FFmpeg
webvttdec.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2012 Clément Bœsch
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU Lesser General Public
8  * License as published by the Free Software Foundation; either
9  * version 2.1 of the License, or (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14  * Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with FFmpeg; if not, write to the Free Software
18  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
21 /**
22  * @file
23  * WebVTT subtitle demuxer
24  * @see http://dev.w3.org/html5/webvtt/
25  */
26 
27 #include "avformat.h"
28 #include "internal.h"
29 #include "subtitles.h"
30 #include "libavutil/bprint.h"
31 #include "libavutil/intreadwrite.h"
32 #include "libavutil/opt.h"
33 
34 typedef struct {
35  const AVClass *class;
37  int kind;
39 
40 static int webvtt_probe(const AVProbeData *p)
41 {
42  const uint8_t *ptr = p->buf;
43 
44  if (AV_RB24(ptr) == 0xEFBBBF)
45  ptr += 3; /* skip UTF-8 BOM */
46  if (!strncmp(ptr, "WEBVTT", 6) &&
47  (!ptr[6] || strchr("\n\r\t ", ptr[6])))
48  return AVPROBE_SCORE_MAX;
49  return 0;
50 }
51 
52 static int64_t read_ts(const char *s)
53 {
54  int hh, mm, ss, ms;
55  if (sscanf(s, "%u:%u:%u.%u", &hh, &mm, &ss, &ms) == 4) return (hh*3600LL + mm*60LL + ss) * 1000LL + ms;
56  if (sscanf(s, "%u:%u.%u", &mm, &ss, &ms) == 3) return ( mm*60LL + ss) * 1000LL + ms;
57  return AV_NOPTS_VALUE;
58 }
59 
61 {
62  WebVTTContext *webvtt = s->priv_data;
63  AVBPrint cue;
64  int res = 0;
66 
67  if (!st)
68  return AVERROR(ENOMEM);
69  avpriv_set_pts_info(st, 64, 1, 1000);
72  st->disposition |= webvtt->kind;
73 
75 
76  for (;;) {
77  int i;
78  int64_t pos;
79  AVPacket *sub;
80  const char *p, *identifier, *settings;
81  int identifier_len, settings_len;
82  int64_t ts_start, ts_end;
83 
84  ff_subtitles_read_chunk(s->pb, &cue);
85 
86  if (!cue.len)
87  break;
88 
89  p = identifier = cue.str;
90  pos = avio_tell(s->pb);
91 
92  /* ignore header chunk */
93  if (!strncmp(p, "\xEF\xBB\xBFWEBVTT", 9) ||
94  !strncmp(p, "WEBVTT", 6) ||
95  !strncmp(p, "NOTE", 4))
96  continue;
97 
98  /* optional cue identifier (can be a number like in SRT or some kind of
99  * chaptering id) */
100  for (i = 0; p[i] && p[i] != '\n' && p[i] != '\r'; i++) {
101  if (!strncmp(p + i, "-->", 3)) {
102  identifier = NULL;
103  break;
104  }
105  }
106  if (!identifier)
107  identifier_len = 0;
108  else {
109  identifier_len = strcspn(p, "\r\n");
110  p += identifier_len;
111  if (*p == '\r')
112  p++;
113  if (*p == '\n')
114  p++;
115  }
116 
117  /* cue timestamps */
118  if ((ts_start = read_ts(p)) == AV_NOPTS_VALUE)
119  break;
120  if (!(p = strstr(p, "-->")))
121  break;
122  p += 2;
123  do p++; while (*p == ' ' || *p == '\t');
124  if ((ts_end = read_ts(p)) == AV_NOPTS_VALUE)
125  break;
126 
127  /* optional cue settings */
128  p += strcspn(p, "\n\r\t ");
129  while (*p == '\t' || *p == ' ')
130  p++;
131  settings = p;
