On Wed, 12 Oct 2011 23:41:40 +0300 (EEST), Martin Storsjö <[email protected]> 
wrote:
> On Wed, 12 Oct 2011, Anton Khirnov wrote:
> 
> > On Wed, 12 Oct 2011 22:30:54 +0200, Anton Khirnov <[email protected]> wrote:
> >>
> >>>      /** Whether the server accepts the x-Dynamic-Rate header */
> >>>      int accept_dynamic_rate;
> >>> +
> >>> +    /**
> >>> +     * Various option flags for the RTSP muxer/demuxer.
> >>> +     */
> >>> +    int rtsp_flags;
> >>>  } RTSPState;
> >>>
> >>> +#define RTSP_FLAG_FILTER_SRC  0x1
> >>
> >> I would make this a separate option -- makes for less code, since you
> >> can set rt->filter_source directly.
> >>
> >
> > Disregard this, i wrote this before looking at the next patch.
> > Though i wonder if using flags is really better. They require more
> > knowledge from the user.
> 
> I kind of prefer flags over int options for something that's really a 
> bool, specifying -filter_source 1 looks worse than -rtsp_flags 
> filter_source I think. Or maybe not? But it's all quite cosmetic which one 
> is chosen in this case, either one is doable for me in this case.
> 
> Flags might feel more scalable if there's going to be many other bool 
> options, but I don't know if that really is the case...

Keep the flags if you prefer them, it's your code after all.

The next patch is ok in that case.

-- 
Anton Khirnov
_______________________________________________
libav-devel mailing list
[email protected]
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to