[MPlayer-dev-eng] fix braindamaged -edlout behavior
Reynaldo H. Verdejo Pinochet
reynaldo at opendot.cl
Tue Jul 4 22:06:17 CEST 2006
On Tue, Jul 04, 2006 at 02:51:42PM +0200, Reimar Doeffinger wrote:
> Hi,
> On Tue, Jul 04, 2006 at 05:36:40AM -0400, Reynaldo H. Verdejo Pinochet wrote:
> > +float begin_skip = 0; ///< start time of the current skip while on edlout mode
>
> I think this maybe should be made double, mostly for the reason given
> below.
>
> > case MP_CMD_EDL_MARK:
> > - if( edl_fd ) {
> > - float v = sh_video ? sh_video->pts :
> > - playing_audio_pts(sh_audio, d_audio, audio_out);
> > - fprintf( edl_fd, "%f %f %d\n", v-2, v, 0 );
> > - }
> > + if( edl_fd ) {
> > + float v = sh_video ? sh_video->pts :
> > + playing_audio_pts(sh_audio, d_audio, audio_out);
> > +
> > + if( begin_skip == 0 )
> > + {
> > + begin_skip = v;
>
> IMHO these both should be double, since 0 might be a valid pts
Really? I mean, sure pts can be a valid timestamp but, but I
wonder if it is really posible to make an 'i' keypress yielding
pts ~ 0 so 0-pts_at_keystroke its bellow float range and v gets
'0' asigned to it.
> timestamp. MP_NOPTS_VALUE should be used instead, but that probably will
> not work with float.
>
> And despite the horrible mixture of tabs and space in the current code,
> resend without cosmetic changes.
Not a problem but keep in mind there are only 3 lines reindented here.
> Also, most of the surrounding code uses spaces, not tabs, so I'd prefer
> the new code to do this as well...
Not a problem neither. will send an updated one, anyway, what do you
guys think about the idea behind this new behavior?
> Greetings,
> Reimar Doeffinger
Regards
Reynaldo
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/mplayer-dev-eng/attachments/20060704/f62423c4/attachment.pgp>
More information about the MPlayer-dev-eng
mailing list