On Tue, Mar 22, 2016 at 09:30:32AM +0200, Martin Storsjö wrote:
> On Mon, 21 Mar 2016, Diego Biurrun wrote:
> > --- a/libavformat/Makefile
> > +++ b/libavformat/Makefile
> > @@ -416,12 +416,12 @@ SKIPHEADERS-$(CONFIG_FFRTMPCRYPT_PROTOCOL) += rtmpdh.h
> > SKIPHEADERS-$(CONFIG_NETWORK) += network.h rtsp.h
> >
> > TESTPROGS = seek \
> > - srtp \
> > url \
> >
> > TESTPROGS-$(CONFIG_MOV_MUXER) += movenc
> > TESTPROGS-$(CONFIG_NETWORK) += noproxy
> > TESTPROGS-$(CONFIG_FFRTMPCRYPT_PROTOCOL) += rtmpdh
> > +TESTPROGS-$(CONFIG_SRTP_PROTOCOL) += srtp
>
> Why?
>
> There's two layers here, srtp.{c,h}, which is an internal API within lavf,
> which is used by the rtpdec subsystem if needed. It is also used by the
> srtp protocol, which is a separate entity, which can be enabled/disabled
> independently. This test only tests the srtp internal API, which is, in
> principle, unrelated to whether the srtp protocol is registered.
>
> I can imagine that you want to disable it if both rtpdec and srtpproto are
> disabled though (if building lavf with most components disabled, I guess
> it doesn't make much sense to test this), and only checking the srtp
> protocol perhaps makes sense in one way if you insist though, for
> simplicity.
You're right, better patch(es) coming up soon.
Thanks, Diego
_______________________________________________
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel
