[FFmpeg-devel] [PATCH] ffmpeg: implement -force_key_frames_expr option
stefasab at gmail.com
Sun Jan 27 17:02:02 CET 2013
On date Thursday 2013-01-24 13:07:36 +0100, Stefano Sabatini encoded:
> On date Tuesday 2013-01-22 01:16:13 +0100, Stefano Sabatini encoded:
> > On date Sunday 2013-01-13 22:48:17 +0100, Nicolas George encoded:
> > > Le quartidi 24 nivôse, an CCXXI, Stefano Sabatini a écrit :
> > > > Also, after some thinking I'm not sure it is a good idea to "pollute"
> > > > libavcodec (and I can't find a simple way to implement this at the
> > > > library level), and I don't want to churn library code more than
> > > > necessary. Thus basically I'm reproposing this patch in the current
> > > > form, which fixes the problem for me.
> > >
> > > I think I agree with that position.
> > >
> > > > A possible alternative to the current approach (which computes the
> > > > time value for the next split) would be to evaluate an expression for
> > > > every frame, so that for example:
> > > > expr:n*10
> > > >
> > > > would become:
> > > > expr:gte(t, n*10)
> > > >
> > > > which would be more flexible.
> > >
> > > If there are enough available variables, then this is better indeed.
> > Updated.
> > Note: an ifelse() ternary function would simplify syntax.
> Updated. I'll push in three days if I read no comments.
> FFmpeg = Fascinating and Freak Monstrous Philosofic Exxagerate Gangster
> From 5ebbeaa8629209f582e097f049225b720eb59f66 Mon Sep 17 00:00:00 2001
> From: Stefano Sabatini <stefasab at gmail.com>
> Date: Sun, 9 Dec 2012 18:40:22 +0100
> Subject: [PATCH] ffmpeg: implement -force_key_frames expression evalution
> doc/ffmpeg.texi | 43 ++++++++++++++++++++++++++++++++++++--
> ffmpeg.c | 62 ++++++++++++++++++++++++++++++++++++++++++++++++++-----
> ffmpeg.h | 14 +++++++++++++
> 3 files changed, 112 insertions(+), 7 deletions(-)
Pushed with a few minor fixes.
FFmpeg = Fancy & Foolish Meaningful Pure EnGine
More information about the ffmpeg-devel