FFmpeg
srtdec.c
Go to the documentation of this file.
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 "internal.h"
25 #include "subtitles.h"
26 #include "libavutil/bprint.h"
27 #include "libavutil/intreadwrite.h"
28 
29 typedef struct {
31 } SRTContext;
32 
33 static int srt_probe(const AVProbeData *p)
34 {
35  int v;
36  char buf[64], *pbuf;
37  FFTextReader tr;
38 
39  ff_text_init_buf(&tr, p->buf, p->buf_size);
40 
41  while (ff_text_peek_r8(&tr) == '\r' || ff_text_peek_r8(&tr) == '\n')
42  ff_text_r8(&tr);
43 
44  /* Check if the first non-empty line is a number. We do not check what the
45  * number is because in practice it can be anything.
46  * Also, that number can be followed by random garbage, so we can not
47  * unfortunately check that we only have a number. */
48  if (ff_subtitles_read_line(&tr, buf, sizeof(buf)) < 0 ||
49  strtol(buf, &pbuf, 10) < 0 || pbuf == buf)
50  return 0;
51 
52  /* Check if the next line matches a SRT timestamp */
53  if (ff_subtitles_read_line(&tr, buf, sizeof(buf)) < 0)
54  return 0;
55  pbuf = buf;
56  if (buf[0] == '-')
57  pbuf++;
58  if (pbuf[0] >= '0' && pbuf[0] <= '9' && strstr(buf, " --> ")
59  && sscanf(buf, "%*d:%*d:%*d%*1[,.]%*d --> %*d:%*d:%*d%*1[,.]%d", &v) == 1)
60  return AVPROBE_SCORE_MAX;
61 
62  return 0;
63 }
64 
65 struct event_info {
67  int duration;
68  int64_t pts;
69  int64_t pos;
70 };
71 
72 static int get_event_info(const char *line, struct event_info *ei)
73 {
74  int hh1, mm1, ss1, ms1;
75  int hh2, mm2, ss2, ms2;
76 
77  ei->x1 = ei->x2 = ei->y1 = ei->y2 = ei->duration = -1;
78  ei->pts = AV_NOPTS_VALUE;
79  ei->pos = -1;
80  if (sscanf(line, "%d:%d:%d%*1[,.]%d --> %d:%d:%d%*1[,.]%d"
81  "%*[ ]X1:%"PRId32" X2:%"PRId32" Y1:%"PRId32" Y2:%"PRId32,
82  &hh1, &mm1, &ss1, &ms1,
83  &hh2, &mm2, &ss2, &ms2,
84  &ei->x1, &ei->x2, &ei->y1, &ei->y2) >= 8) {
85  const int64_t start = (hh1*3600LL + mm1*60LL + ss1) * 1000LL + ms1;
86  const int64_t end = (hh2*3600LL + mm2*60LL + ss2) * 1000LL + ms2;
87  ei->duration = end - start;
88  ei->pts = start;
89  return 0;
90  }
91  return -1;
92 }
93 
94 static int add_event(FFDemuxSubtitlesQueue *q, AVBPrint *buf, char *line_cache,
95  const struct event_info *ei, int append_cache)
96 {
97  if (append_cache && line_cache[0])
98  av_bprintf(buf, "%s\n", line_cache);
99  line_cache[0] = 0;
100 
101  while (buf->len > 0 && buf->str[buf->len - 1] == '\n')
102  buf->str[--buf->len] = 0;
103 
104  if (buf->len) {
105  AVPacket *sub = ff_subtitles_queue_insert(q, buf->str, buf->len, 0);
106  if (!sub)
107  return AVERROR(ENOMEM);
108  av_bprint_clear(buf);
109  sub->pos = ei->pos;
110  sub->pts = ei->pts;
111  sub->duration = ei->duration;
112  if (ei->x1 != -1) {
114  if (p) {
115  AV_WL32(p, ei->x1);
116  AV_WL32(p + 4, ei->y1);
117  AV_WL32(p + 8, ei->x2);
118  AV_WL32(p + 12, ei->y2);
119  }
120  }
121  }
122 
123  return 0;
124 }
125 
127 {
128  SRTContext *srt = s->priv_data;
129  AVBPrint buf;
131  int res = 0;
132  char line[4096], line_cache[4096];
133  int has_event_info = 0;
