/* * Amuse Graphics Movie decoder * * Copyright (c) 2018 Paul B Mahol * * 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 */ #include #include #include #define BITSTREAM_READER_LE #include "avcodec.h" #include "bytestream.h" #include "copy_block.h" #include "get_bits.h" #include "idctdsp.h" #include "internal.h" static const uint8_t unscaled_luma[64] = { 16, 11, 10, 16, 24, 40, 51, 61, 12, 12, 14, 19, 26, 58, 60, 55, 14, 13, 16, 24, 40, 57, 69, 56, 14, 17, 22, 29, 51, 87, 80, 62, 18, 22, 37, 56, 68,109,103, 77, 24, 35, 55, 64, 81,104,113, 92, 49, 64, 78, 87,103,121,120,101, 72, 92, 95, 98, 112,100,103,99 }; static const uint8_t unscaled_chroma[64] = { 17, 18, 24, 47, 99, 99, 99, 99, 18, 21, 26, 66, 99, 99, 99, 99, 24, 26, 56, 99, 99, 99, 99, 99, 47, 66, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99, 99 }; typedef struct MotionVector { int16_t x, y; } MotionVector; typedef struct AGMContext { const AVClass *class; AVCodecContext *avctx; GetBitContext gb; GetByteContext gbyte; int key_frame; int bitstream_size; int compression; int blocks_w; int blocks_h; int size[3]; int plus; int dct; int rgb; unsigned flags; unsigned fflags; uint8_t *output; unsigned padded_output_size; unsigned output_size; MotionVector *mvectors; unsigned mvectors_size; VLC vlc; AVFrame *prev_frame; int luma_quant_matrix[64]; int chroma_quant_matrix[64]; ScanTable scantable; DECLARE_ALIGNED(32, int16_t, block)[64]; int16_t *wblocks; unsigned wblocks_size; int *map; unsigned map_size; IDCTDSPContext idsp; } AGMContext; static int read_code(GetBitContext *gb, int *oskip, int *level, int *map, int mode) { int len = 0, skip = 0, max; if (show_bits(gb, 2)) { switch (show_bits(gb, 4)) { case 1: case 9: len = 1; skip = 3; break; case 2: len = 3; skip = 4; break; case 3: len = 7; skip = 4; break; case 5: case 13: len = 2; skip = 3; break; case 6: len = 4; skip = 4; break; case 7: len = 8; skip = 4; break; case 10: len = 5; skip = 4; break; case 11: len = 9; skip = 4; break; case 14: len = 6; skip = 4; break; case 15: len = ((show_bits(gb, 5) & 0x10) | 0xA0) >> 4; skip = 5; break; default: return AVERROR_INVALIDDATA; } skip_bits(gb, skip); *level = get_bits(gb, len); *map = 1; *oskip = 0; max = 1 << (len - 1); if (*level < max) *level = -(max + *level); } else if (show_bits(gb, 3) & 4) { skip_bits(gb, 3); if (mode == 1) { if (show_bits(gb, 4)) { if (show_bits(gb, 4) == 1) { skip_bits(gb, 4); *oskip = get_bits(gb, 16); } else { *oskip = get_bits(gb, 4); } } else { skip_bits(gb, 4); *oskip = get_bits(gb, 10); } } else if (mode == 0) { *oskip = get_bits(gb, 10); } *level = 0; } else { skip_bits(gb, 3); if (mode == 0) *oskip = get_bits(gb, 4); else if (mode == 1) *oskip = 0; *level = 0; } return 0; } static int decode_intra_blocks(AGMContext *s, GetBitContext *gb, const int *quant_matrix, int *skip, int *dc_level) { const uint8_t *scantable = s->scantable.permutated; int level, ret, map = 0; memset(s->wblocks, 0, s->wblocks_size); for (int i = 0; i < 64; i++) { int16_t *block = s->wblocks + scantable[i]; for (int j = 0; j < s->blocks_w;) { if (*skip > 0) { int rskip; rskip = FFMIN(*skip, s->blocks_w - j); j += rskip; if (i == 0) { for (int k = 0; k < rskip; k++) block[64 * k] = *dc_level * quant_matrix[0]; } block += rskip * 64; *skip -= rskip; } else { ret = read_code(gb, skip, &level, &map, s->flags & 1); if (ret < 0) return ret; if (i == 0) *dc_level += level; block[0] = (i == 0 ? *dc_level : level) * quant_matrix[i]; block += 64; j++; } } } return 0; } static int decode_inter_blocks(AGMContext *s, GetBitContext *gb, const int *quant_matrix, int *skip, int *map) { const uint8_t *scantable = s->scantable.permutated; int level, ret; memset(s->wblocks, 0, s->wblocks_size); memset(s->map, 0, s->map_size); for (int i = 0; i < 64; i++) { int16_t *block = s->wblocks + scantable[i]; for (int j = 0; j < s->blocks_w;) { if (*skip > 0) { int rskip; rskip = FFMIN(*skip, s->blocks_w - j); j += rskip; block += rskip * 64; *skip -= rskip; } else { ret = read_code(gb, skip, &level, &map[j], s->flags & 1); if (ret < 0) return ret; block[0] = level * quant_matrix[i]; block += 64; j++; } } } return 0; } static int decode_intra_block(AGMContext *s, GetBitContext *gb, const int *quant_matrix, int *skip, int *dc_level) { const uint8_t *scantable = s->scantable.permutated; const int offset = s->plus ? 0 : 1024; int16_t *block = s->block; int level, ret, map = 0; memset(block, 0, sizeof(s->block)); if (*skip > 0) { (*skip)--; } else { ret = read_code(gb, skip, &level, &map, s->flags & 1); if (ret < 0) return ret; *dc_level += level; } block[scantable[0]] = offset + *dc_level * quant_matrix[0]; for (int i = 1; i < 64;) { if (*skip > 0) { int rskip; rskip = FFMIN(*skip, 64 - i); i += rskip; *skip -= rskip; } else { ret = read_code(gb, skip, &level, &map, s->flags & 1); if (ret < 0) return ret; block[scantable[i]] = level * quant_matrix[i]; i++; } } return 0; } static int decode_intra_plane(AGMContext *s, GetBitContext *gb, int size, const int *quant_matrix, AVFrame *frame, int plane) { int ret, skip = 0, dc_level = 0; const int offset = s->plus ? 0 : 1024; if ((ret = init_get_bits8(gb, s->gbyte.buffer, size)) < 0) return ret; if (s->flags & 1) { av_fast_padded_malloc(&s->wblocks, &s->wblocks_size, 64 * s->blocks_w * sizeof(*s->wblocks)); if (!s->wblocks) return AVERROR(ENOMEM); for (int y = 0; y < s->blocks_h; y++) { ret = decode_intra_blocks(s, gb, quant_matrix, &skip, &dc_level); if (ret < 0) return ret; for (int x = 0; x < s->blocks_w; x++) { s->wblocks[64 * x] += offset; s->idsp.idct_put(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane], s->wblocks + 64 * x); } } } else { for (int y = 0; y < s->blocks_h; y++) { for (int x = 0; x < s->blocks_w; x++) { ret = decode_intra_block(s, gb, quant_matrix, &skip, &dc_level); if (ret < 0) return ret; s->idsp.idct_put(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane], s->block); } } } align_get_bits(gb); if (get_bits_left(gb) < 0) av_log(s->avctx, AV_LOG_WARNING, "overread\n"); if (get_bits_left(gb) > 0) av_log(s->avctx, AV_LOG_WARNING, "underread: %d\n", get_bits_left(gb)); return 0; } static int decode_inter_block(AGMContext *s, GetBitContext *gb, const int *quant_matrix, int *skip, int *map) { const uint8_t *scantable = s->scantable.permutated; int16_t *block = s->block; int level, ret; memset(block, 0, sizeof(s->block)); for (int i = 0; i < 64;) { if (*skip > 0) { int rskip; rskip = FFMIN(*skip, 64 - i); i += rskip; *skip -= rskip; } else { ret = read_code(gb, skip, &level, map, s->flags & 1); if (ret < 0) return ret; block[scantable[i]] = level * quant_matrix[i]; i++; } } return 0; } static int decode_inter_plane(AGMContext *s, GetBitContext *gb, int size, const int *quant_matrix, AVFrame *frame, AVFrame *prev, int plane) { int ret, skip = 