FFmpeg
tedcaptionsdec.c
Go to the documentation of this file.
1 /*
2  * TED Talks captions format decoder
3  * Copyright (c) 2012 Nicolas George
4  *
5  * This file is part of FFmpeg.
6  *
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.
11  *
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.
16  *
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
20  */
21 
22 #include "libavutil/bprint.h"
23 #include "libavutil/log.h"
24 #include "libavutil/opt.h"
25 #include "avformat.h"
26 #include "internal.h"
27 #include "subtitles.h"
28 
29 typedef struct {
30  AVClass *class;
31  int64_t start_time;
34 
35 static const AVOption tedcaptions_options[] = {
36  { "start_time", "set the start time (offset) of the subtitles, in ms",
38  { .i64 = 15000 }, INT64_MIN, INT64_MAX,
40  { NULL },
41 };
42 
44  .class_name = "tedcaptions_demuxer",
45  .item_name = av_default_item_name,
46  .option = tedcaptions_options,
47  .version = LIBAVUTIL_VERSION_INT,
48 };
49 
50 #define BETWEEN(a, amin, amax) ((unsigned)((a) - (amin)) <= (amax) - (amin))
51 
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)
55 
56 static void av_bprint_utf8(AVBPrint *bp, unsigned c)
57 {
58  int bytes, i;
59 
60  if (c <= 0x7F) {
61  av_bprint_chars(bp, c, 1);
62  return;
63  }
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);
68 }
69 
70 static void next_byte(AVIOContext *pb, int *cur_byte)
71 {
72  uint8_t b;
73  int ret = avio_read(pb, &b, 1);
74  *cur_byte = ret > 0 ? b : ret == 0 ? AVERROR_EOF : ret;
75 }
76 
77 static void skip_spaces(AVIOContext *pb, int *cur_byte)
78 {
79  while (*cur_byte == ' ' || *cur_byte == '\t' ||
80  *cur_byte == '\n' || *cur_byte == '\r')
81  next_byte(pb, cur_byte);
82 }
83 
84 static int expect_byte(AVIOContext *pb, int *cur_byte, uint8_t c)
85 {
86  skip_spaces(pb, cur_byte);
87  if (*cur_byte != c)
88  return ERR_CODE(*cur_byte);
89  next_byte(pb, cur_byte);
90  return 0;
91 }
92 
93 static int parse_string(AVIOContext *pb, int *cur_byte, AVBPrint *bp, int full)
94 {
95  int ret;
96 
97  ret = expect_byte(pb, cur_byte, '"');
98  if (ret < 0)
99  return ret;
100  while (*cur_byte > 0 && *cur_byte != '"') {
101  if (*cur_byte == '\\') {
102  next_byte(pb, cur_byte);
103  if (*cur_byte < 0)
104  return AVERROR_INVALIDDATA;
105  if ((*cur_byte | 32) == 'u') {
106  unsigned chr = 0, i;
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);
112  }
113  av_bprint_utf8(bp, chr);
114  } else {
115  av_bprint_chars(bp, *cur_byte, 1);
116  }
117  } else {
118  av_bprint_chars(bp, *cur_byte, 1);
119  }
120  next_byte(pb, cur_byte);
121  }
122  ret = expect_byte(pb, cur_byte, '"');
123  if (ret < 0)
124  return ret;
125  if (full && !av_bprint_is_complete(bp))
126  return AVERROR(ENOMEM);
127 
128  return 0;
129 }
130 
131 static int parse_label(AVIOContext *pb, int *cur_byte, AVBPrint *bp)
132 {
133  int ret;
134 
136  ret = parse_string(pb, cur_byte, bp, 0);
137  if (ret < 0)
138  return ret;
139  ret = expect_byte(pb, cur_byte, ':');
