libavcodec/srtdec.c

   1 /*
   2  * SubRip subtitle decoder
   3  * Copyright (c) 2010  Aurelien Jacobs <aurel@gnuage.org>
   4  *
   5  * This file is part of Libav.
   6  *
   7  * Libav is free software; you can redistribute it and/or
   8  * modify it under the terms of the GNU Lesser General Public
   9  * License as published by the Free Software Foundation; either
  10  * version 2.1 of the License, or (at your option) any later version.
  11  *
  12  * Libav is distributed in the hope that it will be useful,
  13  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  14  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  15  * Lesser General Public License for more details.
  16  *
  17  * 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  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  20  */
  21
  22 #include "libavutil/avstring.h"
  23 #include "libavutil/common.h"
  24 #include "libavutil/parseutils.h"
  25 #include "avcodec.h"
  26 #include "ass.h"
  27
  28 static int html_color_parse(AVCodecContext *avctx, const char *str)
  29 {
  30     uint8_t rgba[4];
  31     if (av_parse_color(rgba, str, strcspn(str, "\" >"), avctx) < 0)
  32         return -1;
  33     return rgba[0] | rgba[1] << 8 | rgba[2] << 16;
  34 }
  35
  36 enum {
  37     PARAM_UNKNOWN = -1,
  38     PARAM_SIZE,
  39     PARAM_COLOR,
  40     PARAM_FACE,
  41     PARAM_NUMBER
  42 };
  43
  44 typedef struct SrtStack {
  45     char tag[128];
  46     char param[PARAM_NUMBER][128];
  47 } SrtStack;
  48
  49 static const char *srt_to_ass(AVCodecContext *avctx, char *out, char *out_end,
  50                               const char *in, int x1, int y1, int x2, int y2)
  51 {
  52     char c, *param, buffer[128], tmp[128];
  53     int len, tag_close, sptr = 1, line_start = 1, an = 0, end = 0;
  54     SrtStack stack[16];
  55
  56     stack[0].tag[0] = 0;
  57     strcpy(stack[0].param[PARAM_SIZE],  "{\\fs}");
  58     strcpy(stack[0].param[PARAM_COLOR], "{\\c}");
  59     strcpy(stack[0].param[PARAM_FACE],  "{\\fn}");
  60
  61     if (x1 >= 0 && y1 >= 0) {
  62         if (x2 >= 0 && y2 >= 0 && (x2 != x1 || y2 != y1))
  63             out += snprintf(out, out_end-out,
  64                             "{\\an1}{\\move(%d,%d,%d,%d)}", x1, y1, x2, y2);
  65         else
  66             out += snprintf(out, out_end-out, "{\\an1}{\\pos(%d,%d)}", x1, y1);
  67     }
  68
  69     for (; out < out_end && !end && *in; in++) {
  70         switch (*in) {
  71         case '\r':
  72             break;
  73         case '\n':
  74             if (line_start) {
  75                 end = 1;
  76                 break;
  77             }
  78             while (out[-1] == ' ')
  79                 out--;
  80             out += snprintf(out, out_end-out, "\\N");
  81             line_start = 1;
  82             break;
  83         case ' ':
  84             if (!line_start)
  85                 *out++ = *in;
  86             break;
  87         case '{':    /* skip all {\xxx} substrings except for {\an%d}
  88                         and all microdvd like styles such as {Y:xxx} */
  89             an += sscanf(in, "{\\an%*1u}%c", &c) == 1;
  90             if ((an != 1 && sscanf(in, "{\\%*[^}]}%n%c", &len, &c) > 0) ||
  91                 sscanf(in, "{%*1[CcFfoPSsYy]:%*[^}]}%n%c", &len, &c) > 0) {
  92                 in += len - 1;
  93             } else
  94                 *out++ = *in;
  95             break;
  96         case '<':
  97             tag_close = in[1] == '/';
  98             if (sscanf(in+tag_close+1, "%127[^>]>%n%c", buffer, &len,&c) >= 2) {
  99                 if ((param = strchr(buffer, ' ')))
 100                     *param++ = 0;
 101                 if ((!tag_close && sptr < FF_ARRAY_ELEMS(stack)) ||
 102                     ( tag_close && sptr > 0 && !strcmp(stack[sptr-1].tag, buffer))) {
 103                     int i, j, unknown = 0;
 104                     in += len + tag_close;
 105                     if (!tag_close)
 106                         memset(stack+sptr, 0, sizeof(*stack));
 107                     if (!strcmp(buffer, "font")) {
 108                         if (tag_close) {
 109                             for (i=PARAM_NUMBER-1; i>=0; i--)
 110                                 if (stack[sptr-1].param[i][0])
 111                                     for (j=sptr-2; j>=0; j--)
 112                                         if (stack[j].param[i][0]) {
 113                                             out += snprintf(out, out_end-out,
 114                                                             "%s", stack[j].param[i]);
 115                                             break;
 116                                         }
 117                         } else {
 118                             while (param) {
 119                                 if (!strncmp(param, "size=", 5)) {
 120                                     unsigned font_size;
 121                                     param += 5 + (param[5] == '"');
 122                                     if (sscanf(param, "%u", &font_size) == 1) {
 123                                         snprintf(stack[sptr].param[PARAM_SIZE],
 124                                              sizeof(stack[0].param[PARAM_SIZE]),