0; if ((ret = init_get_bits8(gb, s->gbyte.buffer, size)) < 0) return ret; if (s->flags == 3) { av_fast_padded_malloc(&s->wblocks, &s->wblocks_size, 64 * s->blocks_w * sizeof(*s->wblocks)); if (!s->wblocks) return AVERROR(ENOMEM); av_fast_padded_malloc(&s->map, &s->map_size, s->blocks_w * sizeof(*s->map)); if (!s->map) return AVERROR(ENOMEM); for (int y = 0; y < s->blocks_h; y++) { ret = decode_inter_blocks(s, gb, quant_matrix, &skip, s->map); if (ret < 0) return ret; for (int x = 0; x < s->blocks_w; x++) { int shift = plane == 0; int mvpos = (y >> shift) * (s->blocks_w >> shift) + (x >> shift); int orig_mv_x = s->mvectors[mvpos].x; int mv_x = s->mvectors[mvpos].x / (1 + !shift); int mv_y = s->mvectors[mvpos].y / (1 + !shift); int h = s->avctx->coded_height >> !shift; int w = s->avctx->coded_width >> !shift; int map = s->map[x]; if (orig_mv_x >= -32) { if (y * 8 + mv_y < 0 || y * 8 + mv_y >= h || x * 8 + mv_x < 0 || x * 8 + mv_x >= w) return AVERROR_INVALIDDATA; copy_block8(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, prev->data[plane] + ((s->blocks_h - 1 - y) * 8 - mv_y) * prev->linesize[plane] + (x * 8 + mv_x), frame->linesize[plane], prev->linesize[plane], 8); if (map) { s->idsp.idct(s->wblocks + x * 64); for (int i = 0; i < 64; i++) s->wblocks[i + x * 64] = (s->wblocks[i + x * 64] + 1) & 0xFFFC; s->idsp.add_pixels_clamped(&s->wblocks[x*64], frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane]); } } else if (map) { s->idsp.idct_put(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane], s->wblocks + x * 64); } } } } else if (s->flags & 2) { for (int y = 0; y < s->blocks_h; y++) { for (int x = 0; x < s->blocks_w; x++) { int shift = plane == 0; int mvpos = (y >> shift) * (s->blocks_w >> shift) + (x >> shift); int orig_mv_x = s->mvectors[mvpos].x; int mv_x = s->mvectors[mvpos].x / (1 + !shift); int mv_y = s->mvectors[mvpos].y / (1 + !shift); int h = s->avctx->coded_height >> !shift; int w = s->avctx->coded_width >> !shift; int map = 0; ret = decode_inter_block(s, gb, quant_matrix, &skip, &map); if (ret < 0) return ret; if (orig_mv_x >= -32) { if (y * 8 + mv_y < 0 || y * 8 + mv_y >= h || x * 8 + mv_x < 0 || x * 8 + mv_x >= w) return AVERROR_INVALIDDATA; copy_block8(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, prev->data[plane] + ((s->blocks_h - 1 - y) * 8 - mv_y) * prev->linesize[plane] + (x * 8 + mv_x), frame->linesize[plane], prev->linesize[plane], 8); if (map) { s->idsp.idct(s->block); for (int i = 0; i < 64; i++) s->block[i] = (s->block[i] + 1) & 0xFFFC; s->idsp.add_pixels_clamped(s->block, frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane]); } } else if (map) { s->idsp.idct_put(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane], s->block); } } } } else if (s->flags & 1) { av_fast_padded_malloc(&s->wblocks, &s->wblocks_size, 64 * s->blocks_w * sizeof(*s->wblocks)); if (!s->wblocks) return AVERROR(ENOMEM); av_fast_padded_malloc(&s->map, &s->map_size, s->blocks_w * sizeof(*s->map)); if (!s->map) return AVERROR(ENOMEM); for (int y = 0; y < s->blocks_h; y++) { ret = decode_inter_blocks(s, gb, quant_matrix, &skip, s->map); if (ret < 0) return ret; for (int x = 0; x < s->blocks_w; x++) { if (!s->map[x]) continue; s->idsp.idct_add(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane], s->wblocks + 64 * x); } } } else { for (int y = 0; y < s->blocks_h; y++) { for (int x = 0; x < s->blocks_w; x++) { int map = 0; ret = decode_inter_block(s, gb, quant_matrix, &skip, &map); if (ret < 0) return ret; if (!map) continue; s->idsp.idct_add(frame->data[plane] + (s->blocks_h - 1 - y) * 8 * frame->linesize[plane] + x * 8, frame->linesize[plane], s->block); } } } align_get_bits(gb); if (get_bits_left(gb) < 0) av_log(s->avctx, AV_LOG_WARNING, "overread\n"); if (get_bits_left(gb) > 0) av_log(s->avctx, AV_LOG_WARNING, "underread: %d\n", get_bits_left(gb)); return 0; } static void compute_quant_matrix(AGMContext *s, double qscale) { int luma[64], chroma[64]; double f = 1.0 - fabs(qscale); if (!s->key_frame && (s->flags & 2)) { if (qscale >= 0.0) { for (int i = 0; i < 64; i++) { luma[i] = FFMAX(1, 16 * f); chroma[i] = FFMAX(1, 16 * f); } } else { for (int i = 0; i < 64; i++) { luma[i] = FFMAX(1, 16 - qscale * 32); chroma[i] = FFMAX(1, 16 - qscale * 32); } } } else { if (qscale >= 0.0) { for (int i = 0; i < 64; i++) { luma[i] = FFMAX(1, unscaled_luma [(i & 7) * 8 + (i >> 3)] * f); chroma[i] = FFMAX(1, unscaled_chroma[(i & 7) * 8 + (i >> 3)] * f); } } else { for (int i = 0; i < 64; i++) { luma[i] = FFMAX(1, 255.0 - (255 - unscaled_luma [(i & 7) * 8 + (i >> 3)]) * f); chroma[i] = FFMAX(1, 255.0 - (255 - unscaled_chroma[(i & 7) * 8 + (i >> 3)]) * f); } } } for (int i = 0; i < 64; i++) { int pos = ff_zigzag_direct[i]; s->luma_quant_matrix[i] = luma[pos] * ((pos / 8) & 1 ? -1 : 1); s->chroma_quant_matrix[i] = chroma[pos] * ((pos / 8) & 1 ? -1 : 1); } } static int decode_raw_intra_rgb(AVCodecContext *avctx, GetByteContext *gbyte, AVFrame *frame) { uint8_t *dst = frame->data[0] + (avctx->height - 1) * frame->linesize[0]; uint8_t r = 0, g = 0, b = 0; for (int y = 0; y < avctx->height; y++) { for (int x = 0; x < avctx->width; x++) { dst[x*3+0] = bytestream2_get_byte(gbyte) + r; r = dst[x*3+0]; dst[x*3+1] = bytestream2_get_byte(gbyte) + g; g = dst[x*3+1]; dst[x*3+2] = bytestream2_get_byte(gbyte) + b; b = dst[x*3+2]; } dst -= frame->linesize[0]; } return 0; } static int fill_pixels(uint8_t **y0, uint8_t **y1, uint8_t **u, uint8_t **v, int ylinesize, int ulinesize, int vlinesize, uint8_t *fill, int *nx, int *ny, int *np, int w, int h) { uint8_t *y0dst = *y0; uint8_t *y1dst = *y1; uint8_t *udst = *u; uint8_t *vdst = *v; int x = *nx, y = *ny, pos = *np; if (pos == 0) { y0dst[2*x+0] += fill[0]; y0dst[2*x+1] += fill[1]; y1dst[2*x+0] += fill[2]; y1dst[2*x+1] += fill[3]; pos++; } else if (pos == 1) { udst[x] += fill[0]; vdst[x] += fill[1]; x++; if (x >= w) { x = 0; y++; if (y >= h) return 1; y0dst -= 2*ylinesize; y1dst -= 2*ylinesize; udst -= ulinesize; vdst -= vlinesize; } y0dst[2*x+0] += fill[2]; y0dst[2*x+1] += fill[3]; pos++; } else if (pos == 2) { y1dst[2*x+0] += fill[0]; y1dst[2*x+1] += fill[1]; udst[x] += fill[2]; vdst[x] += fill[3]; x++; if (x >= w) { x = 0; y++; if (y >= h) return 1; y0dst -= 2*ylinesize; y1dst -= 2*ylinesize; udst -= ulinesize; vdst -= vlinesize; } pos = 0; } *y0 = y0dst; *y1 = y1dst; *u = udst; *v = vdst; *np = pos; *nx = x; *ny = y; return 0; } static int decode_runlen_rgb(AVCodecContext *avctx, GetByteContext *gbyte, AVFrame *frame) { uint8_t *dst = frame->data[0] + (avctx->height - 1) * frame->linesize[0]; int runlen, y = 0, x = 0; uint8_t fill[4]; unsigned code; while (bytestream2_get_bytes_left(gbyte) > 0) { code = bytestream2_peek_le32(gbyte); runlen = code & 0xFFFFFF; if (code >> 24 == 0x77) { bytestream2_skip(gbyte, 