Conrad, Small buglet with the 0.9.1 liboggz release (go dude!) http://thaumas.net/~giles/xiph/elphel/clips/elphel_00017.ogg causes a float exception in oggzinfo when it tries to calculate the bitrate. It fails to measure the duration and tries to divide by zero. :) There may well be something wrong with the file, although oggz-validate doesn't complain. Also, the configure script doesn't mention the new tools in its target summary. Thanks for the cool new release, -r
On Fri, Apr 08, 2005 at 12:45:08PM -0700, Ralph Giles wrote:> Conrad, > > Small buglet with the 0.9.1 liboggz release (go dude!) > > http://thaumas.net/~giles/xiph/elphel/clips/elphel_00017.ogg causes a > float exception in oggzinfo when it tries to calculate the bitrate. It > fails to measure the duration and tries to divide by zero. :) > > There may well be something wrong with the file, although oggz-validate > doesn't complain.hmm, the file looks ok. (an interesting test video too :) I'm tracking this bug at: http://trac.annodex.net/ticket/117> Also, the configure script doesn't mention the new tools in its target > summary.fixed in svn :) Conrad.
On Sun, Apr 10, 2005 at 01:14:14PM +1000, Conrad Parker wrote:> On Fri, Apr 08, 2005 at 12:45:08PM -0700, Ralph Giles wrote: > > Conrad, > > > > Small buglet with the 0.9.1 liboggz release (go dude!) > > > > http://thaumas.net/~giles/xiph/elphel/clips/elphel_00017.ogg causes a > > float exception in oggzinfo when it tries to calculate the bitrate. It > > fails to measure the duration and tries to divide by zero. :)ok, this is all fixed now in svn :)> http://trac.annodex.net/ticket/117I've closed the ticket, but please check it. The comments include links to the changesets involved, which provide further details. It was basically due to the way oggzinfo calculated the timestamp of the last video frame -- for the files produced by the elphel camera, the frames don't necessarily fit on a single page. Handling this actually involved *removing* a single line of code from oggzinfo :-) Conrad.