libavformat/srtdec.c

   1 /*
   2  * SubRip subtitle demuxer
   3  * Copyright (c) 2010  Aurelien Jacobs <aurel@gnuage.org>
   4  * Copyright (c) 2015  Clément Bœsch <u pkh me>
   5  *
   6  * This file is part of FFmpeg.
   7  *
   8  * FFmpeg is free software; you can redistribute it and/or
   9  * modify it under the terms of the GNU Lesser General Public
  10  * License as published by the Free Software Foundation; either
  11  * version 2.1 of the License, or (at your option) any later version.
  12  *
  13  * FFmpeg is distributed in the hope that it will be useful,
  14  * but WITHOUT ANY WARRANTY; without even the implied warranty of
  15  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
  16  * Lesser General Public License for more details.
  17  *
  18  * You should have received a copy of the GNU Lesser General Public
  19  * License along with FFmpeg; if not, write to the Free Software
  20  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
  21  */
  22
  23 #include "avformat.h"
  24 #include "demux.h"
  25 #include "internal.h"
  26 #include "subtitles.h"
  27 #include "libavutil/bprint.h"
  28 #include "libavutil/intreadwrite.h"
  29
  30 typedef struct {
  31     FFDemuxSubtitlesQueue q;
  32 } SRTContext;
  33
  34 static int srt_probe(const AVProbeData *p)
  35 {
  36     int v;
  37     char buf[64], *pbuf;
  38     FFTextReader tr;
  39
  40     ff_text_init_buf(&tr, p->buf, p->buf_size);
  41
  42     while (ff_text_peek_r8(&tr) == '\r' || ff_text_peek_r8(&tr) == '\n')
  43         ff_text_r8(&tr);
  44
  45     /* Check if the first non-empty line is a number. We do not check what the
  46      * number is because in practice it can be anything.
  47      * Also, that number can be followed by random garbage, so we can not
  48      * unfortunately check that we only have a number. */
  49     if (ff_subtitles_read_line(&tr, buf, sizeof(buf)) < 0 ||
  50         strtol(buf, &pbuf, 10) < 0 || pbuf == buf)
  51         return 0;
  52
  53     /* Check if the next line matches a SRT timestamp */
  54     if (ff_subtitles_read_line(&tr, buf, sizeof(buf)) < 0)
  55         return 0;
  56     pbuf = buf;
  57     if (buf[0] == '-')
  58         pbuf++;
  59     if (pbuf[0] >= '0' && pbuf[0] <= '9' && strstr(buf, " --> ")
  60         && sscanf(buf, "%*d:%*d:%*d%*1[,.]%*d --> %*d:%*d:%*d%*1[,.]%d", &v) == 1)
  61         return AVPROBE_SCORE_MAX;
  62
  63     return 0;
  64 }
  65
  66 struct event_info {
  67     int32_t x1, x2, y1, y2;
  68     int duration;
  69     int64_t pts;
  70     int64_t pos;
  71 };
  72
  73 static int get_event_info(const char *line, struct event_info *ei)
  74 {
  75     int hh1, mm1, ss1, ms1;
  76     int hh2, mm2, ss2, ms2;
  77
  78     ei->x1 = ei->x2 = ei->y1 = ei->y2 = ei->duration = -1;
  79     ei->pts = AV_NOPTS_VALUE;
  80     ei->pos = -1;
  81     if (sscanf(line, "%d:%d:%d%*1[,.]%d --> %d:%d:%d%*1[,.]%d"
  82                "%*[ ]X1:%"PRId32" X2:%"PRId32" Y1:%"PRId32" Y2:%"PRId32,
  83                &hh1, &mm1, &ss1, &ms1,
  84                &hh2, &mm2, &ss2, &ms2,
  85                &ei->x1, &ei->x2, &ei->y1, &ei->y2) >= 8) {
  86         const int64_t start = (hh1*3600LL + mm1*60LL + ss1) * 1000LL + ms1;
  87         const int64_t end   = (hh2*3600LL + mm2*60LL + ss2) * 1000LL + ms2;
  88         ei->duration = end - start;
  89         ei->pts = start;
  90         return 0;
  91     }
  92     return -1;
  93 }
  94
  95 static int add_event(FFDemuxSubtitlesQueue *q, AVBPrint *buf, char *line_cache,
  96                      const struct event_info *ei, int append_cache)
  97 {
  98     if (append_cache && line_cache[0])
  99         av_bprintf(buf, "%s\n", line_cache);
 100     line_cache[0] = 0;
 101     if (!av_bprint_is_complete(buf))
 102         return AVERROR(ENOMEM);
 103
 104     while (buf->len > 0 && buf->str[buf->len - 1] == '\n')
 105         buf->str[--buf->len] = 0;
 106
 107     if (buf->len) {
 108         AVPacket *sub = ff_subtitles_queue_insert_bprint(q, buf, 0);
 109         if (!sub)
 110             return AVERROR(ENOMEM);
 111         av_bprint_clear(buf);
 112         sub->pos = ei->pos;
 113         sub->pts = ei->pts;
 114         sub->duration = ei->duration;