134  struct event_info ei;
135  FFTextReader tr;
136  ff_text_init_avio(s, &tr, s->pb);
137 
138  if (!st)
139  return AVERROR(ENOMEM);
140  avpriv_set_pts_info(st, 64, 1, 1000);
143 
145 
146  line_cache[0] = 0;
147 
148  while (!ff_text_eof(&tr)) {
149  struct event_info tmp_ei;
150  const int64_t pos = ff_text_pos(&tr);
151  ptrdiff_t len = ff_subtitles_read_line(&tr, line, sizeof(line));
152 
153  if (len < 0)
154  break;
155 
156  if (!len || !line[0])
157  continue;
158 
159  if (get_event_info(line, &tmp_ei) < 0) {
160  char *pline;
161 
162  if (!has_event_info)
163  continue;
164 
165  if (line_cache[0]) {
166  /* We got some cache and a new line so we assume the cached
167  * line was actually part of the payload */
168  av_bprintf(&buf, "%s\n", line_cache);
169  line_cache[0] = 0;
170  }
171 
172  /* If the line doesn't start with a number, we assume it's part of
173  * the payload, otherwise is likely an event number preceding the
174  * timing information... but we can't be sure of this yet, so we
175  * cache it */
176  if (strtol(line, &pline, 10) < 0 || line == pline)
177  av_bprintf(&buf, "%s\n", line);
178  else
179  strcpy(line_cache, line);
180  } else {
181  if (has_event_info) {
182  /* We have the information of previous event, append it to the
183  * queue. We insert the cached line if and only if the payload
184  * is empty and the cached line is not a standalone number. */
185  char *pline = NULL;
186  const int standalone_number = strtol(line_cache, &pline, 10) >= 0 && pline && !*pline;
187  res = add_event(&srt->q, &buf, line_cache, &ei, !buf.len && !standalone_number);
188  if (res < 0)
189  goto end;
190  } else {
191  has_event_info = 1;
192  }
193  tmp_ei.pos = pos;
194  ei = tmp_ei;
195  }
196  }
197 
198  /* Append the last event. Here we force the cache to be flushed, because a
199  * trailing number is more likely to be geniune (for example a copyright
200  * date) and not the event index of an inexistant event */
201  if (has_event_info) {
202  res = add_event(&srt->q, &buf, line_cache, &ei, 1);
203  if (res < 0)
204  goto end;
205  }
206 
208 
209 end:
210  if (res < 0)
212  av_bprint_finalize(&buf, NULL);
213  return res;
214 }
215 
217 {
218  SRTContext *srt = s->priv_data;
219  return ff_subtitles_queue_read_packet(&srt->q, pkt);
220 }
221 
222 static int srt_read_seek(AVFormatContext *s, int stream_index,
223  int64_t min_ts, int64_t ts, int64_t max_ts, int flags)
224 {
225  SRTContext *srt = s->priv_data;
226  return ff_subtitles_queue_seek(&srt->q, s, stream_index,
227  min_ts, ts, max_ts, flags);
228 }
229 
231 {
232  SRTContext *srt = s->priv_data;
234  return 0;
235 }
236 
238  .name = "srt",
239  .long_name = NULL_IF_CONFIG_SMALL("SubRip subtitle"),
240  .priv_data_size = sizeof(SRTContext),
244  .read_seek2 = srt_read_seek,
246 };
AVMEDIA_TYPE_SUBTITLE
@ AVMEDIA_TYPE_SUBTITLE
Definition: avutil.h:204
AV_BPRINT_SIZE_UNLIMITED
#define AV_BPRINT_SIZE_UNLIMITED
ff_text_r8
int ff_text_r8(FFTextReader *r)
Return the next byte.
