pgssubdec.c 14.1 KB
Newer Older
1 2 3 4
/*
 * PGS subtitle decoder
 * Copyright (c) 2009 Stephen Backway
 *
5
 * This file is part of Libav.
6
 *
7
 * Libav is free software; you can redistribute it and/or
8 9 10 11
 * 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.
 *
12
 * Libav is distributed in the hope that it will be useful,
13 14 15 16 17
 * 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
18
 * License along with Libav; if not, write to the Free Software
19 20 21 22
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
23
 * @file
24 25 26 27 28 29
 * PGS subtitle decoder
 */

#include "avcodec.h"
#include "dsputil.h"
#include "bytestream.h"
30
#include "libavutil/colorspace.h"
31
#include "libavutil/imgutils.h"
32 33 34 35 36 37 38 39 40 41 42 43 44 45 46

#define RGBA(r,g,b,a) (((a) << 24) | ((r) << 16) | ((g) << 8) | (b))

enum SegmentType {
    PALETTE_SEGMENT      = 0x14,
    PICTURE_SEGMENT      = 0x15,
    PRESENTATION_SEGMENT = 0x16,
    WINDOW_SEGMENT       = 0x17,
    DISPLAY_SEGMENT      = 0x80,
};

typedef struct PGSSubPresentation {
    int x;
    int y;
    int id_number;
47
    int object_number;
48 49 50 51 52 53 54
} PGSSubPresentation;

typedef struct PGSSubPicture {
    int          w;
    int          h;
    uint8_t      *rle;
    unsigned int rle_buffer_size, rle_data_len;
55
    unsigned int rle_remaining_len;
56 57 58 59 60 61 62 63 64 65
} PGSSubPicture;

typedef struct PGSSubContext {
    PGSSubPresentation presentation;
    uint32_t           clut[256];
    PGSSubPicture      picture;
} PGSSubContext;

static av_cold int init_decoder(AVCodecContext *avctx)
{
66
    avctx->pix_fmt = PIX_FMT_PAL8;
67 68 69 70 71 72 73 74 75 76 77 78 79 80 81

    return 0;
}

static av_cold int close_decoder(AVCodecContext *avctx)
{
    PGSSubContext *ctx = avctx->priv_data;

    av_freep(&ctx->picture.rle);
    ctx->picture.rle_buffer_size  = 0;

    return 0;
}

/**
82
 * Decode the RLE data.
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136
 *
 * The subtitle is stored as an Run Length Encoded image.
 *
 * @param avctx contains the current codec context
 * @param sub pointer to the processed subtitle data
 * @param buf pointer to the RLE data to process
 * @param buf_size size of the RLE data to process
 */
static int decode_rle(AVCodecContext *avctx, AVSubtitle *sub,
                      const uint8_t *buf, unsigned int buf_size)
{
    const uint8_t *rle_bitmap_end;
    int pixel_count, line_count;

    rle_bitmap_end = buf + buf_size;

    sub->rects[0]->pict.data[0] = av_malloc(sub->rects[0]->w * sub->rects[0]->h);

    if (!sub->rects[0]->pict.data[0])
        return -1;

    pixel_count = 0;
    line_count  = 0;

    while (buf < rle_bitmap_end && line_count < sub->rects[0]->h) {
        uint8_t flags, color;
        int run;

        color = bytestream_get_byte(&buf);
        run   = 1;

        if (color == 0x00) {
            flags = bytestream_get_byte(&buf);
            run   = flags & 0x3f;
            if (flags & 0x40)
                run = (run << 8) + bytestream_get_byte(&buf);
            color = flags & 0x80 ? bytestream_get_byte(&buf) : 0;
        }

        if (run > 0 && pixel_count + run <= sub->rects[0]->w * sub->rects[0]->h) {
            memset(sub->rects[0]->pict.data[0] + pixel_count, color, run);
            pixel_count += run;
        } else if (!run) {
            /*
             * New Line. Check if correct pixels decoded, if not display warning
             * and adjust bitmap pointer to correct new line position.
             */
            if (pixel_count % sub->rects[0]->w > 0)
                av_log(avctx, AV_LOG_ERROR, "Decoded %d pixels, when line should be %d pixels\n",
                       pixel_count % sub->rects[0]->w, sub->rects[0]->w);
            line_count++;
        }
    }

