[FFmpeg-devel] qt-faststart refactor exit code

Frank Barchard fbarchard
Mon Jun 29 21:36:51 CEST 2009


Hey Ramiro!  I use your autobuilds :)  Very helpful, thanks.  It would be
nice if qt-faststart were in bin/ prebuilds.
Re PRIu64
Done.  Thanks!  That allowed me to remove the casting to unsigned long long
too.

Updated diff, not inverted this time.


On Fri, Jun 26, 2009 at 5:26 PM, Ramiro Polla <ramiro.polla at gmail.com>wrote:

> On Fri, Jun 26, 2009 at 8:45 PM, Frank Barchard<fbarchard at google.com>
> wrote:
> > Diego had asked I break out the exit cleanup of the qt-faststart update
> > The old tool had some leaks during error handling.  Doing a single
> cleanup
> > at the end reduces the code size and makes it more robust to errors.
> >
> > I've also added printfs which help diagnose problems and moved the error
> > checks after the printfs, so you can see what they refer to.
> > For the printf, I ran into an issue with %llu on MinGW and had to ifdef a
> > formatter to use I64.
>
> There's PRIu64 defined in inttypes.h for that, as in:
> printf("value: %"PRIu64"\n", something_64bit);
>
> > Tested with linux, cygwin, mingw, vc and icl.
> > For vc / icl I used a patched inttypes.h, so no code change is necessary.
> >  Built like this:
> > icl /Ox /I\inttypes qt-faststart_exit.c /Feqt-faststart_exit_icl.exe
> >
> > This diff assumes you have the 'update' for 'free' atom removal, so I've
> > also included full source
>
> Isn't the diff inverted?
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at mplayerhq.hu
> https://lists.mplayerhq.hu/mailman/listinfo/ffmpeg-devel
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: qt-fastexit2.diff
Type: application/octet-stream
Size: 9734 bytes
Desc: not available
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090629/945aaf15/attachment.obj>



More information about the ffmpeg-devel mailing list