140  if (ret < 0)
141  return ret;
142  return 0;
143 }
144 
145 static int parse_boolean(AVIOContext *pb, int *cur_byte, int *result)
146 {
147  static const char * const text[] = { "false", "true" };
148  const char *p;
149  int i;
150 
151  skip_spaces(pb, cur_byte);
152  for (i = 0; i < 2; i++) {
153  p = text[i];
154  if (*cur_byte != *p)
155  continue;
156  for (; *p; p++, next_byte(pb, cur_byte))
157  if (*cur_byte != *p)
158  return AVERROR_INVALIDDATA;
159  if (BETWEEN(*cur_byte | 32, 'a', 'z'))
160  return AVERROR_INVALIDDATA;
161  *result = i;
162  return 0;
163  }
164  return AVERROR_INVALIDDATA;
165 }
166 
167 static int parse_int(AVIOContext *pb, int *cur_byte, int64_t *result)
168 {
169  int64_t val = 0;
170 
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  val = val * 10 + (*cur_byte - '0');
176  next_byte(pb, cur_byte);
177  }
178  *result = val;
179  return 0;
180 }
181 
183 {
184  int ret, cur_byte, start_of_par;
185  AVBPrint label, content;
186  int64_t pos, start, duration;
187  AVPacket *pkt;
188 
190 
191  next_byte(pb, &cur_byte);
192  ret = expect_byte(pb, &cur_byte, '{');
193  if (ret < 0)
194  return AVERROR_INVALIDDATA;
195  ret = parse_label(pb, &cur_byte, &label);
196  if (ret < 0 || strcmp(label.str, "captions"))
197  return AVERROR_INVALIDDATA;
198  ret = expect_byte(pb, &cur_byte, '[');
199  if (ret < 0)
200  return AVERROR_INVALIDDATA;
201  while (1) {
202  start = duration = AV_NOPTS_VALUE;
203  ret = expect_byte(pb, &cur_byte, '{');
204  if (ret < 0)
205  goto fail;
206  pos = avio_tell(pb) - 1;
207  while (1) {
208  ret = parse_label(pb, &cur_byte, &label);
209  if (ret < 0)
210  goto fail;
211  if (!strcmp(label.str, "startOfParagraph")) {
212  ret = parse_boolean(pb, &cur_byte, &start_of_par);
213  if (ret < 0)
214  goto fail;
215  } else if (!strcmp(label.str, "content")) {
216  ret = parse_string(pb, &cur_byte, &content, 1);
217  if (ret < 0)
218  goto fail;
219  } else if (!strcmp(label.str, "startTime")) {
220  ret = parse_int(pb, &cur_byte, &start);
221  if (ret < 0)
222  goto fail;
223  } else if (!strcmp(label.str, "duration")) {
224  ret = parse_int(pb, &cur_byte, &duration);
225  if (ret < 0)
226  goto fail;
227  } else {
228  ret = AVERROR_INVALIDDATA;
229  goto fail;
230  }
231  skip_spaces(pb, &cur_byte);
232  if (cur_byte != ',')
233  break;
234  next_byte(pb, &cur_byte);
235  }
236  ret = expect_byte(pb, &cur_byte, '}');
237  if (ret < 0)
238  goto fail;
239 
240  if (!content.size || start == AV_NOPTS_VALUE ||
241  duration == AV_NOPTS_VALUE) {
242  ret = AVERROR_INVALIDDATA;
243  goto fail;
244  }
245  pkt = ff_subtitles_queue_insert(subs, content.str, content.len, 0);
246  if (!pkt) {
247  ret = AVERROR(ENOMEM);
248  goto fail;
249  }
250  pkt->pos = pos;
251  pkt->pts = start;
252  pkt->duration = duration;
253  av_bprint_clear(&content);
254 
255  skip_spaces(pb, &cur_byte);
256  if (cur_byte != ',')
257  break;
258  next_byte(pb, &cur_byte);
259  }
260  ret = expect_byte(pb, &cur_byte, ']');
