On Wed, May 4, 2011 at 11:05 AM, Ronald S. Bultje <[email protected]> wrote:
>
> Hi,
>
> 2011/5/4 Måns Rullgård <[email protected]>:
> > Diego Biurrun <[email protected]> writes:
> >
> >> On Wed, May 04, 2011 at 01:47:30PM -0400, Ronald S. Bultje wrote:
> >>>
> >>> 2011/5/4 Diego Biurrun <[email protected]>:
> >>> > --- a/libavcodec/Makefile
> >>> > +++ b/libavcodec/Makefile
> >>> > @@ -546,6 +546,7 @@ OBJS-$(CONFIG_SPDIF_DEMUXER)           += 
> >>> > aacadtsdec.o mpeg4audio.o
> >>> > +OBJS-$(CONFIG_WTV_DEMUXER)             += mpeg4audio.o mpegaudiodata.o
> >>> >
> >>> > --- a/libavformat/Makefile
> >>> > +++ b/libavformat/Makefile
> >>> > @@ -295,7 +295,8 @@ OBJS-$(CONFIG_WEBM_MUXER)                += 
> >>> > matroskaenc.o matroska.o \
> >>> >  OBJS-$(CONFIG_WSVQA_DEMUXER)             += westwood.o
> >>> > -OBJS-$(CONFIG_WTV_DEMUXER)               += wtv.o asf.o asfdec.o 
> >>> > mpegts.o riff.o
> >>> > +OBJS-$(CONFIG_WTV_DEMUXER)               += wtv.o asfdec.o asf.o 
> >>> > asfcrypt.o \
> >>> > +                                            avlanguage.o mpegts.o 
> >>> > isom.o riff.o
> >>> >  OBJS-$(CONFIG_WV_DEMUXER)                += wv.o apetag.o
> >>>
> >>> How much of each file does it really need?
> >>
> >> I did not investigate, it is a recursive dependency of sorts:
> >> asfdec.o needs asfcrypt.o and avlanguage.o, mpegts.o needs
> >> isom.o and the libavcodec deps.
> >
> > Then it is perhaps better to break up the chain into several explicit steps.
>
> Before we go there.
>
> _How much_ of each file does it need? Does it use every function in
> mpegts.c, and does that use every function in isom.c/libavcodec? Or
> are there things that could be split because we only need mpegts:x
> whereas the independent piece mpegts:y needs isom.c:z (which should
> then be split from the rest of isom.c, if that makes sense)?
>

Some of these files are already the result of the process you just
described. E.g. isom.c are the mov parts that get reused.
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to