137 138 139 140 141
    if (pixel_count < sub->rects[0]->w * sub->rects[0]->h) {
        av_log(avctx, AV_LOG_ERROR, "Insufficient RLE data for subtitle\n");
        return -1;
    }

142
    av_dlog(avctx, "Pixel Count = %d, Area = %d\n", pixel_count, sub->rects[0]->w * sub->rects[0]->h);
143 144 145 146 147

    return 0;
}

/**
148
 * Parse the picture segment packet.
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165
 *
 * The picture segment contains details on the sequence id,
 * width, height and Run Length Encoded (RLE) bitmap data.
 *
 * @param avctx contains the current codec context
 * @param buf pointer to the packet to process
 * @param buf_size size of packet to process
 * @todo TODO: Enable support for RLE data over multiple packets
 */
static int parse_picture_segment(AVCodecContext *avctx,
                                  const uint8_t *buf, int buf_size)
{
    PGSSubContext *ctx = avctx->priv_data;

    uint8_t sequence_desc;
    unsigned int rle_bitmap_len, width, height;

166 167 168 169
    if (buf_size <= 4)
        return -1;
    buf_size -= 4;

170 171 172 173 174 175 176
    /* skip 3 unknown bytes: Object ID (2 bytes), Version Number */
    buf += 3;

    /* Read the Sequence Description to determine if start of RLE data or appended to previous RLE */
    sequence_desc = bytestream_get_byte(&buf);

    if (!(sequence_desc & 0x80)) {
177 178 179
        /* Additional RLE data */
        if (buf_size > ctx->picture.rle_remaining_len)
            return -1;
180

181 182
        memcpy(ctx->picture.rle + ctx->picture.rle_data_len, buf, buf_size);
        ctx->picture.rle_data_len += buf_size;
183
        ctx->picture.rle_remaining_len -= buf_size;
184

185
        return 0;
186 187
    }

188 189 190 191 192 193
    if (buf_size <= 7)
        return -1;
    buf_size -= 7;

    /* Decode rle bitmap length, stored size includes width/height data */
    rle_bitmap_len = bytestream_get_be24(&buf) - 2*2;
194 195 196 197 198 199

    /* Get bitmap dimensions from data */
    width  = bytestream_get_be16(&buf);
    height = bytestream_get_be16(&buf);

    /* Make sure the bitmap is not too large */
200
    if (avctx->width < width || avctx->height < height) {
201
        av_log(avctx, AV_LOG_ERROR, "Bitmap dimensions larger than video.\n");
202 203 204 205 206 207 208 209 210 211 212
        return -1;
    }

    ctx->picture.w = width;
    ctx->picture.h = height;

    av_fast_malloc(&ctx->picture.rle, &ctx->picture.rle_buffer_size, rle_bitmap_len);

    if (!ctx->picture.rle)
        return -1;

213 214 215
    memcpy(ctx->picture.rle, buf, buf_size);
    ctx->picture.rle_data_len = buf_size;
    ctx->picture.rle_remaining_len = rle_bitmap_len - buf_size;
216 217 218 219 220

    return 0;
}

/**
221
 * Parse the palette segment packet.
222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247
 *
 * The palette segment contains details of the palette,
 * a maximum of 256 colors can be defined.
 *
 * @param avctx contains the current codec context
 * @param buf pointer to the packet to process
 * @param buf_size size of packet to process
 */
static void parse_palette_segment(AVCodecContext *avctx,
                                  const uint8_t *buf, int buf_size)
{
    PGSSubContext *ctx = avctx->priv_data;

    const uint8_t *buf_end = buf + buf_size;
    const uint8_t *cm      = ff_cropTbl + MAX_NEG_CROP;
    int color_id;
    int y, cb, cr, alpha;
    int r, g, b, r_add, g_add, b_add;

    /* Skip two null bytes */
    buf += 2;

    while (buf < buf_end) {
        color_id  = bytestream_get_byte(&buf);
        y         = bytestream_get_byte(&buf);
        cr        = bytestream_get_byte(&buf);
248
        cb        = bytestream_get_byte(&buf);
249 250 251 252 253
        alpha     = bytestream_get_byte(&buf);

        YUV_TO_RGB1(cb, cr);
        YUV_TO_RGB2(r, g, b, y);