261  if (ret < 0)
262  goto fail;
263  ret = expect_byte(pb, &cur_byte, '}');
264  if (ret < 0)
265  goto fail;
266  skip_spaces(pb, &cur_byte);
267  if (cur_byte != AVERROR_EOF)
268  ret = ERR_CODE(cur_byte);
269 fail:
270  av_bprint_finalize(&content, NULL);
271  return ret;
272 }
273 
275 {
277  AVStream *st = avformat_new_stream(avf, NULL);
278  int ret, i;
279  AVPacket *last;
280 
281  if (!st)
282  return AVERROR(ENOMEM);
283 
284  ret = parse_file(avf->pb, &tc->subs);
285  if (ret < 0) {
286  if (ret == AVERROR_INVALIDDATA)
287  av_log(avf, AV_LOG_ERROR, "Syntax error near offset %"PRId64".\n",
288  avio_tell(avf->pb));
290  return ret;
291  }
293  for (i = 0; i < tc->subs.nb_subs; i++)
294  tc->subs.subs[i].pts += tc->start_time;
295 
296  last = &tc->subs.subs[tc->subs.nb_subs - 1];
299  avpriv_set_pts_info(st, 64, 1, 1000);
300  st->probe_packets = 0;
301  st->start_time = 0;
302  st->duration = last->pts + last->duration;
303  st->cur_dts = 0;
304 
305  return 0;
306 }
307 
309 {
311 
312  return ff_subtitles_queue_read_packet(&tc->subs, packet);
313 }
314 
316 {
318 
320  return 0;
321 }
322 
324 {
325  static const char *const tags[] = {
326  "\"captions\"", "\"duration\"", "\"content\"",
327  "\"startOfParagraph\"", "\"startTime\"",
328  };
329  unsigned i, count = 0;
330  const char *t;
331 
332  if (p->buf[strspn(p->buf, " \t\r\n")] != '{')
333  return 0;
334  for (i = 0; i < FF_ARRAY_ELEMS(tags); i++) {
335  if (!(t = strstr(p->buf, tags[i])))
336  continue;
337  t += strlen(tags[i]);
338  t += strspn(t, " \t\r\n");
339  if (*t == ':')
340  count++;
341  }
342  return count == FF_ARRAY_ELEMS(tags) ? AVPROBE_SCORE_MAX :
343  count ? AVPROBE_SCORE_EXTENSION : 0;
344 }
345 
346 static int tedcaptions_read_seek(AVFormatContext *avf, int stream_index,
347  int64_t min_ts, int64_t ts, int64_t max_ts,
348  int flags)
349 {
351  return ff_subtitles_queue_seek(&tc->subs, avf, stream_index,
352  min_ts, ts, max_ts, flags);
353 }
354 
356  .name = "tedcaptions",
357  .long_name = NULL_IF_CONFIG_SMALL("TED Talks captions"),
358  .priv_data_size = sizeof(TEDCaptionsDemuxer),
359  .priv_class = &tedcaptions_demuxer_class,
364  .read_seek2 = tedcaptions_read_seek,
365 };
#define NULL
Definition: coverity.c:32
Bytestream IO Context.
Definition: avio.h:161
#define AVERROR_INVALIDDATA
Invalid data found when processing input.
Definition: error.h:59
static int tedcaptions_read_seek(AVFormatContext *avf, int stream_index, int64_t min_ts, int64_t ts, int64_t max_ts, int flags)
AVOption.
Definition: opt.h:248
#define AV_OPT_FLAG_SUBTITLE_PARAM
Definition: opt.h:282
#define LIBAVUTIL_VERSION_INT
Definition: version.h:85
int64_t pos
byte position in stream, -1 if unknown
Definition: packet.h:383
void avpriv_set_pts_info(AVStream *s, int pts_wrap_bits, unsigned int pts_num, unsigned int pts_den)
Set the time base and wrapping info for a given stream.
Definition: utils.c:4869
int probe_packets
Number of packets to buffer for codec probing.
