[FFmpeg-devel] [PATCH] lavfi: add dejudder filter to remove judder produced by partially telecined material.

Paul B Mahol onemda at gmail.com
Sun Feb 9 16:19:55 CET 2014


On 2/9/14, Nicholas Robbins <nickrobbins at yahoo.com> wrote:
>> On Saturday, February 8, 2014 7:26 PM, Lukasz Marek
>> <lukasz.m.luki at gmail.com> wrote:
>
>>>>>    +    int a, b, c, d;
>>>  a--d are indexes to a ring buffer. I could name them "previous,
>> second-previous, ultimate, penultimate" but that seemed cumbersome. Would
>> a
>> comment here about what they are be helpful?
>>
>> maybe i1, i2, i3, i4 is better than a,b,c,d. At least it says they are
>> indexes.
>
> Ok, I'll change them to i1, i2, i3, i4.
>
>>>>>    +static int filter_frame(AVFilterLink *inlink, AVFrame *frame)
>>>>>    +{
>>>>>    +    int i;
>>>>>    +    AVFilterContext *ctx  = inlink->dst;
>>>>>    +    AVFilterLink *outlink = ctx->outputs[0];
>>>>>    +    DejudderContext *dj   = ctx->priv;
>>>>
>>>>>    +    int64_t *judbuff      = dj->ringbuff;
>>>>
>>>>  you can probably remove the indirection and leave optims to the
>>>>  compiler
>>>
>>>  If I understand you, you are suggesting I replace all my judbuff's in
>> the code with "inlink->dst->priv->ringbuff". This seems a
>> little cumbersome. Is that what you mean? Other filters seem to do what
>> I've
>> done here.
>>
>> I think Stefano referred the last one.
>
> So replace all my "judbuff"s with "dj->ringbuff"s? Is that what you mean?

You do not need to comply to every nonsense someone have on its mind.

>
> --
> Nicholas Robbins
>
> _______________________________________________
> ffmpeg-devel mailing list
> ffmpeg-devel at ffmpeg.org
> http://ffmpeg.org/mailman/listinfo/ffmpeg-devel
>


More information about the ffmpeg-devel mailing list