[position++];
So we know that at line 451 (before the post-increment), position >= 0
and position < size() - 1.
That guarantees we're looking up a valid vector element, except in one
case -- when the vector is empty. The old code has an explicit check
for this, the new code doesn't.
If values is empty, "values.size() - channels" will wrap around
silently, because values.size() is unsigned. (How apposite, following
that thread about unsigned versus signed ints the other day!)
> I did this because calling repaint() for every preview was causing
> too much flicker, so I ended up calling it only when the preview
> queue was emptied. However now that it seems calling update() works,
> I can get the old behavior back if you prefer.
Yes, I prefer -- AudioPreviewThread is clever enough to try to do
smaller previews first, precisely to accommodate incremental refresh.
Chris
-------------------------------------------------------
SF.Net email is sponsored by: Discover Easy Linux Migration Strategies
from IBM. Find simple to follow Roadmaps, straightforward articles,
informative Webcasts and more! Get everything you need to get up to
speed, fast. http://ads.osdn.com/?ad_id=7477&alloc_id=16492&op=click
_______________________________________________
Rosegarden-devel mailing list
[email protected] - use the link below to unsubscribe
https://lists.sourceforge.net/lists/listinfo/rosegarden-devel
