[FFmpeg-devel] [PATCH 1/2] avformat/utils: Export coded dimensions unconditionally

[Hendrik Leppkes]

On Sun, Jun 5, 2016 at 5:13 AM, Michael Niedermayer
<michael at niedermayer.cc> wrote:
> This fixes a API regression
> Probably fixes Ticket5451
>
> Signed-off-by: Michael Niedermayer <michael at niedermayer.cc>
> ---
>  libavformat/utils.c   |    4 ++--
>  libavformat/version.h |    2 +-
>  2 files changed, 3 insertions(+), 3 deletions(-)
>
> diff --git a/libavformat/utils.c b/libavformat/utils.c
> index ac056c4..7ca0a12 100644
> --- a/libavformat/utils.c
> +++ b/libavformat/utils.c
> @@ -3789,8 +3789,6 @@ FF_DISABLE_DEPRECATION_WARNINGS
>              av_codec_set_lowres(st->codec, av_codec_get_lowres(st->internal->avctx));
>              st->codec->width = st->internal->avctx->width;
>              st->codec->height = st->internal->avctx->height;
> -            st->codec->coded_width = st->internal->avctx->coded_width;
> -            st->codec->coded_height = st->internal->avctx->coded_height;
>          }
>
>          if (st->codec->codec_tag != MKTAG('t','m','c','d'))
> @@ -3807,6 +3805,8 @@ FF_DISABLE_DEPRECATION_WARNINGS
>          }
>
>          // Fields unavailable in AVCodecParameters
> +        st->codec->coded_width = st->internal->avctx->coded_width;
> +        st->codec->coded_height = st->internal->avctx->coded_height;
>          st->codec->properties = st->internal->avctx->properties;
>  FF_ENABLE_DEPRECATION_WARNINGS
>  #endif
> diff --git a/libavformat/version.h b/libavformat/version.h
> index c92a23f..2f79b7f 100644
> --- a/libavformat/version.h
> +++ b/libavformat/version.h
> @@ -29,7 +29,7 @@
>
>  #include "libavutil/version.h"
>
> -// When bumping major check Ticket5467, 5421 for regressing
> +// When bumping major check Ticket5467, 5421, 5451 for regressing
>  // Also please add any ticket numbers that you belive might regress here
>  #define LIBAVFORMAT_VERSION_MAJOR  57
>  #define LIBAVFORMAT_VERSION_MINOR  37
> --
> 1.7.9.5

The ticket in question accesses st->codec to get this field. st->codec
is going away, so there is no point in testing this "regression",
since they need to update their API usage first for it to even
compile.
Heck the same thing applies to the other tickets you listed here, all
of those access information from st->codec which is going away
entirely, so you can't really "test" these cases. Someone would have
to update the code to use codecpar first, and while doing this
migration they would need to preserve whatever behavior exists today.

- Hendrik