mirror of https://git.ffmpeg.org/ffmpeg.git
373 lines
11 KiB
C
373 lines
11 KiB
C
/*
|
|
* Electronic Arts TGV Video Decoder
|
|
* Copyright (c) 2007-2008 Peter Ross
|
|
*
|
|
* 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 St, Fifth Floor, Boston, MA 02110-1301 USA
|
|
*/
|
|
|
|
/**
|
|
* @file
|
|
* Electronic Arts TGV Video Decoder
|
|
* by Peter Ross (pross@xvid.org)
|
|
*
|
|
* Technical details here:
|
|
* http://wiki.multimedia.cx/index.php?title=Electronic_Arts_TGV
|
|
*/
|
|
|
|
#include "libavutil/imgutils.h"
|
|
#include "libavutil/mem.h"
|
|
|
|
#define BITSTREAM_READER_LE
|
|
#include "avcodec.h"
|
|
#include "get_bits.h"
|
|
#include "codec_internal.h"
|
|
#include "internal.h"
|
|
|
|
#define EA_PREAMBLE_SIZE 8
|
|
#define kVGT_TAG MKTAG('k', 'V', 'G', 'T')
|
|
|
|
typedef struct TgvContext {
|
|
AVCodecContext *avctx;
|
|
AVFrame *last_frame;
|
|
uint8_t *frame_buffer;
|
|
int width,height;
|
|
uint32_t palette[AVPALETTE_COUNT];
|
|
|
|
int (*mv_codebook)[2];
|
|
uint8_t (*block_codebook)[16];
|
|
int num_mvs; ///< current length of mv_codebook
|
|
int num_blocks_packed; ///< current length of block_codebook
|
|
} TgvContext;
|
|
|
|
static av_cold int tgv_decode_init(AVCodecContext *avctx)
|
|
{
|
|
TgvContext *s = avctx->priv_data;
|
|
s->avctx = avctx;
|
|
avctx->framerate = (AVRational){ 15, 1 };
|
|
avctx->pix_fmt = AV_PIX_FMT_PAL8;
|
|
|
|
s->last_frame = av_frame_alloc();
|
|
if (!s->last_frame)
|
|
return AVERROR(ENOMEM);
|
|
|
|
return 0;
|
|
}
|
|
|
|
/**
|
|
* Unpack buffer
|
|
* @return 0 on success, -1 on critical buffer underflow
|
|
*/
|
|
static int unpack(const uint8_t *src, const uint8_t *src_end,
|
|
uint8_t *dst, int width, int height)
|
|
{
|
|
uint8_t *dst_end = dst + width*height;
|
|
int size, size1, size2, offset, run;
|
|
uint8_t *dst_start = dst;
|
|
|
|
if (src[0] & 0x01)
|
|
src += 5;
|
|
else
|
|
src += 2;
|
|
|
|
if (src_end - src < 3)
|
|
return AVERROR_INVALIDDATA;
|
|
size = AV_RB24(src);
|
|
src += 3;
|
|
|
|
while (size > 0 && src < src_end) {
|
|
|
|
/* determine size1 and size2 */
|
|
size1 = (src[0] & 3);
|
|
if (src[0] & 0x80) { // 1
|
|
if (src[0] & 0x40 ) { // 11
|
|
if (src[0] & 0x20) { // 111
|
|
if (src[0] < 0xFC) // !(111111)
|
|
size1 = (((src[0] & 31) + 1) << 2);
|
|
src++;
|
|
size2 = 0;
|
|
} else { // 110
|
|
offset = ((src[0] & 0x10) << 12) + AV_RB16(&src[1]) + 1;
|
|
size2 = ((src[0] & 0xC) << 6) + src[3] + 5;
|
|
src += 4;
|
|
}
|
|
} else { // 10
|
|
size1 = ((src[1] & 0xC0) >> 6);
|
|
offset = (AV_RB16(&src[1]) & 0x3FFF) + 1;
|
|
size2 = (src[0] & 0x3F) + 4;
|
|
src += 3;
|
|
}
|
|
} else { // 0
|
|
offset = ((src[0] & 0x60) << 3) + src[1] + 1;
|
|
size2 = ((src[0] & 0x1C) >> 2) + 3;
|
|
src += 2;
|
|
}
|
|
|
|
|
|
/* fetch strip from src */
|
|
if (size1 > src_end - src)
|
|
break;
|
|
|
|
if (size1 > 0) {
|
|
size -= size1;
|
|
run = FFMIN(size1, dst_end - dst);
|
|
memcpy(dst, src, run);
|
|
dst += run;
|
|
src += run;
|
|
}
|
|
|
|
if (size2 > 0) {
|
|
if (dst - dst_start < offset)
|
|
return 0;
|
|
size -= size2;
|
|
run = FFMIN(size2, dst_end - dst);
|
|
av_memcpy_backptr(dst, offset, run);
|
|
dst += run;
|
|
}
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
/**
|
|
* Decode inter-frame
|
|
* @return 0 on success, -1 on critical buffer underflow
