2 * TED Talks captions format decoder
3 * Copyright (c) 2012 Nicolas George
5 * This file is part of FFmpeg.
7 * FFmpeg 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.
12 * FFmpeg 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.
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with FFmpeg; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
22 #include "libavutil/bprint.h"
23 #include "libavutil/log.h"
24 #include "libavutil/opt.h"
27 #include "subtitles.h"
32 FFDemuxSubtitlesQueue subs;
35 static const AVOption tedcaptions_options[] = {
36 { "start_time", "set the start time (offset) of the subtitles, in ms",
37 offsetof(TEDCaptionsDemuxer, start_time), AV_OPT_TYPE_INT64,
38 { .i64 = 15000 }, INT64_MIN, INT64_MAX,
39 AV_OPT_FLAG_SUBTITLE_PARAM | AV_OPT_FLAG_DECODING_PARAM },
43 static const AVClass tedcaptions_demuxer_class = {
44 .class_name = "tedcaptions_demuxer",
45 .item_name = av_default_item_name,
46 .option = tedcaptions_options,
47 .version = LIBAVUTIL_VERSION_INT,
50 #define BETWEEN(a, amin, amax) ((unsigned)((a) - (amin)) <= (amax) - (amin))
52 #define HEX_DIGIT_TEST(c) (BETWEEN(c, '0', '9') || BETWEEN((c) | 32, 'a', 'z'))
53 #define HEX_DIGIT_VAL(c) ((c) <= '9' ? (c) - '0' : ((c) | 32) - 'a' + 10)
54 #define ERR_CODE(c) ((c) < 0 ? (c) : AVERROR_INVALIDDATA)
56 static void av_bprint_utf8(AVBPrint *bp, unsigned c)
61 av_bprint_chars(bp, c, 1);
64 bytes = (av_log2(c) - 2) / 5;
65 av_bprint_chars(bp, (c >> (bytes * 6)) | ((0xFF80 >> bytes) & 0xFF), 1);
66 for (i = bytes - 1; i >= 0; i--)
67 av_bprint_chars(bp, ((c >> (i * 6)) & 0x3F) | 0x80, 1);
70 static void next_byte(AVIOContext *pb, int *cur_byte)
73 int ret = avio_read(pb, &b, 1);
74 *cur_byte = ret > 0 ? b : ret == 0 ? AVERROR_EOF : ret;
77 static void skip_spaces(AVIOContext *pb, int *cur_byte)
79 while (*cur_byte == ' ' || *cur_byte == '\t' ||
80 *cur_byte == '\n' || *cur_byte == '\r')
81 next_byte(pb, cur_byte);
84 static int expect_byte(AVIOContext *pb, int *cur_byte, uint8_t c)
86 skip_spaces(pb, cur_byte);
88 return ERR_CODE(*cur_byte);
89 next_byte(pb, cur_byte);
93 static int parse_string(AVIOContext *pb, int *cur_byte, AVBPrint *bp, int full)
97 ret = expect_byte(pb, cur_byte, '"');
100 while (*cur_byte > 0 && *cur_byte != '"') {
101 if (*cur_byte == '\\') {
102 next_byte(pb, cur_byte);
104 return AVERROR_INVALIDDATA;
105 if ((*cur_byte | 32) == 'u') {
107 for (i = 0; i < 4; i++) {
108 next_byte(pb, cur_byte);
109 if (!HEX_DIGIT_TEST(*cur_byte))
110 return ERR_CODE(*cur_byte);
111 chr = chr * 16 + HEX_DIGIT_VAL(*cur_byte);
113 av_bprint_utf8(bp, chr);
115 av_bprint_chars(bp, *cur_byte, 1);
118 av_bprint_chars(bp, *cur_byte, 1);
120 next_byte(pb, cur_byte);
122 ret = expect_byte(pb, cur_byte, '"');
125 if (full && !av_bprint_is_complete(bp))
126 return AVERROR(ENOMEM);
131 static int parse_label(AVIOContext *pb, int *cur_byte, AVBPrint *bp)
135 av_bprint_init(bp, 0, AV_BPRINT_SIZE_AUTOMATIC);
136 ret = parse_string(pb, cur_byte, bp, 0);
139 ret = expect_byte(pb, cur_byte, ':');
145 static int parse_boolean(AVIOContext *pb, int *cur_byte, int *result)
147 static const char * const text[] = { "false", "true" };
151 skip_spaces(pb, cur_byte);
152 for (i = 0; i < 2; i++) {
156 for (; *p; p++, next_byte(pb, cur_byte))
158 return AVERROR_INVALIDDATA;
159 if (BETWEEN(*cur_byte | 32, 'a', 'z'))
160 return AVERROR_INVALIDDATA;
164 return AVERROR_INVALIDDATA;
167 static int parse_int(AVIOContext *pb, int *cur_byte, int64_t *result)
171 skip_spaces(pb, cur_byte);
172 if ((unsigned)*cur_byte - '0' > 9)
173 return AVERROR_INVALIDDATA;
174 while (BETWEEN(*cur_byte, '0', '9')) {
175 if (val > INT_MAX/10 - (*cur_byte - '0'))
176 return AVERROR_INVALIDDATA;
177 val = val * 10 + (*cur_byte - '0');
178 next_byte(pb, cur_byte);
184 static int parse_file(AVIOContext *pb, FFDemuxSubtitlesQueue *subs)
186 int ret, cur_byte, start_of_par;
187 AVBPrint label, content;
188 int64_t pos, start, duration;