254
        av_dlog(avctx, "Color %d := (%d,%d,%d,%d)\n", color_id, r, g, b, alpha);
255 256 257 258 259 260 261

        /* Store color in palette */
        ctx->clut[color_id] = RGBA(r,g,b,alpha);
    }
}

/**
262
 * Parse the presentation segment packet.
263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279
 *
 * The presentation segment contains details on the video
 * width, video height, x & y subtitle position.
 *
 * @param avctx contains the current codec context
 * @param buf pointer to the packet to process
 * @param buf_size size of packet to process
 * @todo TODO: Implement cropping
 * @todo TODO: Implement forcing of subtitles
 */
static void parse_presentation_segment(AVCodecContext *avctx,
                                       const uint8_t *buf, int buf_size)
{
    PGSSubContext *ctx = avctx->priv_data;

    int x, y;

280 281
    int w = bytestream_get_be16(&buf);
    int h = bytestream_get_be16(&buf);
282

283
    av_dlog(avctx, "Video Dimensions %dx%d\n",
284
            w, h);
285
    if (av_image_check_size(w, h, 0, avctx) >= 0)
286
        avcodec_set_dimensions(avctx, w, h);
287 288 289 290 291 292

    /* Skip 1 bytes of unknown, frame rate? */
    buf++;

    ctx->presentation.id_number = bytestream_get_be16(&buf);

Reimar Döffinger's avatar
Reimar Döffinger committed
293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317
    /*
     * Skip 3 bytes of unknown:
     *     state
     *     palette_update_flag (0x80),
     *     palette_id_to_use,
     */
    buf += 3;

    ctx->presentation.object_number = bytestream_get_byte(&buf);
    if (!ctx->presentation.object_number)
        return;

    /*
     * Skip 4 bytes of unknown:
     *     object_id_ref (2 bytes),
     *     window_id_ref,
     *     composition_flag (0x80 - object cropped, 0x40 - object forced)
     */
    buf += 4;

    x = bytestream_get_be16(&buf);
    y = bytestream_get_be16(&buf);

    /* TODO If cropping, cropping_x, cropping_y, cropping_width, cropping_height (all 2 bytes).*/

318
    av_dlog(avctx, "Subtitle Placement x=%d, y=%d\n", x, y);
Reimar Döffinger's avatar
Reimar Döffinger committed
319 320 321 322 323 324

    if (x > avctx->width || y > avctx->height) {
        av_log(avctx, AV_LOG_ERROR, "Subtitle out of video bounds. x = %d, y = %d, video width = %d, video height = %d.\n",
               x, y, avctx->width, avctx->height);
        x = 0; y = 0;
    }
325

Reimar Döffinger's avatar
Reimar Döffinger committed
326 327 328
    /* Fill in dimensions */
    ctx->presentation.x = x;
    ctx->presentation.y = y;
329 330 331
}

/**
332
 * Parse the display segment packet.
333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357
 *
 * The display segment controls the updating of the display.
 *
 * @param avctx contains the current codec context
 * @param data pointer to the data pertaining the subtitle to display
 * @param buf pointer to the packet to process
 * @param buf_size size of packet to process
 * @todo TODO: Fix start time, relies on correct PTS, currently too late
 *
 * @todo TODO: Fix end time, normally cleared by a second display
 * @todo       segment, which is currently ignored as it clears
 * @todo       the subtitle too early.
 */
static int display_end_segment(AVCodecContext *avctx, void *data,
                               const uint8_t *buf, int buf_size)
{
    AVSubtitle    *sub = data;
    PGSSubContext *ctx = avctx->priv_data;

    /*
     *      The end display time is a timeout value and is only reached
     *      if the next subtitle is later then timeout or subtitle has
     *      not been cleared by a subsequent empty display command.
     */

358
    memset(sub, 0, sizeof(*sub));
359 360 361 362
    // Blank if last object_number was 0.
    // Note that this may be wrong for more complex subtitles.
    if (!ctx->presentation.object_number)
        return 1;
363 364 365 366
    sub->start_display_time = 0;
    sub->end_display_time   = 20000;
    sub->format             = 0;

