[FFmpeg-devel] [PATCH] png parser
Peter Holik
peter
Mon Jun 8 13:54:39 CEST 2009
>> On Thu, Jun 04, 2009 at 10:40:55AM +0200, Peter Holik wrote:
>>> > On Mon, Jun 01, 2009 at 12:27:52PM +0200, Peter Holik wrote:
>>> >> Hi!
>>> >>
>>> >> I've now reworked my patch to really parse a png for end of frame.
>>> >>
>>> >> Now you can filter out png's of any stream like
>>> >>
>>> >> cat garbage a.png garbage b.png garbage | ffmpeg -f image2pipe -vcodec png -i - c%d.png
>>> >>
>>> >> cu Peter
>>> >> Makefile | 1
>>> >> allcodecs.c | 1
>>> >> png_parser.c | 124 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 3 files
> changed, 126 insertions(+)
>>> >> 63f7cd1bb826726f1d2a9e7254410cd7ec98a81c png-parser.patch
>>> >> diff --git a/libavcodec/Makefile b/libavcodec/Makefile
>>> >> index 14ef9f4..189231b 100644
>>> >> --- a/libavcodec/Makefile
>>> >> +++ b/libavcodec/Makefile
>>> >> @@ -411,6 +411,7 @@ OBJS-$(CONFIG_MLP_PARSER) += mlp_parser.o mlp.o
>>> >> OBJS-$(CONFIG_MPEG4VIDEO_PARSER) += mpeg4video_parser.o h263.o mpeg12data.o
>>> mpegvideo.o
>>> >> error_resilience.o
>>> >> OBJS-$(CONFIG_MPEGAUDIO_PARSER) += mpegaudio_parser.o mpegaudiodecheader.o
>>> >> mpegaudiodata.o
>>> >> OBJS-$(CONFIG_MPEGVIDEO_PARSER) += mpegvideo_parser.o mpeg12.o mpeg12data.o
>>> mpegvideo.o
>>> >> error_resilience.o
>>> >> +OBJS-$(CONFIG_PNG_PARSER) += png_parser.o
>>> >> OBJS-$(CONFIG_PNM_PARSER) += pnm_parser.o pnm.o OBJS-$(CONFIG_VC1_PARSER)
> += vc1_parser.o
>>> >> OBJS-$(CONFIG_VP3_PARSER) += vp3_parser.o
>>> >> diff --git a/libavcodec/allcodecs.c b/libavcodec/allcodecs.c index e8d5d27..362021f 100644
>>> >> --- a/libavcodec/allcodecs.c
>>> >> +++ b/libavcodec/allcodecs.c
>>> >> @@ -334,6 +334,7 @@ void avcodec_register_all(void)
>>> >> REGISTER_PARSER (MPEG4VIDEO, mpeg4video);
>>> >> REGISTER_PARSER (MPEGAUDIO, mpegaudio);
>>> >> REGISTER_PARSER (MPEGVIDEO, mpegvideo);
>>> >> + REGISTER_PARSER (PNG, png);
>>> >> REGISTER_PARSER (PNM, pnm);
>>> >> REGISTER_PARSER (VC1, vc1);
>>> >> REGISTER_PARSER (VP3, vp3);
>>> >> diff --git a/libavcodec/png_parser.c b/libavcodec/png_parser.c new file mode 100644
>>> >> index 0000000..64f2356
>>> >> --- /dev/null
>>> >> +++ b/libavcodec/png_parser.c
>>> >> @@ -0,0 +1,124 @@
>>> >> +/*
>>> >> + * PNG parser
>>> >> + * Copyright (c) 2009 Peter Holik
>>> >> + *
>>> >> + * This file is part of FFmpeg.
>>> >> + *
>>> >> + * FFmpeg is free software; you can redistribute it and/or
>>> >> + * modify it under the terms of the GNU Lesser General Public + * License as published by
> the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any
> later version. + *
>>> >> + * FFmpeg is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY;
> without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR
> PURPOSE. See the GNU + * Lesser General Public License for more details.
>>> >> + *
>>> >> + * You should have received a copy of the GNU Lesser General Public + * License along with
> FFmpeg; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth
> Floor, Boston, MA 02110-1301 USA + */
>>> >> +
>>> >> +/**
>>> >> + * @file libavcodec/png_parser.c
>>> >> + * PNG parser
>>> >> + */
>>> >> +
>>> >> +#include "parser.h"
>>> >> +#include "bytestream.h"
>>> >> +#include "png.h"
>>> >> +
>>> >> +static int png_parse(AVCodecParserContext *s, AVCodecContext *avctx, +
> const uint8_t **poutbuf, int *poutbuf_size, + const uint8_t *buf, int
> buf_size)
>>> >> +{
>>> >> + ParseContext *pc = s->priv_data;
>>> >> + uint8_t *bytestream;
>>> >> + uint8_t *bytestream_start;
>>> >> + uint8_t *bytestream_end;
>>> >> +
>>> >> + /* EOF considered as end of frame */
>>> >> + if (buf_size == 0)
>>> >> + goto flush;
>>> >> +
>>> >> + *poutbuf = NULL;
>>> >> + *poutbuf_size = 0;
>>> >> +
>>> >> + /* first add new buffer to parsebuffer */
>>> >> + bytestream_start = av_fast_realloc(pc->buffer, &pc->buffer_size, +
> pc->index + buf_size + FF_INPUT_BUFFER_PADDING_SIZE); + if(!bytestream_start)
>>> >> + goto fail;
>>> >> + pc->buffer = bytestream_start;
>>> >> + memcpy(&pc->buffer[pc->index], buf, buf_size);
>>> >
>>> > mandatory copying is unacceptable
>>> > the case where each input is exactly a png frame for example should be free of any
> timeconsuming operation
>>> >
>>> > maybe ff_combine_frame() would be usefull
>>>
>>> But ff_combine_frame also does memcpy, so there is no difference to my approach.
>>
>> ff_combine_frame() does not always memcpy, your code does always
>
> change my patch to only memcpy when input is not exactly a png frame.
>
> cu Peter
ping
More information about the ffmpeg-devel
mailing list