|
|
*/
|
|
static int tgv_decode_inter(TgvContext *s, AVFrame *frame,
|
|
const uint8_t *buf, const uint8_t *buf_end)
|
|
{
|
|
int num_mvs;
|
|
int num_blocks_raw;
|
|
int num_blocks_packed;
|
|
int vector_bits;
|
|
int i,j,x,y;
|
|
GetBitContext gb;
|
|
int mvbits;
|
|
const uint8_t *blocks_raw;
|
|
|
|
if(buf_end - buf < 12)
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
num_mvs = AV_RL16(&buf[0]);
|
|
num_blocks_raw = AV_RL16(&buf[2]);
|
|
num_blocks_packed = AV_RL16(&buf[4]);
|
|
vector_bits = AV_RL16(&buf[6]);
|
|
buf += 12;
|
|
|
|
if (vector_bits > MIN_CACHE_BITS || !vector_bits) {
|
|
av_log(s->avctx, AV_LOG_ERROR,
|
|
"Invalid value for motion vector bits: %d\n", vector_bits);
|
|
return AVERROR_INVALIDDATA;
|
|
}
|
|
|
|
/* allocate codebook buffers as necessary */
|
|
if (num_mvs > s->num_mvs) {
|
|
int err = av_reallocp_array(&s->mv_codebook, num_mvs, sizeof(*s->mv_codebook));
|
|
if (err < 0) {
|
|
s->num_mvs = 0;
|
|
return err;
|
|
}
|
|
s->num_mvs = num_mvs;
|
|
}
|
|
|
|
if (num_blocks_packed > s->num_blocks_packed) {
|
|
int err;
|
|
if ((err = av_reallocp(&s->block_codebook, num_blocks_packed * 16)) < 0) {
|
|
s->num_blocks_packed = 0;
|
|
return err;
|
|
}
|
|
s->num_blocks_packed = num_blocks_packed;
|
|
}
|
|
|
|
/* read motion vectors */
|
|
mvbits = (num_mvs * 2 * 10 + 31) & ~31;
|
|
|
|
if (buf_end - buf < (mvbits>>3) + 16*num_blocks_raw + 8*num_blocks_packed)
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
init_get_bits(&gb, buf, mvbits);
|
|
for (i = 0; i < num_mvs; i++) {
|
|
s->mv_codebook[i][0] = get_sbits(&gb, 10);
|
|
s->mv_codebook[i][1] = get_sbits(&gb, 10);
|
|
}
|
|
buf += mvbits >> 3;
|
|
|
|
/* note ptr to uncompressed blocks */
|
|
blocks_raw = buf;
|
|
buf += num_blocks_raw * 16;
|
|
|
|
/* read compressed blocks */
|
|
init_get_bits(&gb, buf, (buf_end - buf) << 3);
|
|
for (i = 0; i < num_blocks_packed; i++) {
|
|
int tmp[4];
|
|
for (j = 0; j < 4; j++)
|
|
tmp[j] = get_bits(&gb, 8);
|
|
for (j = 0; j < 16; j++)
|
|
s->block_codebook[i][15-j] = tmp[get_bits(&gb, 2)];
|
|
}
|
|
|
|
if (get_bits_left(&gb) < vector_bits *
|
|
(s->avctx->height / 4) * (s->avctx->width / 4))
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
/* read vectors and build frame */
|
|
for (y = 0; y < s->avctx->height / 4; y++)
|
|
for (x = 0; x < s->avctx->width / 4; x++) {
|
|
unsigned int vector = get_bits(&gb, vector_bits);
|
|
const uint8_t *src;
|
|
ptrdiff_t src_stride;
|
|
|
|
if (vector < num_mvs) {
|
|
int mx = x * 4 + s->mv_codebook[vector][0];
|
|
int my = y * 4 + s->mv_codebook[vector][1];
|
|
|
|
if (mx < 0 || mx + 4 > s->avctx->width ||
|
|
my < 0 || my + 4 > s->avctx->height) {
|
|
av_log(s->avctx, AV_LOG_ERROR, "MV %d %d out of picture\n", mx, my);
|
|
continue;
|
|
}
|
|
|
|
src = s->last_frame->data[0] + mx + my * s->last_frame->linesize[0];
|
|
src_stride = s->last_frame->linesize[0];
|
|
} else {
|
|
int offset = vector - num_mvs;
|
|
if (offset < num_blocks_raw)
|
|
src = blocks_raw + 16*offset;
|
|
else if (offset - num_blocks_raw < num_blocks_packed)
|
|
src = s->block_codebook[offset - num_blocks_raw];
|
|
else
|
|
continue;
|
|
src_stride = 4;
|
|
}
|
|
|
|
for (j = 0; j < 4; j++)
|
|
for (i = 0; i < 4; i++)
|
|
