blob: 0aeb8a63f489bc081f13f8a71e4022346ca1bea2
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; |
36 | FFDemuxSubtitlesQueue q; |
37 | int kind; |
38 | } WebVTTContext; |
39 | |
40 | static int webvtt_probe(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 | |
60 | static int webvtt_read_header(AVFormatContext *s) |
61 | { |
62 | WebVTTContext *webvtt = s->priv_data; |
63 | AVBPrint header, cue; |
64 | int res = 0; |
65 | AVStream *st = avformat_new_stream(s, NULL); |
66 | |
67 | if (!st) |
68 | return AVERROR(ENOMEM); |
69 | avpriv_set_pts_info(st, 64, 1, 1000); |
70 | st->codecpar->codec_type = AVMEDIA_TYPE_SUBTITLE; |
71 | st->codecpar->codec_id = AV_CODEC_ID_WEBVTT; |
72 | st->disposition |= webvtt->kind; |
73 | |
74 | av_bprint_init(&header, 0, AV_BPRINT_SIZE_UNLIMITED); |
75 | av_bprint_init(&cue, 0, AV_BPRINT_SIZE_UNLIMITED); |
76 | |
77 | for (;;) { |
78 | int i; |
79 | int64_t pos; |
80 | AVPacket *sub; |
81 | const char *p, *identifier, *settings; |
82 | int identifier_len, settings_len; |
83 | int64_t ts_start, ts_end; |
84 | |
85 | ff_subtitles_read_chunk(s->pb, &cue); |
86 | |
87 | if (!cue.len) |
88 | break; |
89 | |
90 | p = identifier = cue.str; |
91 | pos = avio_tell(s->pb); |
92 | |
93 | /* ignore header chunk */ |
94 | if (!strncmp(p, "\xEF\xBB\xBFWEBVTT", 9) || |
95 | !strncmp(p, "WEBVTT", 6) || |
96 | !strncmp(p, "NOTE", 4)) |
97 | continue; |
98 | |
99 | /* optional cue identifier (can be a number like in SRT or some kind of |
100 | * chaptering id) */ |
101 | for (i = 0; p[i] && p[i] != '\n' && p[i] != '\r'; i++) { |
102 | if (!strncmp(p + i, "-->", 3)) { |
103 | identifier = NULL; |
104 | break; |
105 | } |
106 | } |
107 | if (!identifier) |
108 | identifier_len = 0; |
109 | else { |
110 | identifier_len = strcspn(p, "\r\n"); |
111 | p += identifier_len; |
112 | if (*p == '\r') |
113 | p++; |
114 | if (*p == '\n') |
115 | p++; |
116 | } |
117 | |
118 | /* cue timestamps */ |
119 | if ((ts_start = read_ts(p)) == AV_NOPTS_VALUE) |
120 | break; |
121 | if (!(p = strstr(p, "-->"))) |
122 | break; |
123 | p += 2; |
124 | do p++; while (*p == ' ' || *p == '\t'); |
125 | if ((ts_end = read_ts(p)) == AV_NOPTS_VALUE) |
126 | break; |
127 | |
128 | /* optional cue settings */ |
129 | p += strcspn(p, "\n\t "); |
130 | while (*p == '\t' || *p == ' ') |
131 | p++; |
132 | settings = p; |
133 | settings_len = strcspn(p, "\r\n"); |
134 | p += settings_len; |
135 | if (*p == '\r') |
136 | p++; |
137 | if (*p == '\n') |
138 | p++; |
139 | |
140 | /* create packet */ |
141 | sub = ff_subtitles_queue_insert(&webvtt->q, p, strlen(p), 0); |
142 | if (!sub) { |
143 | res = AVERROR(ENOMEM); |
144 | goto end; |
145 | } |
146 | sub->pos = pos; |
147 | sub->pts = ts_start; |
148 | sub->duration = ts_end - ts_start; |
149 | |
150 | #define SET_SIDE_DATA(name, type) do { \ |
151 | if (name##_len) { \ |
152 | uint8_t *buf = av_packet_new_side_data(sub, type, name##_len); \ |
153 | if (!buf) { \ |
154 | res = AVERROR(ENOMEM); \ |
155 | goto end; \ |
156 | } \ |
157 | memcpy(buf, name, name##_len); \ |
158 | } \ |
159 | } while (0) |
160 | |
161 | SET_SIDE_DATA(identifier, AV_PKT_DATA_WEBVTT_IDENTIFIER); |
162 | SET_SIDE_DATA(settings, AV_PKT_DATA_WEBVTT_SETTINGS); |
163 | } |
164 | |
165 | ff_subtitles_queue_finalize(s, &webvtt->q); |
166 | |
167 | end: |
168 | av_bprint_finalize(&cue, NULL); |
169 | av_bprint_finalize(&header, NULL); |
170 | return res; |
171 | } |
172 | |
173 | static int webvtt_read_packet(AVFormatContext *s, AVPacket *pkt) |
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 | |
187 | static int webvtt_read_close(AVFormatContext *s) |
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 |
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, 0, "webvtt_kind" }, |
200 | { "captions", "WebVTT captions kind", 0, AV_OPT_TYPE_CONST, { .i64 = AV_DISPOSITION_CAPTIONS }, INT_MIN, INT_MAX, 0, "webvtt_kind" }, |
201 | { "descriptions", "WebVTT descriptions kind", 0, AV_OPT_TYPE_CONST, { .i64 = AV_DISPOSITION_DESCRIPTIONS }, INT_MIN, INT_MAX, 0, "webvtt_kind" }, |
202 | { "metadata", "WebVTT metadata kind", 0, AV_OPT_TYPE_CONST, { .i64 = AV_DISPOSITION_METADATA }, INT_MIN, INT_MAX, 0, "webvtt_kind" }, |
203 | { NULL } |
204 | }; |
205 | |
206 | static const AVClass webvtt_demuxer_class = { |
207 | .class_name = "WebVTT demuxer", |
208 | .item_name = av_default_item_name, |
209 | .option = options, |
210 | .version = LIBAVUTIL_VERSION_INT, |
211 | }; |
212 | |
213 | AVInputFormat ff_webvtt_demuxer = { |
214 | .name = "webvtt", |
215 | .long_name = NULL_IF_CONFIG_SMALL("WebVTT subtitle"), |
216 | .priv_data_size = sizeof(WebVTTContext), |
217 | .read_probe = webvtt_probe, |
218 | .read_header = webvtt_read_header, |
219 | .read_packet = webvtt_read_packet, |
220 | .read_seek2 = webvtt_read_seek, |
221 | .read_close = webvtt_read_close, |
222 | .extensions = "vtt", |
223 | .priv_class = &webvtt_demuxer_class, |
224 | }; |
225 |