Definition: avformat.h:1086
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
const char * av_default_item_name(void *ptr)
Return the context name.
Definition: log.c:235
int av_log2(unsigned v)
Definition: intmath.c:26
void ff_subtitles_queue_clean(FFDemuxSubtitlesQueue *q)
Remove and destroy all the subtitles packets.
Definition: subtitles.c:302
#define tc
Definition: regdef.h:69
static AVPacket pkt
static void av_bprint_utf8(AVBPrint *bp, unsigned c)
AVInputFormat ff_tedcaptions_demuxer
int av_bprint_finalize(AVBPrint *buf, char **ret_str)
Finalize a print buffer.
Definition: bprint.c:235
#define BETWEEN(a, amin, amax)
Format I/O context.
Definition: avformat.h:1351
int64_t cur_dts
Definition: avformat.h:1079
const char * class_name
The name of the class; usually it is the same name as the context structure type to which the AVClass...
Definition: log.h:72
static const AVClass tedcaptions_demuxer_class
static int64_t start_time
Definition: ffplay.c:332
uint8_t
#define av_cold
Definition: attributes.h:88
FFDemuxSubtitlesQueue subs
AVOptions.
int ff_subtitles_queue_read_packet(FFDemuxSubtitlesQueue *q, AVPacket *pkt)
Generic read_packet() callback for subtitles demuxers using this queue system.
Definition: subtitles.c:212
int64_t duration
Duration of this packet in AVStream->time_base units, 0 if unknown.
Definition: packet.h:381
Undefined Behavior In the C some operations are like signed integer dereferencing freed accessing outside allocated Undefined Behavior must not occur in a C it is not safe even if the output of undefined operations is unused The unsafety may seem nit picking but Optimizing compilers have in fact optimized code on the assumption that no undefined Behavior occurs Optimizing code based on wrong assumptions can and has in some cases lead to effects beyond the output of computations The signed integer overflow problem in speed critical code Code which is highly optimized and works with signed integers sometimes has the problem that often the output of the computation does not c
Definition: undefined.txt:32
static int tedcaptions_read_packet(AVFormatContext *avf, AVPacket *packet)
AVStream * avformat_new_stream(AVFormatContext *s, const AVCodec *c)
Add a new stream to a media file.
Definition: utils.c:4450
static void skip_spaces(AVIOContext *pb, int *cur_byte)
int64_t duration
Definition: movenc.c:63
#define AVERROR_EOF
End of file.
Definition: error.h:55
static av_cold int read_close(AVFormatContext *ctx)
Definition: libcdio.c:145
static av_always_inline int64_t avio_tell(AVIOContext *s)
ftell() equivalent for AVIOContext.
Definition: avio.h:557
#define av_log(a,...)
int avio_read(AVIOContext *s, unsigned char *buf, int size)
Read size bytes from AVIOContext into buf.
Definition: aviobuf.c:637
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
Definition: log.h:194
#define AV_BPRINT_SIZE_UNLIMITED
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:153
void av_bprint_init(AVBPrint *buf, unsigned size_init, unsigned size_max)
Definition: bprint.c:69
unsigned int pos
Definition: spdifenc.c:410
#define ERR_CODE(c)
enum AVMediaType codec_type
General type of the encoded data.
Definition: codec_par.h:56
static int parse_label(AVIOContext *pb, int *cur_byte, AVBPrint *bp)
GLsizei count
Definition: opengl_enc.c:108
#define fail()
Definition: checkasm.h:123
static int tedcaptions_read_close(AVFormatContext *avf)
unsigned char * buf
Buffer must have AVPROBE_PADDING_SIZE of extra allocated bytes filled with zero.
Definition: avformat.h:443
#define b
Definition: input.c:41
static int parse_file(AVIOContext *pb, FFDemuxSubtitlesQueue *subs)
static int av_bprint_is_complete(const AVBPrint *buf)
Test if the print buffer is complete (not truncated).