132  settings_len = strcspn(p, "\r\n");
133  p += settings_len;
134  if (*p == '\r')
135  p++;
136  if (*p == '\n')
137  p++;
138 
139  /* create packet */
140  sub = ff_subtitles_queue_insert(&webvtt->q, p, strlen(p), 0);
141  if (!sub) {
142  res = AVERROR(ENOMEM);
143  goto end;
144  }
145  sub->pos = pos;
146  sub->pts = ts_start;
147  sub->duration = ts_end - ts_start;
148 
149 #define SET_SIDE_DATA(name, type) do { \
150  if (name##_len) { \
151  uint8_t *buf = av_packet_new_side_data(sub, type, name##_len); \
152  if (!buf) { \
153  res = AVERROR(ENOMEM); \
154  goto end; \
155  } \
156  memcpy(buf, name, name##_len); \
157  } \
158 } while (0)
159 
162  }
163 
164  ff_subtitles_queue_finalize(s, &webvtt->q);
165 
166 end:
167  if (res < 0)
168  ff_subtitles_queue_clean(&webvtt->q);
169  av_bprint_finalize(&cue, NULL);
170  return res;
171 }
172 
174 {
175  WebVTTContext *webvtt = s->priv_data;
176  return ff_subtitles_queue_read_packet(&webvtt->q, pkt);
177 }
178 
179 static int webvtt_read_seek(AVFormatContext *s, int stream_index,
180  int64_t min_ts, int64_t ts, int64_t max_ts, int flags)
181 {
182  WebVTTContext *webvtt = s->priv_data;
183  return ff_subtitles_queue_seek(&webvtt->q, s, stream_index,
184  min_ts, ts, max_ts, flags);
185 }
186 
188 {
189  WebVTTContext *webvtt = s->priv_data;
190  ff_subtitles_queue_clean(&webvtt->q);
191  return 0;
192 }
193 
194 #define OFFSET(x) offsetof(WebVTTContext, x)
195 #define KIND_FLAGS AV_OPT_FLAG_SUBTITLE_PARAM|AV_OPT_FLAG_DECODING_PARAM
196 
197 static const AVOption options[] = {
198  { "kind", "Set kind of WebVTT track", OFFSET(kind), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, KIND_FLAGS, "webvtt_kind" },
199  { "subtitles", "WebVTT subtitles kind", 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, INT_MIN, INT_MAX, KIND_FLAGS, "webvtt_kind" },
200  { "captions", "WebVTT captions kind", 0, AV_OPT_TYPE_CONST, { .i64 = AV_DISPOSITION_CAPTIONS }, INT_MIN, INT_MAX, KIND_FLAGS, "webvtt_kind" },
201  { "descriptions", "WebVTT descriptions kind", 0, AV_OPT_TYPE_CONST, { .i64 = AV_DISPOSITION_DESCRIPTIONS }, INT_MIN, INT_MAX, KIND_FLAGS, "webvtt_kind" },
202  { "metadata", "WebVTT metadata kind", 0, AV_OPT_TYPE_CONST, { .i64 = AV_DISPOSITION_METADATA }, INT_MIN, INT_MAX, KIND_FLAGS, "webvtt_kind" },
203  { NULL }
204 };
205 
207  .class_name = "WebVTT demuxer",
208  .item_name = av_default_item_name,
209  .option = options,
210  .version = LIBAVUTIL_VERSION_INT,
211 };
212 
214  .name = "webvtt",
215  .long_name = NULL_IF_CONFIG_SMALL("WebVTT subtitle"),
216  .priv_data_size = sizeof(WebVTTContext),
220  .read_seek2 = webvtt_read_seek,
222  .extensions = "vtt",
223  .priv_class = &webvtt_demuxer_class,
224 };
#define AV_DISPOSITION_METADATA
Definition: avformat.h:862
#define NULL
Definition: coverity.c:32
static int webvtt_read_close(AVFormatContext *s)
Definition: webvttdec.c:187
The optional first identifier line of a WebVTT cue.
Definition: packet.h:196
AVOption.
