[FFmpeg-devel] [PATCH] fftools/ffmpeg.c: allow forcing input framerate on streamcopy

[Leo Izen]

On 10/20/18 6:47 PM, Michael Niedermayer wrote:
> On Fri, Oct 19, 2018 at 07:29:28PM -0400, Leo Izen wrote:
>> On 10/19/18 3:02 PM, Carl Eugen Hoyos wrote:
>>> 2018-10-19 20:39 GMT+02:00, Leo Izen <leo.izen at gmail.com>:
>>>> On 10/19/18 2:26 PM, Carl Eugen Hoyos wrote:
>>>>> 2018-10-19 4:58 GMT+02:00, Leo Izen <leo.izen at gmail.com>:
>>>>>> ---
>>>>>>    fftools/ffmpeg.c | 8 +++++---
>>>>>>    1 file changed, 5 insertions(+), 3 deletions(-)
>>>>>>
>>>>>> diff --git a/fftools/ffmpeg.c b/fftools/ffmpeg.c
>>>>>> index da4259a9a8..5d68194676 100644
>>>>>> --- a/fftools/ffmpeg.c
>>>>>> +++ b/fftools/ffmpeg.c
>>>>>> @@ -2045,12 +2045,14 @@ static void do_streamcopy(InputStream *ist,
>>>>>> OutputStream *ost, const AVPacket *p
>>>>>>        if (ost->enc_ctx->codec_type == AVMEDIA_TYPE_VIDEO)
>>>>>>            ost->sync_opts++;
>>>>>>
>>>>>> -    if (pkt->pts != AV_NOPTS_VALUE)
>>>>>> +    if (ist->framerate.num)
>>>>>> +        opkt.pts = av_rescale_q(ist->pts, AV_TIME_BASE_Q,
>>>>>> ost->mux_timebase) - ost_tb_start_time;
>>>>>> +    else if (pkt->pts != AV_NOPTS_VALUE)
>>>>>>            opkt.pts = av_rescale_q(pkt->pts, ist->st->time_base,
>>>>>> ost->mux_timebase) - ost_tb_start_time;
>>>>>>        else
>>>>>>            opkt.pts = AV_NOPTS_VALUE;
>>>>>>
>>>>>> -    if (pkt->dts == AV_NOPTS_VALUE)
>>>>>> +    if (pkt->dts == AV_NOPTS_VALUE || ist->framerate.num)
>>>>>>            opkt.dts = av_rescale_q(ist->dts, AV_TIME_BASE_Q,
>>>>>> ost->mux_timebase);
>>>>>>        else
>>>>>>            opkt.dts = av_rescale_q(pkt->dts, ist->st->time_base,
>>>>>> ost->mux_timebase);
>>>>>> @@ -2602,7 +2604,7 @@ static int process_input_packet(InputStream *ist,
>>>>>> const AVPacket *pkt, int no_eo
>>>>>>            avpkt = *pkt;
>>>>>>        }
>>>>>>
>>>>>> -    if (pkt && pkt->dts != AV_NOPTS_VALUE) {
>>>>>> +    if (pkt && pkt->dts != AV_NOPTS_VALUE && !ist->framerate.num) {
>>>>>>            ist->next_dts = ist->dts = av_rescale_q(pkt->dts,
>>>>>> ist->st->time_base, AV_TIME_BASE_Q);
>>>>>>            if (ist->dec_ctx->codec_type != AVMEDIA_TYPE_VIDEO ||
>>>>>> !ist->decoding_needed)
>>>>>>                ist->next_pts = ist->pts = ist->dts;
>>>>> How can this be tested?
>>>>>
>>>>> Carl Eugen
>>>> I'm not entirely sure. I ran "make fate" and it passed, and I
>>>> successfully rescaled a 30fps clip to 15fps, 20fps, 45fps, and 60fps
>>>> using -r:v as an input option. I'm not entirely sure what the standard
>>>> procedure is for performing more rigorous tests.
>>> What I meant was:
>>> Which kind of input and output did you use to test your patch?
>>>
>>> Carl Eugen
>>> _______________________________________________
>> It worked perfectly at various framerates for AVI files. It also worked in
>> practice for MP4 and Matroska files, although my patch doesn't properly set
>> the avg_frame_rate and r_frame_rate of the output stream. Intelligent
>> players like MPV play it fine and ignore the incorrect *_frame_rate
>> metadata. Anyway, here's an updated patch that fixes that problem so it
>> shouldn't be an issue.
>>
>> Leo Izen
>>
>> ---
>>   fftools/ffmpeg.c | 19 ++++++++++++++-----
>>   1 file changed, 14 insertions(+), 5 deletions(-)
>>
>> diff --git a/fftools/ffmpeg.c b/fftools/ffmpeg.c
>> index da4259a9a8..6e81716795 100644
>> --- a/fftools/ffmpeg.c
>> +++ b/fftools/ffmpeg.c
>> @@ -2045,12 +2045,14 @@ static void do_streamcopy(InputStream *ist,
>> OutputStream *ost, const AVPacket *p
> This will not apply as it has too many newlines
>
> [...]

What do you mean by "this will not apply"? Do you mean attempting to 
apply the patch fails? What do newline characters have anything to do 
with it?

Leo Izen