4); for (int i = 0; i < 4; i++) fill[i] = bytestream2_get_byte(gbyte); while (runlen > 0) { runlen--; for (int i = 0; i < 4; i++) { dst[x] += fill[i]; x++; if (x >= frame->width * 3) { x = 0; y++; dst -= frame->linesize[0]; if (y >= frame->height) return 0; } } } } else { for (int i = 0; i < 4; i++) fill[i] = bytestream2_get_byte(gbyte); for (int i = 0; i < 4; i++) { dst[x] += fill[i]; x++; if (x >= frame->width * 3) { x = 0; y++; dst -= frame->linesize[0]; if (y >= frame->height) return 0; } } } } return 0; } static int decode_runlen(AVCodecContext *avctx, GetByteContext *gbyte, AVFrame *frame) { uint8_t *y0dst = frame->data[0] + (avctx->height - 1) * frame->linesize[0]; uint8_t *y1dst = y0dst - frame->linesize[0]; uint8_t *udst = frame->data[1] + ((avctx->height >> 1) - 1) * frame->linesize[1]; uint8_t *vdst = frame->data[2] + ((avctx->height >> 1) - 1) * frame->linesize[2]; int runlen, y = 0, x = 0, pos = 0; uint8_t fill[4]; unsigned code; while (bytestream2_get_bytes_left(gbyte) > 0) { code = bytestream2_peek_le32(gbyte); runlen = code & 0xFFFFFF; if (code >> 24 == 0x77) { bytestream2_skip(gbyte, 4); for (int i = 0; i < 4; i++) fill[i] = bytestream2_get_byte(gbyte); while (runlen > 0) { runlen--; if (fill_pixels(&y0dst, &y1dst, &udst, &vdst, frame->linesize[0], frame->linesize[1], frame->linesize[2], fill, &x, &y, &pos, avctx->width / 2, avctx->height / 2)) return 0; } } else { for (int i = 0; i < 4; i++) fill[i] = bytestream2_get_byte(gbyte); if (fill_pixels(&y0dst, &y1dst, &udst, &vdst, frame->linesize[0], frame->linesize[1], frame->linesize[2], fill, &x, &y, &pos, avctx->width / 2, avctx->height / 2)) return 0; } } return 0; } static int decode_raw_intra(AVCodecContext *avctx, GetByteContext *gbyte, AVFrame *frame) { uint8_t *y0dst = frame->data[0] + (avctx->height - 1) * frame->linesize[0]; uint8_t *y1dst = y0dst - frame->linesize[0]; uint8_t *udst = frame->data[1] + ((avctx->height >> 1) - 1) * frame->linesize[1]; uint8_t *vdst = frame->data[2] + ((avctx->height >> 1) - 1) * frame->linesize[2]; uint8_t ly0 = 0, ly1 = 0, ly2 = 0, ly3 = 0, lu = 0, lv = 0; for (int y = 0; y < avctx->height / 2; y++) { for (int x = 0; x < avctx->width / 2; x++) { y0dst[x*2+0] = bytestream2_get_byte(gbyte) + ly0; ly0 = y0dst[x*2+0]; y0dst[x*2+1] = bytestream2_get_byte(gbyte) + ly1; ly1 = y0dst[x*2+1]; y1dst[x*2+0] = bytestream2_get_byte(gbyte) + ly2; ly2 = y1dst[x*2+0]; y1dst[x*2+1] = bytestream2_get_byte(gbyte) + ly3; ly3 = y1dst[x*2+1]; udst[x] = bytestream2_get_byte(gbyte) + lu; lu = udst[x]; vdst[x] = bytestream2_get_byte(gbyte) + lv; lv = vdst[x]; } y0dst -= 2*frame->linesize[0]; y1dst -= 2*frame->linesize[0]; udst -= frame->linesize[1]; vdst -= frame->linesize[2]; } return 0; } static int decode_intra(AVCodecContext *avctx, GetBitContext *gb, AVFrame *frame) { AGMContext *s = avctx->priv_data; int ret; compute_quant_matrix(s, (2 * s->compression - 100) / 100.0); s->blocks_w = avctx->coded_width >> 3; s->blocks_h = avctx->coded_height >> 3; ret = decode_intra_plane(s, gb, s->size[0], s->luma_quant_matrix, frame, 0); if (ret < 0) return ret; bytestream2_skip(&s->gbyte, s->size[0]); s->blocks_w = avctx->coded_width >> 4; s->blocks_h = avctx->coded_height >> 4; ret = decode_intra_plane(s, gb, s->size[1], s->chroma_quant_matrix, frame, 2); if (ret < 0) return ret; bytestream2_skip(&s->gbyte, s->size[1]); s->blocks_w = avctx->coded_width >> 4; s->blocks_h = avctx->coded_height >> 4; ret = decode_intra_plane(s, gb, s->size[2], s->chroma_quant_matrix, frame, 1); if (ret < 0) return ret; return 0; } static int decode_motion_vectors(AVCodecContext *avctx, GetBitContext *gb) { AGMContext *s = avctx->priv_data; int nb_mvs = ((avctx->height + 15) >> 4) * ((avctx->width + 15) >> 4); int ret, skip = 0, value, map; av_fast_padded_malloc(&s->mvectors, &s->mvectors_size, nb_mvs * sizeof(*s->mvectors)); if (!s->mvectors) return AVERROR(ENOMEM); if ((ret = init_get_bits8(gb, s->gbyte.buffer, bytestream2_get_bytes_left(&s->gbyte) - (s->size[0] + s->size[1] + s->size[2]))) < 0) return ret; memset(s->mvectors, 0, sizeof(*s->mvectors) * nb_mvs); for (int i = 0; i < nb_mvs; i++) { ret = read_code(gb, &skip, &value, &map, 1); if (ret < 0) return ret; s->mvectors[i].x = value; i += skip; } for (int i = 0; i < nb_mvs; i++) { ret = read_code(gb, &skip, &value, &map, 1); if (ret < 0) return ret; s->mvectors[i].y = value; i += skip; } if (get_bits_left(gb) <= 0) return AVERROR_INVALIDDATA; skip = (get_bits_count(gb) >> 3) + 1; bytestream2_skip(&s->gbyte, skip); return 0; } static int decode_inter(AVCodecContext *avctx, GetBitContext *gb, AVFrame *frame, AVFrame *prev) { AGMContext *s = avctx->priv_data; int ret; compute_quant_matrix(s, (2 * s->compression - 100) / 100.0); if (s->flags & 2) { ret = decode_motion_vectors(avctx, gb); if (ret < 0) return ret; } s->blocks_w = avctx->coded_width >> 3; s->blocks_h = avctx->coded_height >> 3; ret = decode_inter_plane(s, gb, s->size[0], s->luma_quant_matrix, frame, prev, 0); if (ret < 0) return ret; bytestream2_skip(&s->gbyte, s->size[0]); s->blocks_w = avctx->coded_width >> 4; s->blocks_h = avctx->coded_height >> 4; ret = decode_inter_plane(s, gb, s->size[1], s->chroma_quant_matrix, frame, prev, 2); if (ret < 0) return ret; bytestream2_skip(&s->gbyte, s->size[1]); s->blocks_w = avctx->coded_width >> 4; s->blocks_h = avctx->coded_height >> 4; ret = decode_inter_plane(s, gb, s->size[2], s->chroma_quant_matrix, frame, prev, 1); if (ret < 0) return ret; return 0; } typedef struct Node { int parent; int child[2]; } Node; static void get_tree_codes(uint32_t *codes, Node *nodes, int idx, uint32_t pfx, int bitpos) { if (idx < 256 && idx >= 0) { codes[idx] = pfx; } else if (idx >= 0) { get_tree_codes(codes, nodes, nodes[idx].child[0], pfx + (0 << bitpos), bitpos + 1); get_tree_codes(codes, nodes, nodes[idx].child[1], pfx + (1 << bitpos), bitpos + 1); } } static int make_new_tree(const uint8_t *bitlens, uint32_t *codes) { int zlcount = 0, curlen, idx, nindex, last, llast; int blcounts[32] = { 0 }; int syms[8192]; Node nodes[512]; int node_idx[1024]; int old_idx[512]; for (int i = 0; i < 256; i++) { int bitlen = bitlens[i]; int blcount = blcounts[bitlen]; zlcount += bitlen < 1; syms[(bitlen << 8) + blcount] = i; blcounts[bitlen]++; } for (int i = 0; i < 512; i++) { nodes[i].child[0] = -1; nodes[i].child[1] = -1; } for (int i = 0; i < 256; i++) { node_idx[i] = 257 + i; } curlen = 1; node_idx[512] = 256; last = 255; nindex = 1; for (curlen = 1; curlen < 32; curlen++) { if (blcounts[curlen] > 0) { int max_zlcount = zlcount + blcounts[curlen]; for (int i = 0; zlcount < 256 && zlcount < max_zlcount; zlcount++, i++) { int p = node_idx[nindex - 1 + 512]; int ch = syms[256 * curlen + i]; if (nindex <= 0) return AVERROR_INVALIDDATA; if (nodes[p].child[0] == -1) { nodes[p].child[0] = ch; } else { nodes[p].child[1] = ch; nindex--; } nodes[ch].parent = p; } } llast = last - 1; idx = 0; while (nindex > 0) { int p, ch; last = llast - idx; p = node_idx[nindex - 1 + 512]; ch = node_idx[last]; if (nodes[p].child[0] == -1) { nodes[p].child[0] = ch; } else { nodes[p].child[1] = ch; nindex--; } old_idx[idx] = ch; nodes[ch].parent = p; if (idx == llast) goto next; idx++; if (nindex <= 0) { for (int i = 0; i < idx; i++) node_idx[512 + i] = old_idx[i]; } } nindex = idx; } next: get_tree_codes(codes, nodes, 256, 0, 0); return 0; } static int build_huff(const uint8_t *bitlen, VLC *vlc) { uint32_t new_codes[256]; uint8_t bits[256]; uint8_t symbols[256]; uint32_t codes[256]; int nb_codes = 0; int ret = make_new_tree(bitlen, new_codes); if (ret < 0) return ret; for (int i = 0; i < 256; i++) { if (bitlen[i]) { bits[nb_codes] = bitlen[i]; codes[nb_codes] = new_codes[i]; symbols[nb_codes] = i; nb_codes++; } } ff_free_vlc(vlc); return ff_init_vlc_sparse(vlc, 13, nb_codes, bits, 1, 1, codes, 4, 4, symbols, 1, 1, INIT_VLC_LE); } static int decode_huffman2(AVCodecContext *avctx, int header, int size) { AGMContext *s = avctx->priv_data; GetBitContext *gb = &s->gb; uint8_t lens[256]; int ret, x, len; if ((ret = init_get_bits8(gb, s->gbyte.buffer, bytestream2_get_bytes_left(&s->gbyte))) < 0) return ret; s->output_size = get_bits_long(gb, 32); av_fast_padded_malloc(&s->output, &s->padded_output_size, s->output_size); if (!s->output) return AVERROR(ENOMEM); x = get_bits(gb, 1); len = 4 + get_bits(gb, 1); if (x) { int cb[8] = { 0 }; int count = get_bits(gb, 3) + 1; for (int i = 0; i < count; i++) cb[i] = get_bits(gb, len); for (int i = 0; i < 256; i++) { int idx = get_bits(gb, 3); lens[i] = cb[idx]; } } else { for (int i = 0; i < 256; i++) lens[i] = get_bits(gb, len); } if ((ret = build_huff(lens, &s->vlc)) < 0) return ret; x = 0; while (get_bits_left(gb) > 0 && x < s->output_size) { int val = get_vlc2(gb, s->vlc.table, s->vlc.bits, 3); if (val < 0) return AVERROR_INVALIDDATA; s->output[x++] = val; } return 0; } static int decode_frame(AVCodecContext *avctx, void *data, int *got_frame, AVPacket *avpkt) { AGMContext *s = avctx->priv_data; GetBitContext *gb = &s->gb; GetByteContext *gbyte = &s->gbyte; AVFrame *frame = data; int w, h, width, height, header; unsigned compressed_size; long skip; int ret; if (!avpkt->size) return 0; bytestream2_init(gbyte, avpkt->data, avpkt->size); header = bytestream2_get_le32(gbyte); s->fflags = bytestream2_get_le32(gbyte); s->bitstream_size = s->fflags & 0x1FFFFFFF; s->fflags >>= 29; av_log(avctx, AV_LOG_DEBUG, "fflags: %X\n", s->fflags); if (avpkt->size < s->bitstream_size + 8) return AVERROR_INVALIDDATA; s->key_frame = (avpkt->flags & AV_PKT_FLAG_KEY); frame->key_frame = s->key_frame; frame->pict_type = s->key_frame ? AV_PICTURE_TYPE_I : AV_PICTURE_TYPE_P; if (header) { if (avctx->codec_tag == MKTAG('A', 'G', 'M', '0') || avctx->codec_tag == MKTAG('A', 'G', 'M', '1')) return AVERROR_PATCHWELCOME; else ret = decode_huffman2(avctx, header, (avpkt->size - s->bitstream_size) - 8); if (ret < 0) return ret; bytestream2_init(gbyte, s->output, s->output_size); } else if (!s->dct) { bytestream2_skip(gbyte, 4); } if (s->dct) { s->flags = 0; w = bytestream2_get_le32(gbyte); h = bytestream2_get_le32(gbyte); if (w == INT32_MIN || h == INT32_MIN) return AVERROR_INVALIDDATA; if (w < 0) { w = -w; s->flags |= 2; } if (h < 0) { h = -h; s->flags |= 1; } width = avctx->width; height = avctx->height; if (w < width || h < height || w & 7 || h & 7) return AVERROR_INVALIDDATA; ret = ff_set_dimensions(avctx, w, h); if (ret < 0) return ret; avctx->width = width; avctx->height = height; s->compression = bytestream2_get_le32(gbyte); if (s->compression < 0 || s->compression > 100) return AVERROR_INVALIDDATA; for (int i = 0; i < 3; i++) s->size[i] = bytestream2_get_le32(gbyte); if (header) { compressed_size = s->output_size; skip = 8LL; } else { compressed_size = avpkt->size; skip = 32LL; } if (s->size[0] < 0 || s->size[1] < 0 || s->size[2] < 0 || skip + s->size[0] + s->size[1] + s->size[2] > compressed_size) { return AVERROR_INVALIDDATA; } } if ((ret = ff_get_buffer(avctx, frame, AV_GET_BUFFER_FLAG_REF)) < 0) return ret; if (frame->key_frame) { if (!s->dct && !s->rgb) ret = decode_raw_intra(avctx, gbyte, frame); else if (!s->dct && s->rgb) ret = decode_raw_intra_rgb(avctx, gbyte, frame); else ret = decode_intra(avctx, gb, frame); } else { if (!s->prev_frame->data[0]) { av_log(avctx, AV_LOG_ERROR, "Missing reference frame.\n"); return AVERROR_INVALIDDATA; } if (s->prev_frame-> width != frame->width || s->prev_frame->height != frame->height) return AVERROR_INVALIDDATA; if (!(s->flags & 2)) { ret = av_frame_copy(frame, s->prev_frame); if (ret < 0) return ret; } if (s->dct) { ret = decode_inter(avctx, gb, frame, s->prev_frame); } else if (!s->dct && !s->rgb) { ret = decode_runlen(avctx, gbyte, frame); } else { ret = decode_runlen_rgb(avctx, gbyte, frame); } } if (ret < 0) return ret; av_frame_unref(s->prev_frame); if ((ret = av_frame_ref(s->prev_frame, frame)) < 0) return ret; frame->crop_top = avctx->coded_height - avctx->height; frame->crop_left = avctx->coded_width - avctx->width; *got_frame = 1; return avpkt->size; } static av_cold int decode_init(AVCodecContext *avctx) { AGMContext *s = avctx->priv_data; s->rgb = avctx->codec_tag == MKTAG('A', 'G', 'M', '4'); avctx->pix_fmt = s->rgb ? AV_PIX_FMT_BGR24 : AV_PIX_FMT_YUV420P; s->avctx = avctx; s->plus = avctx->codec_tag == MKTAG('A', 'G', 'M', '3') || avctx->codec_tag == MKTAG('A', 'G', 'M', '7'); s->dct = avctx->codec_tag != MKTAG('A', 'G', 'M', '4') && avctx->codec_tag != MKTAG('A', 'G', 'M', '5'); avctx->idct_algo = FF_IDCT_SIMPLE; ff_idctdsp_init(&s->idsp, avctx); ff_init_scantable(s->idsp.idct_permutation, &s->scantable, ff_zigzag_direct); s->prev_frame = av_frame_alloc(); if (!s->prev_frame) return AVERROR(ENOMEM); return 0; } static void decode_flush(AVCodecContext *avctx) { AGMContext *s = avctx->priv_data; av_frame_unref(s->prev_frame); } static av_cold int decode_close(AVCodecContext *avctx) { AGMContext *s = avctx->priv_data; ff_free_vlc(&s->vlc); av_frame_free(&s->prev_frame); av_freep(&s->mvectors); s->mvectors_size = 0; av_freep(&s->wblocks); s->wblocks_size = 0; av_freep(&s->output); s->padded_output_size = 0; av_freep(&s->map); s->map_size = 0; return 0; } AVCodec ff_agm_decoder = { .name = "agm", .long_name = NULL_IF_CONFIG_SMALL("Amuse Graphics Movie"), .type = AVMEDIA_TYPE_VIDEO, .id = AV_CODEC_ID_AGM, .priv_data_size = sizeof(AGMContext), .init = decode_init, .close = decode_close, .decode = decode_frame, .flush = decode_flush, .capabilities = AV_CODEC_CAP_DR1, .caps_internal = FF_CODEC_CAP_INIT_THREADSAFE | FF_CODEC_CAP_INIT_CLEANUP | FF_CODEC_CAP_EXPORTS_CROPPING, };