191 av_bprint_init(&content, 0, AV_BPRINT_SIZE_UNLIMITED);
193 next_byte(pb, &cur_byte);
194 ret = expect_byte(pb, &cur_byte, '{');
196 return AVERROR_INVALIDDATA;
197 ret = parse_label(pb, &cur_byte, &label);
198 if (ret < 0 || strcmp(label.str, "captions"))
199 return AVERROR_INVALIDDATA;
200 ret = expect_byte(pb, &cur_byte, '[');
202 return AVERROR_INVALIDDATA;
204 start = duration = AV_NOPTS_VALUE;
205 ret = expect_byte(pb, &cur_byte, '{');
208 pos = avio_tell(pb) - 1;
210 ret = parse_label(pb, &cur_byte, &label);
213 if (!strcmp(label.str, "startOfParagraph")) {
214 ret = parse_boolean(pb, &cur_byte, &start_of_par);
217 } else if (!strcmp(label.str, "content")) {
218 ret = parse_string(pb, &cur_byte, &content, 1);
221 } else if (!strcmp(label.str, "startTime")) {
222 ret = parse_int(pb, &cur_byte, &start);
225 } else if (!strcmp(label.str, "duration")) {
226 ret = parse_int(pb, &cur_byte, &duration);
230 ret = AVERROR_INVALIDDATA;
233 skip_spaces(pb, &cur_byte);
236 next_byte(pb, &cur_byte);
238 ret = expect_byte(pb, &cur_byte, '}');
242 if (!content.size || start == AV_NOPTS_VALUE ||
243 duration == AV_NOPTS_VALUE) {
244 ret = AVERROR_INVALIDDATA;
247 pkt = ff_subtitles_queue_insert(subs, content.str, content.len, 0);
249 ret = AVERROR(ENOMEM);
254 pkt->duration = duration;
255 av_bprint_clear(&content);
257 skip_spaces(pb, &cur_byte);
260 next_byte(pb, &cur_byte);
262 ret = expect_byte(pb, &cur_byte, ']');
265 ret = expect_byte(pb, &cur_byte, '}');
268 skip_spaces(pb, &cur_byte);
269 if (cur_byte != AVERROR_EOF)
270 ret = ERR_CODE(cur_byte);
272 av_bprint_finalize(&content, NULL);
276 static av_cold int tedcaptions_read_header(AVFormatContext *avf)
278 TEDCaptionsDemuxer *tc = avf->priv_data;
279 AVStream *st = avformat_new_stream(avf, NULL);
284 return AVERROR(ENOMEM);
286 ret = parse_file(avf->pb, &tc->subs);
288 if (ret == AVERROR_INVALIDDATA)
289 av_log(avf, AV_LOG_ERROR, "Syntax error near offset %"PRId64".\n",
291 ff_subtitles_queue_clean(&tc->subs);
294 ff_subtitles_queue_finalize(avf, &tc->subs);
295 for (i = 0; i < tc->subs.nb_subs; i++)
296 tc->subs.subs[i]->pts += tc->start_time;
298 last = tc->subs.subs[tc->subs.nb_subs - 1];
299 st->codecpar->codec_type = AVMEDIA_TYPE_SUBTITLE;
300 st->codecpar->codec_id = AV_CODEC_ID_TEXT;
301 avpriv_set_pts_info(st, 64, 1, 1000);
302 st->probe_packets = 0;
304 st->duration = last->pts + last->duration;
310 static int tedcaptions_read_packet(AVFormatContext *avf, AVPacket *packet)
312 TEDCaptionsDemuxer *tc = avf->priv_data;
314 return ff_subtitles_queue_read_packet(&tc->subs, packet);
317 static int tedcaptions_read_close(AVFormatContext *avf)
319 TEDCaptionsDemuxer *tc = avf->priv_data;
321 ff_subtitles_queue_clean(&tc->subs);
325 static av_cold int tedcaptions_read_probe(const AVProbeData *p)
327 static const char *const tags[] = {
328 "\"captions\"", "\"duration\"", "\"content\"",
329 "\"startOfParagraph\"", "\"startTime\"",
331 unsigned i, count = 0;
334 if (p->buf[strspn(p->buf, " \t\r\n")] != '{')
336 for (i = 0; i < FF_ARRAY_ELEMS(tags); i++) {
337 if (!(t = strstr(p->buf, tags[i])))
339 t += strlen(tags[i]);
340 t += strspn(t, " \t\r\n");
344 return count == FF_ARRAY_ELEMS(tags) ? AVPROBE_SCORE_MAX :
345 count ? AVPROBE_SCORE_EXTENSION : 0;
348 static int tedcaptions_read_seek(AVFormatContext *avf, int stream_index,
349 int64_t min_ts, int64_t ts, int64_t max_ts,
352 TEDCaptionsDemuxer *tc = avf->priv_data;
353 return ff_subtitles_queue_seek(&tc->subs, avf, stream_index,
354 min_ts, ts, max_ts, flags);
357 const AVInputFormat ff_tedcaptions_demuxer = {
358 .name = "tedcaptions",
359 .long_name = NULL_IF_CONFIG_SMALL("TED Talks captions"),
360 .priv_data_size = sizeof(TEDCaptionsDemuxer),
361 .priv_class = &tedcaptions_demuxer_class,
362 .read_header = tedcaptions_read_header,
363 .read_packet = tedcaptions_read_packet,
364 .read_close = tedcaptions_read_close,
365 .read_probe = tedcaptions_read_probe,
366 .read_seek2 = tedcaptions_read_seek,