 115         if (ei->x1 != -1) {
 116             uint8_t *p = av_packet_new_side_data(sub, AV_PKT_DATA_SUBTITLE_POSITION, 16);
 117             if (p) {
 118                 AV_WL32(p,      ei->x1);
 119                 AV_WL32(p +  4, ei->y1);
 120                 AV_WL32(p +  8, ei->x2);
 121                 AV_WL32(p + 12, ei->y2);
 122             }
 123         }
 124     }
 125
 126     return 0;
 127 }
 128
 129 static int srt_read_header(AVFormatContext *s)
 130 {
 131     SRTContext *srt = s->priv_data;
 132     AVBPrint buf;
 133     AVStream *st = avformat_new_stream(s, NULL);
 134     int res = 0;
 135     char line[4096], line_cache[4096];
 136     int has_event_info = 0;
 137     struct event_info ei;
 138     FFTextReader tr;
 139     ff_text_init_avio(s, &tr, s->pb);
 140
 141     if (!st)
 142         return AVERROR(ENOMEM);
 143     avpriv_set_pts_info(st, 64, 1, 1000);
 144     st->codecpar->codec_type = AVMEDIA_TYPE_SUBTITLE;
 145     st->codecpar->codec_id   = AV_CODEC_ID_SUBRIP;
 146
 147     av_bprint_init(&buf, 0, AV_BPRINT_SIZE_UNLIMITED);
 148
 149     line_cache[0] = 0;
 150
 151     while (!ff_text_eof(&tr)) {
 152         struct event_info tmp_ei;
 153         const int64_t pos = ff_text_pos(&tr);
 154         ptrdiff_t len = ff_subtitles_read_line(&tr, line, sizeof(line));
 155
 156         if (len < 0)
 157             break;
 158
 159         if (!len || !line[0])
 160             continue;
 161
 162         if (get_event_info(line, &tmp_ei) < 0) {
 163             char *pline;
 164
 165             if (!has_event_info)
 166                 continue;
 167
 168             if (line_cache[0]) {
 169                 /* We got some cache and a new line so we assume the cached
 170                  * line was actually part of the payload */
 171                 av_bprintf(&buf, "%s\n", line_cache);
 172                 line_cache[0] = 0;
 173             }
 174
 175             /* If the line doesn't start with a number, we assume it's part of
 176              * the payload, otherwise is likely an event number preceding the
 177              * timing information... but we can't be sure of this yet, so we
 178              * cache it */
 179             if (strtol(line, &pline, 10) < 0 || line == pline)
 180                 av_bprintf(&buf, "%s\n", line);
 181             else
 182                 strcpy(line_cache, line);
 183         } else {
 184             if (has_event_info) {
 185                 /* We have the information of previous event, append it to the
 186                  * queue. We insert the cached line if and only if the payload
 187                  * is empty and the cached line is not a standalone number. */
 188                 char *pline = NULL;
 189                 const int standalone_number = strtol(line_cache, &pline, 10) >= 0 && pline && !*pline;
 190                 res = add_event(&srt->q, &buf, line_cache, &ei, !buf.len && !standalone_number);
 191                 if (res < 0)
 192                     goto end;
 193             } else {
 194                 has_event_info = 1;
 195             }
 196             tmp_ei.pos = pos;
 197             ei = tmp_ei;
 198         }
 199     }
 200
 201     /* Append the last event. Here we force the cache to be flushed, because a
 202      * trailing number is more likely to be genuine (for example a copyright
 203      * date) and not the event index of an inexistent event */
 204     if (has_event_info) {
 205         res = add_event(&srt->q, &buf, line_cache, &ei, 1);
 206         if (res < 0)
 207             goto end;
 208     }
 209
 210     ff_subtitles_queue_finalize(s, &srt->q);
 211
 212 end:
 213     av_bprint_finalize(&buf, NULL);
 214     return res;
 215 }
 216
 217 const FFInputFormat ff_srt_demuxer = {
 218     .p.name      = "srt",
 219     .p.long_name = NULL_IF_CONFIG_SMALL("SubRip subtitle"),
 220     .priv_data_size = sizeof(SRTContext),
 221     .flags_internal = FF_INFMT_FLAG_INIT_CLEANUP,
 222     .read_probe  = srt_probe,
 223     .read_header = srt_read_header,
 224     .read_packet = ff_subtitles_read_packet,
 225     .read_seek2  = ff_subtitles_read_seek,
 226     .read_close  = ff_subtitles_read_close,
 227 };