On Fri, 13 Jan 2017 13:29:55 +0100
Anton Khirnov <an...@khirnov.net> wrote:

> Quoting wm4 (2017-01-13 12:09:23)
> > On Fri, 13 Jan 2017 12:04:36 +0100
> > Anton Khirnov <an...@khirnov.net> wrote:
> >   
> > > Before this commit, AVIOContext is to be freed with a plain av_free(),
> > > which prevents us from adding any deeper structure to it.
> > > ---
> > >  doc/APIchanges        |  3 +++
> > >  libavformat/avio.h    |  8 ++++++++
> > >  libavformat/aviobuf.c | 17 ++++++++++++++---
> > >  libavformat/version.h |  4 ++--
> > >  4 files changed, 27 insertions(+), 5 deletions(-)
> > > 
> > > diff --git a/doc/APIchanges b/doc/APIchanges
> > > index c8c2a21..1cab9f4 100644
> > > --- a/doc/APIchanges
> > > +++ b/doc/APIchanges
> > > @@ -13,6 +13,9 @@ libavutil:     2015-08-28
> > >  
> > >  API changes, most recent first:
> > >  
> > > +2016-xx-xx - xxxxxxx - lavf 57.11.0 - avio.h
> > > +  Add avio_context_free(). From now on it must be used for freeing 
> > > AVIOContext.
> > > +  
> > 
> > As an API user I'm saying "lol no". This implicitly breaks every API
> > user without warning, and you'd only know once it actually breaks (and
> > you've spent a few hours of debugging). Unless you read APIchanges,
> > which not many are likely to do.  
> 
> I'm not breaking anything just yet, and not planning to in the immediate
> future. The (preliminary) transition plan I have in mind involved
> deprecating avio_alloc_context() and replacing it with
> avio_context_alloc(), with a properly big-endian-namespaced name and
> most likely a different signature. That will give you all the
> deprecation warnings you want. The exact shape of that new API is to be
> determined, but it seemed to me that it's best to add the destructor as
> soon as possible, since the API for that is very straightforward.

Uh, ok, I guess that works.

> (also, if you're a libav API user and don't read APIchanges at least
> once per major bump then you deserve everything you get =p)
> 

Even if I do I can miss something like that easily.

In general I don't really think you can expect users to read that,
unless you present it on a golden tablet on every major release.
_______________________________________________
libav-devel mailing list
libav-devel@libav.org
https://lists.libav.org/mailman/listinfo/libav-devel

Reply via email to