Definition: bprint.h:185
static void next_byte(AVIOContext *pb, int *cur_byte)
static av_cold int tedcaptions_read_header(AVFormatContext *avf)
#define FF_ARRAY_ELEMS(a)
static int read_header(FFV1Context *f)
Definition: ffv1dec.c:527
Stream structure.
Definition: avformat.h:876
static int read_packet(void *opaque, uint8_t *buf, int buf_size)
Definition: avio_reading.c:42
#define HEX_DIGIT_TEST(c)
int ff_subtitles_queue_seek(FFDemuxSubtitlesQueue *q, AVFormatContext *s, int stream_index, int64_t min_ts, int64_t ts, int64_t max_ts, int flags)
Update current_sub_idx to emulate a seek.
Definition: subtitles.c:250
#define AV_BPRINT_SIZE_AUTOMATIC
AVIOContext * pb
I/O context.
Definition: avformat.h:1393
Describe the class of an AVClass context structure.
Definition: log.h:67
#define AV_OPT_FLAG_DECODING_PARAM
a generic parameter which can be set by the user for demuxing or decoding
Definition: opt.h:279
#define AVPROBE_SCORE_EXTENSION
score for file extension
Definition: avformat.h:451
This structure contains the data a format has to probe a file.
Definition: avformat.h:441
#define flags(name, subs,...)
Definition: cbs_av1.c:560
static int read_probe(const AVProbeData *pd)
Definition: jvdec.c:55
void av_bprint_clear(AVBPrint *buf)
Reset the string to "" but keep internal allocated data.
Definition: bprint.c:227
int64_t duration
Decoding: duration of the stream, in stream time base.
Definition: avformat.h:925
#define AVPROBE_SCORE_MAX
maximum score
Definition: avformat.h:453
#define HEX_DIGIT_VAL(c)
Main libavformat public API header.
static int expect_byte(AVIOContext *pb, int *cur_byte, uint8_t c)
raw UTF-8 text
Definition: codec_id.h:517
static av_cold int tedcaptions_read_probe(const AVProbeData *p)
int64_t start_time
Decoding: pts of the first frame of the stream in presentation order, in stream time base...
Definition: avformat.h:915
static const AVOption tedcaptions_options[]
AVPacket * subs
array of subtitles packets
Definition: subtitles.h:103
void * priv_data
Format private data.
Definition: avformat.h:1379
static int parse_string(AVIOContext *pb, int *cur_byte, AVBPrint *bp, int full)
and forward the result(frame or status change) to the corresponding input.If nothing is possible
const char * name
A comma separated list of short names for the format.
Definition: avformat.h:650
static int parse_boolean(AVIOContext *pb, int *cur_byte, int *result)
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1023
AVPacket * ff_subtitles_queue_insert(FFDemuxSubtitlesQueue *q, const uint8_t *event, size_t len, int merge)
Insert a new subtitle event.
Definition: subtitles.c:111
Filter the word “frame” indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later.That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another.Frame references ownership and permissions
static double val(void *priv, double ch)
Definition: aeval.c:76
This structure stores compressed data.
Definition: packet.h:340
int nb_subs
number of subtitles packets
Definition: subtitles.h:104
void ff_subtitles_queue_finalize(void *log_ctx, FFDemuxSubtitlesQueue *q)
Set missing durations, sort subtitles by PTS (and then byte position), and drop duplicated events...
Definition: subtitles.c:194
int64_t pts
Presentation timestamp in AVStream->time_base units; the time at which the decompressed packet will b...
Definition: packet.h:356
int i
Definition: input.c:407
#define AV_NOPTS_VALUE
Undefined timestamp value.
Definition: avutil.h:248
void av_bprint_chars(AVBPrint *buf, char c, unsigned n)
Append char c n times to a print buffer.
Definition: bprint.c:140
static int parse_int(AVIOContext *pb, int *cur_byte, int64_t *result)