 125                                              "{\\fs%u}", font_size);
 126                                     }
 127                                 } else if (!strncmp(param, "color=", 6)) {
 128                                     param += 6 + (param[6] == '"');
 129                                     snprintf(stack[sptr].param[PARAM_COLOR],
 130                                          sizeof(stack[0].param[PARAM_COLOR]),
 131                                          "{\\c&H%X&}",
 132                                          html_color_parse(avctx, param));
 133                                 } else if (!strncmp(param, "face=", 5)) {
 134                                     param += 5 + (param[5] == '"');
 135                                     len = strcspn(param,
 136                                                   param[-1] == '"' ? "\"" :" ");
 137                                     av_strlcpy(tmp, param,
 138                                                FFMIN(sizeof(tmp), len+1));
 139                                     param += len;
 140                                     snprintf(stack[sptr].param[PARAM_FACE],
 141                                              sizeof(stack[0].param[PARAM_FACE]),
 142                                              "{\\fn%s}", tmp);
 143                                 }
 144                                 if ((param = strchr(param, ' ')))
 145                                     param++;
 146                             }
 147                             for (i=0; i<PARAM_NUMBER; i++)
 148                                 if (stack[sptr].param[i][0])
 149                                     out += snprintf(out, out_end-out,
 150                                                     "%s", stack[sptr].param[i]);
 151                         }
 152                     } else if (!buffer[1] && strspn(buffer, "bisu") == 1) {
 153                         out += snprintf(out, out_end-out,
 154                                         "{\\%c%d}", buffer[0], !tag_close);
 155                     } else {
 156                         unknown = 1;
 157                         snprintf(tmp, sizeof(tmp), "</%s>", buffer);
 158                     }
 159                     if (tag_close) {
 160                         sptr--;
 161                     } else if (unknown && !strstr(in, tmp)) {
 162                         in -= len + tag_close;
 163                         *out++ = *in;
 164                     } else
 165                         av_strlcpy(stack[sptr++].tag, buffer,
 166                                    sizeof(stack[0].tag));
 167                     break;
 168                 }
 169             }
 170         default:
 171             *out++ = *in;
 172             break;
 173         }
 174         if (*in != ' ' && *in != '\r' && *in != '\n')
 175             line_start = 0;
 176     }
 177
 178     out = FFMIN(out, out_end-3);
 179     while (!strncmp(out-2, "\\N", 2))
 180         out -= 2;
 181     while (out[-1] == ' ')
 182         out--;
 183     out += snprintf(out, out_end-out, "\r\n");
 184     return in;
 185 }
 186
 187 static const char *read_ts(const char *buf, int *ts_start, int *ts_end,
 188                            int *x1, int *y1, int *x2, int *y2)
 189 {
 190     int i, hs, ms, ss, he, me, se;
 191
 192     for (i=0; i<2; i++) {
 193         /* try to read timestamps in either the first or second line */
 194         int c = sscanf(buf, "%d:%2d:%2d%*1[,.]%3d --> %d:%2d:%2d%*1[,.]%3d"
 195                        "%*[ ]X1:%d X2:%d Y1:%d Y2:%d",
 196                        &hs, &ms, &ss, ts_start, &he, &me, &se, ts_end,
 197                        x1, x2, y1, y2);
 198         buf += strcspn(buf, "\n") + 1;
 199         if (c >= 8) {
 200             *ts_start = 100*(ss + 60*(ms + 60*hs)) + *ts_start/10;
 201             *ts_end   = 100*(se + 60*(me + 60*he)) + *ts_end  /10;
 202             return buf;
 203         }
 204     }
 205     return NULL;
 206 }
 207
 208 static int srt_decode_frame(AVCodecContext *avctx,
 209                             void *data, int *got_sub_ptr, AVPacket *avpkt)
 210 {
 211     AVSubtitle *sub = data;
 212     int ts_start, ts_end, x1 = -1, y1 = -1, x2 = -1, y2 = -1;
 213     char buffer[2048];
 214     const char *ptr = avpkt->data;
 215     const char *end = avpkt->data + avpkt->size;
 216
 217     if (avpkt->size <= 0)
 218         return avpkt->size;
 219
 220     ff_ass_init(sub);
 221
 222     while (ptr < end && *ptr) {
 223         ptr = read_ts(ptr, &ts_start, &ts_end, &x1, &y1, &x2, &y2);
 224         if (!ptr)
 225             break;
 226         ptr = srt_to_ass(avctx, buffer, buffer+sizeof(buffer), ptr,
 227                          x1, y1, x2, y2);
 228         ff_ass_add_rect(sub, buffer, ts_start, ts_end, 0);
 229     }
 230
 231     *got_sub_ptr = sub->num_rects > 0;
 232     return avpkt->size;
 233 }
 234
 235 AVCodec ff_srt_decoder = {
 236     .name         = "srt",
 237     .long_name    = NULL_IF_CONFIG_SMALL("SubRip subtitle"),
 238     .type         = AVMEDIA_TYPE_SUBTITLE,
 239     .id           = AV_CODEC_ID_SRT,
 240     .init         = ff_ass_subtitle_header_default,
 241     .decode       = srt_decode_frame,
 242 };