Definition: subtitles.c:65
AVERROR
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
avformat_new_stream
AVStream * avformat_new_stream(AVFormatContext *s, const AVCodec *c)
Add a new stream to a media file.
Definition: utils.c:4526
AVCodecParameters::codec_type
enum AVMediaType codec_type
General type of the encoded data.
Definition: codec_par.h:56
AV_WL32
#define AV_WL32(p, v)
Definition: intreadwrite.h:426
av_bprint_finalize
int av_bprint_finalize(AVBPrint *buf, char **ret_str)
Finalize a print buffer.
Definition: bprint.c:235
av_bprint_init
void av_bprint_init(AVBPrint *buf, unsigned size_init, unsigned size_max)
Definition: bprint.c:69
get_event_info
static int get_event_info(const char *line, struct event_info *ei)
Definition: srtdec.c:72
end
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:92
event_info
Definition: srtdec.c:65
SRTContext
Definition: srtenc.c:32
add_event
static int add_event(FFDemuxSubtitlesQueue *q, AVBPrint *buf, char *line_cache, const struct event_info *ei, int append_cache)
Definition: srtdec.c:94
SRTContext::q
FFDemuxSubtitlesQueue q
Definition: srtdec.c:30
AVPacket::duration
int64_t duration
Duration of this packet in AVStream->time_base units, 0 if unknown.
Definition: packet.h:373
FFTextReader
Definition: subtitles.h:40
AVProbeData::buf_size
int buf_size
Size of buf except extra allocated bytes.
Definition: avformat.h:444
ff_subtitles_read_line
ptrdiff_t ff_subtitles_read_line(FFTextReader *tr, char *buf, size_t size)
Read a line of text.
Definition: subtitles.c:411
srt_read_close
static int srt_read_close(AVFormatContext *s)
Definition: srtdec.c:230
AVPROBE_SCORE_MAX
#define AVPROBE_SCORE_MAX
maximum score
Definition: avformat.h:453
read_close
static av_cold int read_close(AVFormatContext *ctx)
Definition: libcdio.c:145
event_info::pts
int64_t pts
Definition: srtdec.c:68
ff_subtitles_queue_seek
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
AVInputFormat
Definition: avformat.h:636
ff_text_init_buf
void ff_text_init_buf(FFTextReader *r, void *buf, size_t size)
Similar to ff_text_init_avio(), but sets it up to read from a bounded buffer.
Definition: subtitles.c:53
event_info::y1
int32_t y1
Definition: srtdec.c:66
intreadwrite.h
srt_probe
static int srt_probe(const AVProbeData *p)
Definition: srtdec.c:33
s
#define s(width, name)
Definition: cbs_vp9.c:257
AVInputFormat::name
const char * name
A comma separated list of short names for the format.
Definition: avformat.h:641
AVProbeData::buf
unsigned char * buf
Buffer must have AVPROBE_PADDING_SIZE of extra allocated bytes filled with zero.
Definition: avformat.h:443
ff_subtitles_queue_read_packet
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
av_packet_new_side_data
uint8_t * av_packet_new_side_data(AVPacket *pkt, enum AVPacketSideDataType type, int size)
Allocate new information of a packet.
Definition: avpacket.c:332
int32_t
int32_t
Definition: audio_convert.c:194
event_info::duration
int duration
Definition: srtdec.c:67
AVFormatContext
Format I/O context.
Definition: avformat.h:1335
internal.h
AVStream::codecpar
AVCodecParameters * codecpar
Codec parameters associated with this stream.
Definition: avformat.h:1012
read_header
static int read_header(FFV1Context *f)
Definition: ffv1dec.c:527
NULL
#define NULL
Definition: coverity.c:32
read_probe
static int read_probe(const AVProbeData *pd)
Definition: jvdec.c:55
AVProbeData
This structure contains the data a format has to probe a file.
