[FFmpeg-devel] avcodec/hapqa_extract_bsf : add bsf filter for haqqa (to hapq or hapalpha only) conversion
James Almer
jamrial at gmail.com
Sun Mar 11 20:33:54 EET 2018
On 3/11/2018 3:20 PM, Martin Vignali wrote:
> 2018-03-11 18:45 GMT+01:00 James Almer <jamrial at gmail.com>:
>
>> On 3/11/2018 2:08 PM, Martin Vignali wrote:
>>> Work for me, changed.
>>>
>>> New patchs in attach, with these changes.
>>
>> Please combine the last three patches (bsf, docs and changelog entry)
>> into a single one.
>>
>
> Changed
>
>>
>>> diff --git a/libavcodec/Makefile b/libavcodec/Makefile
>>> index ff6c9f8b2c..80e54f9e6d 100644
>>> --- a/libavcodec/Makefile
>>> +++ b/libavcodec/Makefile
>>> @@ -1040,6 +1040,7 @@ OBJS-$(CONFIG_DUMP_EXTRADATA_BSF) +=
>> dump_extradata_bsf.o
>>> OBJS-$(CONFIG_DCA_CORE_BSF) += dca_core_bsf.o
>>> OBJS-$(CONFIG_EXTRACT_EXTRADATA_BSF) += extract_extradata_bsf.o
>> \
>>> h2645_parse.o
>>> +OBJS-$(CONFIG_HAPQA_EXTRACT_BSF) += hapqa_extract_bsf.o
>>
>> Missing hap.o
>>
>
> Changed
>
>>
>> Also, HAPQA goes after H264.
>>
>
> Changed (also in bitstream_filters.c)
>
>
>>
>>> OBJS-$(CONFIG_H264_METADATA_BSF) += h264_metadata_bsf.o
>>> OBJS-$(CONFIG_H264_MP4TOANNEXB_BSF) += h264_mp4toannexb_bsf.o
>>> OBJS-$(CONFIG_H264_REDUNDANT_PPS_BSF) += h264_redundant_pps_bsf.o
>>
>>> +
>>> + av_packet_move_ref(out, in);
>>> + out->data += start_section_size;
>>> + out->size = target_packet_size;
>>> +
>>> + ret = av_packet_copy_props(out, in);
>>
>> No need to call av_packet_copy_props(). You moved the reference from
>> "in" to "out" right before this, which includes all properties, so "in"
>> is now an empty packet and av_packet_copy_props() will just reset all
>> properties from "out".
>>
>
> Thanks for the explanations
> Removed.
>
>
>>
>> You should add a new fate test using one of the existing samples to make
>> sure the output of this bsf is the expected one.
>>
>>
> I plan to add fate test later.
> I would like to apply the previous fate hap patch before.
>
> New patchs in attach
>
> Martin
Should be ok now.
More information about the ffmpeg-devel
mailing list