On Sun, Sep 14, 2014 at 07:43:27AM -0700, Deb Mukherjee wrote:
> Handles decoding of new VP9 profiles 1-3 with different color sampling
> and bit-depths.
>
> For high bitdepth (profiles 2 and 3) support, we currently need to link
> with the highbitdepth branch of libvpx with --enable-experimental
Deb Mukherjee google.com> writes:
> +if (avctx->codec_id == AV_CODEC_ID_VP8) {
> +if (img->fmt != VPX_IMG_FMT_I420)
Wouldn't (avctx->codec_id == ... && img->fmt != ...)
be simpler?
> +return 0;
Shouldn't this be return AVERROR_INVALIDDATA?
Thank you, Carl Eugen
_
On Fri, Sep 12, 2014 at 04:07:16PM -0700, Deb Mukherjee wrote:
> Handles decoding of new VP9 profiles 1-3 with different color sampling
> and bit-depths.
>
> For high bitdepth (profiles 2 and 3) support, we currently need to link
> with the highbitdepth branch of libvpx with --enable-experimental
On 22/08/14 5:31 PM, Deb Mukherjee wrote:
> Handles new VP9 profiles 1-3 with different color sampling and
> bit-depths.
> ---
> libavcodec/libvpxdec.c | 69
> +++---
> 1 file changed, 60 insertions(+), 9 deletions(-)
>
> diff --git a/libavcodec/libvpx
On 22.08.2014, at 22:31, Deb Mukherjee wrote:
> +// returns 0 on success, 1 on unsupported
Please choose one of the standard ways, either 1 for success and 0 failure, or
negative failure, 0 or positive success. This variant, at least in the context
of FFmpeg, is just confusing/misleading
Though
On Fri, 22 Aug 2014 13:31:06 -0700
Deb Mukherjee wrote:
> Handles new VP9 profiles 1-3 with different color sampling and
> bit-depths.
> ---
> -static int vp8_decode(AVCodecContext *avctx,
> +static int vpx_decode(AVCodecContext *avctx,
> -static av_cold int vp8_free(AVCodecContext *avctx)
> +s