Definition: avformat.h:441
AV_PKT_DATA_SUBTITLE_POSITION
@ AV_PKT_DATA_SUBTITLE_POSITION
Subtitle event position.
Definition: packet.h:183
ff_subtitles_queue_insert
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
ff_text_init_avio
void ff_text_init_avio(void *s, FFTextReader *r, AVIOContext *pb)
Initialize the FFTextReader from the given AVIOContext.
Definition: subtitles.c:27
NULL_IF_CONFIG_SMALL
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
Definition: internal.h:188
avpriv_set_pts_info
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:4948
AV_NOPTS_VALUE
#define AV_NOPTS_VALUE
Undefined timestamp value.
Definition: avutil.h:248
ff_subtitles_queue_finalize
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
FFDemuxSubtitlesQueue
Definition: subtitles.h:102
ff_subtitles_queue_clean
void ff_subtitles_queue_clean(FFDemuxSubtitlesQueue *q)
Remove and destroy all the subtitles packets.
Definition: subtitles.c:302
line
Definition: graph2dot.c:48
event_info::pos
int64_t pos
Definition: srtdec.c:69
bprint.h
AVPacket::pts
int64_t pts
Presentation timestamp in AVStream->time_base units; the time at which the decompressed packet will b...
Definition: packet.h:348
ff_text_pos
int64_t ff_text_pos(FFTextReader *r)
Return the byte position of the next byte returned by ff_text_r8().
Definition: subtitles.c:60
AV_CODEC_ID_SUBRIP
@ AV_CODEC_ID_SUBRIP
Definition: codec_id.h:526
uint8_t
uint8_t
Definition: audio_convert.c:194
len
int len
Definition: vorbis_enc_data.h:452
read_packet
static int read_packet(void *opaque, uint8_t *buf, int buf_size)
Definition: avio_reading.c:42
AVStream
Stream structure.
Definition: avformat.h:865
event_info::x2
int32_t x2
Definition: srtdec.c:66
srt_read_header
static int srt_read_header(AVFormatContext *s)
Definition: srtdec.c:126
pos
unsigned int pos
Definition: spdifenc.c:412
avformat.h
av_bprintf
void av_bprintf(AVBPrint *buf, const char *fmt,...)
Definition: bprint.c:94
ff_srt_demuxer
AVInputFormat ff_srt_demuxer
Definition: srtdec.c:237
ff_text_peek_r8
int ff_text_peek_r8(FFTextReader *r)
Like ff_text_r8(), but don't remove the byte from the buffer.
Definition: subtitles.c:97
event_info::y2
int32_t y2
Definition: srtdec.c:66
av_bprint_clear
void av_bprint_clear(AVBPrint *buf)
Reset the string to "" but keep internal allocated data.
Definition: bprint.c:227
subtitles.h
pkt
static AVPacket pkt
Definition: demuxing_decoding.c:54
ff_text_eof
int ff_text_eof(FFTextReader *r)
Return non-zero if EOF was reached.
Definition: subtitles.c:92
srt_read_packet
static int srt_read_packet(AVFormatContext *s, AVPacket *pkt)
Definition: srtdec.c:216
srt_read_seek
static int srt_read_seek(AVFormatContext *s, int stream_index, int64_t min_ts, int64_t ts, int64_t max_ts, int flags)
Definition: srtdec.c:222
event_info::x1
int32_t x1
Definition: srtdec.c:66
AVCodecParameters::codec_id
enum AVCodecID codec_id
Specific type of the encoded data (the codec used).
Definition: codec_par.h:60
AVPacket
This structure stores compressed data.
Definition: packet.h:332
AVPacket::pos
int64_t pos
byte position in stream, -1 if unknown
Definition: packet.h:375
flags
#define flags(name, subs,...)
Definition: cbs_av1.c:565