frame->data[0][(y * 4 + j) * frame->linesize[0] + (x * 4 + i)] =
|
|
src[j * src_stride + i];
|
|
}
|
|
|
|
return 0;
|
|
}
|
|
|
|
static int tgv_decode_frame(AVCodecContext *avctx, AVFrame *frame,
|
|
int *got_frame, AVPacket *avpkt)
|
|
{
|
|
const uint8_t *buf = avpkt->data;
|
|
int buf_size = avpkt->size;
|
|
TgvContext *s = avctx->priv_data;
|
|
const uint8_t *buf_end = buf + buf_size;
|
|
int chunk_type, ret;
|
|
|
|
if (buf_end - buf < EA_PREAMBLE_SIZE)
|
|
return AVERROR_INVALIDDATA;
|
|
|
|
chunk_type = AV_RL32(&buf[0]);
|
|
buf += EA_PREAMBLE_SIZE;
|
|
|
|
if (chunk_type == kVGT_TAG) {
|
|
int pal_count, i;
|
|
if(buf_end - buf < 12) {
|
|
av_log(avctx, AV_LOG_WARNING, "truncated header\n");
|
|
return AVERROR_INVALIDDATA;
|
|
}
|
|
|
|
s->width = AV_RL16(&buf[0]);
|
|
s->height = AV_RL16(&buf[2]);
|
|
if (s->avctx->width != s->width || s->avctx->height != s->height) {
|
|
av_freep(&s->frame_buffer);
|
|
av_frame_unref(s->last_frame);
|
|
if ((ret = ff_set_dimensions(s->avctx, s->width, s->height)) < 0)
|
|
return ret;
|
|
}
|
|
|
|
pal_count = AV_RL16(&buf[6]);
|
|
buf += 12;
|
|
for(i = 0; i < pal_count && i < AVPALETTE_COUNT && buf_end - buf >= 3; i++) {
|
|
s->palette[i] = 0xFFU << 24 | AV_RB24(buf);
|
|
buf += 3;
|
|
}
|
|
if (buf_end - buf < 5) {
|
|
return AVERROR_INVALIDDATA;
|
|
}
|
|
}
|
|
|
|
if ((ret = ff_get_buffer(avctx, frame, AV_GET_BUFFER_FLAG_REF)) < 0)
|
|
return ret;
|
|
|
|
memcpy(frame->data[1], s->palette, AVPALETTE_SIZE);
|
|
|
|
if (chunk_type == kVGT_TAG) {
|
|
int y;
|
|
frame->key_frame = 1;
|
|
frame->pict_type = AV_PICTURE_TYPE_I;
|
|
|
|
if (!s->frame_buffer &&
|
|
!(s->frame_buffer = av_mallocz(s->width * s->height)))
|
|
return AVERROR(ENOMEM);
|
|
|
|
if (unpack(buf, buf_end, s->frame_buffer, s->avctx->width, s->avctx->height) < 0) {
|
|
av_log(avctx, AV_LOG_WARNING, "truncated intra frame\n");
|
|
return AVERROR_INVALIDDATA;
|
|
}
|
|
for (y = 0; y < s->height; y++)
|
|
memcpy(frame->data[0] + y * frame->linesize[0],
|
|
s->frame_buffer + y * s->width,
|
|
s->width);
|
|
} else {
|
|
if (!s->last_frame->data[0]) {
|
|
av_log(avctx, AV_LOG_WARNING, "inter frame without corresponding intra frame\n");
|
|
return buf_size;
|
|
}
|
|
frame->key_frame = 0;
|
|
frame->pict_type = AV_PICTURE_TYPE_P;
|
|
if (tgv_decode_inter(s, frame, buf, buf_end) < 0) {
|
|
av_log(avctx, AV_LOG_WARNING, "truncated inter frame\n");
|
|
return AVERROR_INVALIDDATA;
|
|
}
|
|
}
|
|
|
|
av_frame_unref(s->last_frame);
|
|
if ((ret = av_frame_ref(s->last_frame, frame)) < 0)
|
|
return ret;
|
|
|
|
*got_frame = 1;
|
|
|
|
return buf_size;
|
|
}
|
|
|
|
static av_cold int tgv_decode_end(AVCodecContext *avctx)
|
|
{
|
|
TgvContext *s = avctx->priv_data;
|
|
av_frame_free(&s->last_frame);
|
|
av_freep(&s->frame_buffer);
|
|
av_freep(&s->mv_codebook);
|
|
av_freep(&s->block_codebook);
|
|
return 0;
|
|
}
|
|
|
|
const FFCodec ff_eatgv_decoder = {
|
|
.p.name = "eatgv",
|
|
.p.long_name = NULL_IF_CONFIG_SMALL("Electronic Arts TGV video"),
|
|
.p.type = AVMEDIA_TYPE_VIDEO,
|
|
.p.id = AV_CODEC_ID_TGV,
|
|
.priv_data_size = sizeof(TgvContext),
|
|
.init = tgv_decode_init,
|
|
.close = tgv_decode_end,
|
|
FF_CODEC_DECODE_CB(tgv_decode_frame),
|
|
.p.capabilities = AV_CODEC_CAP_DR1,
|
|
.caps_internal = FF_CODEC_CAP_INIT_THREADSAFE,
|
|
};
|