Definition: opt.h:248
#define KIND_FLAGS
Definition: webvttdec.c:195
static int webvtt_read_header(AVFormatContext *s)
Definition: webvttdec.c:60
#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:4871
void ff_subtitles_read_chunk(AVIOContext *pb, AVBPrint *buf)
Same as ff_subtitles_read_text_chunk(), but read from an AVIOContext.
Definition: subtitles.c:402
static const AVOption options[]
Definition: webvttdec.c:197
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
void ff_subtitles_queue_clean(FFDemuxSubtitlesQueue *q)
Remove and destroy all the subtitles packets.
Definition: subtitles.c:302
static AVPacket pkt
static int webvtt_read_seek(AVFormatContext *s, int stream_index, int64_t min_ts, int64_t ts, int64_t max_ts, int flags)
Definition: webvttdec.c:179
int av_bprint_finalize(AVBPrint *buf, char **ret_str)
Finalize a print buffer.
Definition: bprint.c:235
Format I/O context.
Definition: avformat.h:1239
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 int webvtt_probe(const AVProbeData *p)
Definition: webvttdec.c:40
uint8_t
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
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:92
int64_t duration
Duration of this packet in AVStream->time_base units, 0 if unknown.
Definition: packet.h:381
AVStream * avformat_new_stream(AVFormatContext *s, const AVCodec *c)
Add a new stream to a media file.
Definition: utils.c:4453
static const AVClass webvtt_demuxer_class
Definition: webvttdec.c:206
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_DISPOSITION_CAPTIONS
To specify text track kind (different from subtitles default).
Definition: avformat.h:860
#define AV_BPRINT_SIZE_UNLIMITED
AVInputFormat ff_webvtt_demuxer
Definition: webvttdec.c:213
#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
enum AVMediaType codec_type
General type of the encoded data.
Definition: codec_par.h:56
unsigned char * buf
Buffer must have AVPROBE_PADDING_SIZE of extra allocated bytes filled with zero.
Definition: avformat.h:443
#define ss(width, name, subs,...)
Definition: cbs_vp9.c:261
#define s(width, name)
Definition: cbs_vp9.c:257
static int read_header(FFV1Context *f)
Definition: ffv1dec.c:527
Stream structure.
Definition: avformat.h:880
static int read_packet(void *opaque, uint8_t *buf, int buf_size)
Definition: avio_reading.c:42
#define SET_SIDE_DATA(name, type)
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_DISPOSITION_DESCRIPTIONS
Definition: avformat.h:861
uint64_t_TMPL AV_WL64 unsigned int_TMPL AV_WL32 unsigned int_TMPL AV_WL24 unsigned int_TMPL AV_WL16 uint64_t_TMPL AV_WB64 unsigned int_TMPL AV_WB32 unsigned int_TMPL AV_RB24
Definition: bytestream.h:91
AVIOContext * pb
I/O context.
Definition: avformat.h:1281
Describe the class of an AVClass context structure.
Definition: log.h:67
#define OFFSET(x)
Definition: webvttdec.c:194
This structure contains the data a format has to probe a file.
Definition: avformat.h:441
static int64_t read_ts(const char *s)
Definition: webvttdec.c:52
#define flags(name, subs,...)
Definition: cbs_av1.c:560
static int read_probe(const AVProbeData *pd)
Definition: jvdec.c:55
#define AVPROBE_SCORE_MAX
maximum score
Definition: avformat.h:453
Main libavformat public API header.
int disposition
AV_DISPOSITION_* bit field.
Definition: avformat.h:933
FFDemuxSubtitlesQueue q
Definition: webvttdec.c:36
The optional settings (rendering instructions) that immediately follow the timestamp specifier of a W...
Definition: packet.h:202
void * priv_data
Format private data.
Definition: avformat.h:1267
const char * name
A comma separated list of short names for the format.
Definition: avformat.h:652
static float sub(float src0, float src1)
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1045
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 int webvtt_read_packet(AVFormatContext *s, AVPacket *pkt)
Definition: webvttdec.c:173
This structure stores compressed data.
Definition: packet.h:340
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