Reimar Döffinger's avatar
Reimar Döffinger committed
367 368 369
    sub->rects     = av_mallocz(sizeof(*sub->rects));
    sub->rects[0]  = av_mallocz(sizeof(*sub->rects[0]));
    sub->num_rects = 1;
370 371 372 373 374 375 376 377 378 379

    sub->rects[0]->x    = ctx->presentation.x;
    sub->rects[0]->y    = ctx->presentation.y;
    sub->rects[0]->w    = ctx->picture.w;
    sub->rects[0]->h    = ctx->picture.h;
    sub->rects[0]->type = SUBTITLE_BITMAP;

    /* Process bitmap */
    sub->rects[0]->pict.linesize[0] = ctx->picture.w;

380 381 382 383
    if (ctx->picture.rle) {
        if (ctx->picture.rle_remaining_len)
            av_log(avctx, AV_LOG_ERROR, "RLE data length %u is %u bytes shorter than expected\n",
                   ctx->picture.rle_data_len, ctx->picture.rle_remaining_len);
384 385
        if(decode_rle(avctx, sub, ctx->picture.rle, ctx->picture.rle_data_len) < 0)
            return 0;
386
    }
387 388
    /* Allocate memory for colors */
    sub->rects[0]->nb_colors    = 256;
389
    sub->rects[0]->pict.data[1] = av_mallocz(AVPALETTE_SIZE);
390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406

    memcpy(sub->rects[0]->pict.data[1], ctx->clut, sub->rects[0]->nb_colors * sizeof(uint32_t));

    return 1;
}

static int decode(AVCodecContext *avctx, void *data, int *data_size,
                  AVPacket *avpkt)
{
    const uint8_t *buf = avpkt->data;
    int buf_size       = avpkt->size;

    const uint8_t *buf_end;
    uint8_t       segment_type;
    int           segment_length;
    int i;

407
    av_dlog(avctx, "PGS sub packet:\n");
408 409

    for (i = 0; i < buf_size; i++) {
410
        av_dlog(avctx, "%02x ", buf[i]);
411
        if (i % 16 == 15)
412
            av_dlog(avctx, "\n");
413 414 415
    }

    if (i & 15)
416
        av_dlog(avctx, "\n");
417 418 419 420 421 422 423 424 425 426 427 428 429 430

    *data_size = 0;

    /* Ensure that we have received at a least a segment code and segment length */
    if (buf_size < 3)
        return -1;

    buf_end = buf + buf_size;

    /* Step through buffer to identify segments */
    while (buf < buf_end) {
        segment_type   = bytestream_get_byte(&buf);
        segment_length = bytestream_get_be16(&buf);

431
        av_dlog(avctx, "Segment Length %d, Segment Type %x\n", segment_length, segment_type);
432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470

        if (segment_type != DISPLAY_SEGMENT && segment_length > buf_end - buf)
            break;

        switch (segment_type) {
        case PALETTE_SEGMENT:
            parse_palette_segment(avctx, buf, segment_length);
            break;
        case PICTURE_SEGMENT:
            parse_picture_segment(avctx, buf, segment_length);
            break;
        case PRESENTATION_SEGMENT:
            parse_presentation_segment(avctx, buf, segment_length);
            break;
        case WINDOW_SEGMENT:
            /*
             * Window Segment Structure (No new information provided):
             *     2 bytes: Unkown,
             *     2 bytes: X position of subtitle,
             *     2 bytes: Y position of subtitle,
             *     2 bytes: Width of subtitle,
             *     2 bytes: Height of subtitle.
             */
            break;
        case DISPLAY_SEGMENT:
            *data_size = display_end_segment(avctx, data, buf, segment_length);
            break;
        default:
            av_log(avctx, AV_LOG_ERROR, "Unknown subtitle segment type 0x%x, length %d\n",
                   segment_type, segment_length);
            break;
        }

        buf += segment_length;
    }

    return buf_size;
}

471
AVCodec ff_pgssub_decoder = {
472 473 474 475 476 477 478
    .name           = "pgssub",
    .type           = AVMEDIA_TYPE_SUBTITLE,
    .id             = CODEC_ID_HDMV_PGS_SUBTITLE,
    .priv_data_size = sizeof(PGSSubContext),
    .init           = init_decoder,
    .close          = close_decoder,
    .decode         = decode,
479 480
    .long_name = NULL_IF_CONFIG_SMALL("HDMV Presentation Graphic Stream subtitles"),
};