[FFmpeg-devel] [RFC] The Big Bump

[Stefano Sabatini]

On date Thursday 2011-02-03 19:40:07 +0100, Michael Niedermayer encoded:
> On Thu, Feb 03, 2011 at 06:46:34PM +0100, Stefano Sabatini wrote:
> > On date Thursday 2011-02-03 16:35:31 +0100, Anton Khirnov encoded:
> > > Hi,
> > > it's been suggested on IRC that we've accumulated enough new APIs and
> > > the associated cruft so the time to bump major for lavf and lavc is
> > > nigh. We should definitely do that before 0.7.
> > > 
> > > The following patch series bumps lavf and lavc and removes all the
> > > deprecated stuff (though i certainly don't expect it to be applied right
> > > now as is). The diffstat looks pretty nice, removing over a thousand
> > > LOC.
> > > 
> > > Please comment.
> > 
> > There is currently a problem with lavu error codes, AVERROR(EINVAL)
> > and AVERROR_INVALIDDATA share the same code, I proposed sometime ago
> > to bump lavu and all the rest of the libav* libraries, at the time the
> > idea was rejected, I wonder if this is a good occasion for doing it.
> 
> maybe but you should discuss this with users of the libs, theres some chance
> things like this might cause a chain reaction of soname bumps througout libs
> and i mean more than you listed
> 

> And i think it might be possible to fix things like this in a less disruptive
> way
> That is introducing a new error code and switching to that gradually deprecating
> the old and then eventually droping it

I see, I'll try to elaborate this solution.

> > This would require to bump all the major numbers (lavu, lavc, lavf,
> > lavfi, lavdev, lavcore, lsws), check which option you prefer:
> > * we don't care
> > * it's overkill
> > * can be delayed more
> > * can be done now
> 
> IMHO as proposed "it's overkill"  but thats just my oppinion
-- 
FFmpeg = Faithful and Fundamental Martial Puristic Explosive Geisha