Add support for ID3v2 year tag
[libav.git] / libavformat / mp3.c
1 /*
2 * MP3 muxer and demuxer
3 * Copyright (c) 2003 Fabrice Bellard.
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 #include "avformat.h"
22 #include "mpegaudio.h"
23 #include "avstring.h"
24 #include "mpegaudiodecheader.h"
25
26 #define ID3v2_HEADER_SIZE 10
27 #define ID3v1_TAG_SIZE 128
28
29 #define ID3v1_GENRE_MAX 125
30
31 static const char *id3v1_genre_str[ID3v1_GENRE_MAX + 1] = {
32 [0] = "Blues",
33 [1] = "Classic Rock",
34 [2] = "Country",
35 [3] = "Dance",
36 [4] = "Disco",
37 [5] = "Funk",
38 [6] = "Grunge",
39 [7] = "Hip-Hop",
40 [8] = "Jazz",
41 [9] = "Metal",
42 [10] = "New Age",
43 [11] = "Oldies",
44 [12] = "Other",
45 [13] = "Pop",
46 [14] = "R&B",
47 [15] = "Rap",
48 [16] = "Reggae",
49 [17] = "Rock",
50 [18] = "Techno",
51 [19] = "Industrial",
52 [20] = "Alternative",
53 [21] = "Ska",
54 [22] = "Death Metal",
55 [23] = "Pranks",
56 [24] = "Soundtrack",
57 [25] = "Euro-Techno",
58 [26] = "Ambient",
59 [27] = "Trip-Hop",
60 [28] = "Vocal",
61 [29] = "Jazz+Funk",
62 [30] = "Fusion",
63 [31] = "Trance",
64 [32] = "Classical",
65 [33] = "Instrumental",
66 [34] = "Acid",
67 [35] = "House",
68 [36] = "Game",
69 [37] = "Sound Clip",
70 [38] = "Gospel",
71 [39] = "Noise",
72 [40] = "AlternRock",
73 [41] = "Bass",
74 [42] = "Soul",
75 [43] = "Punk",
76 [44] = "Space",
77 [45] = "Meditative",
78 [46] = "Instrumental Pop",
79 [47] = "Instrumental Rock",
80 [48] = "Ethnic",
81 [49] = "Gothic",
82 [50] = "Darkwave",
83 [51] = "Techno-Industrial",
84 [52] = "Electronic",
85 [53] = "Pop-Folk",
86 [54] = "Eurodance",
87 [55] = "Dream",
88 [56] = "Southern Rock",
89 [57] = "Comedy",
90 [58] = "Cult",
91 [59] = "Gangsta",
92 [60] = "Top 40",
93 [61] = "Christian Rap",
94 [62] = "Pop/Funk",
95 [63] = "Jungle",
96 [64] = "Native American",
97 [65] = "Cabaret",
98 [66] = "New Wave",
99 [67] = "Psychadelic",
100 [68] = "Rave",
101 [69] = "Showtunes",
102 [70] = "Trailer",
103 [71] = "Lo-Fi",
104 [72] = "Tribal",
105 [73] = "Acid Punk",
106 [74] = "Acid Jazz",
107 [75] = "Polka",
108 [76] = "Retro",
109 [77] = "Musical",
110 [78] = "Rock & Roll",
111 [79] = "Hard Rock",
112 [80] = "Folk",
113 [81] = "Folk-Rock",
114 [82] = "National Folk",
115 [83] = "Swing",
116 [84] = "Fast Fusion",
117 [85] = "Bebob",
118 [86] = "Latin",
119 [87] = "Revival",
120 [88] = "Celtic",
121 [89] = "Bluegrass",
122 [90] = "Avantgarde",
123 [91] = "Gothic Rock",
124 [92] = "Progressive Rock",
125 [93] = "Psychedelic Rock",
126 [94] = "Symphonic Rock",
127 [95] = "Slow Rock",
128 [96] = "Big Band",
129 [97] = "Chorus",
130 [98] = "Easy Listening",
131 [99] = "Acoustic",
132 [100] = "Humour",
133 [101] = "Speech",
134 [102] = "Chanson",
135 [103] = "Opera",
136 [104] = "Chamber Music",
137 [105] = "Sonata",
138 [106] = "Symphony",
139 [107] = "Booty Bass",
140 [108] = "Primus",
141 [109] = "Porn Groove",
142 [110] = "Satire",
143 [111] = "Slow Jam",
144 [112] = "Club",
145 [113] = "Tango",
146 [114] = "Samba",
147 [115] = "Folklore",
148 [116] = "Ballad",
149 [117] = "Power Ballad",
150 [118] = "Rhythmic Soul",
151 [119] = "Freestyle",
152 [120] = "Duet",
153 [121] = "Punk Rock",
154 [122] = "Drum Solo",
155 [123] = "A capella",
156 [124] = "Euro-House",
157 [125] = "Dance Hall",
158 };
159
160 /* buf must be ID3v2_HEADER_SIZE byte long */
161 static int id3v2_match(const uint8_t *buf)
162 {
163 return (buf[0] == 'I' &&
164 buf[1] == 'D' &&
165 buf[2] == '3' &&
166 buf[3] != 0xff &&
167 buf[4] != 0xff &&
168 (buf[6] & 0x80) == 0 &&
169 (buf[7] & 0x80) == 0 &&
170 (buf[8] & 0x80) == 0 &&
171 (buf[9] & 0x80) == 0);
172 }
173
174 static unsigned int id3v2_get_size(ByteIOContext *s, int len)
175 {
176 int v=0;
177 while(len--)
178 v= (v<<7) + (get_byte(s)&0x7F);
179 return v;
180 }
181
182 static void id3v2_read_ttag(AVFormatContext *s, int taglen, char *dst, int dstlen)
183 {
184 char *q;
185 int len;
186
187 if(taglen < 1)
188 return;
189
190 taglen--; /* account for encoding type byte */
191 dstlen--; /* Leave space for zero terminator */
192
193 switch(get_byte(&s->pb)) { /* encoding type */
194
195 case 0: /* ISO-8859-1 (0 - 255 maps directly into unicode) */
196 q = dst;
197 while(taglen--) {
198 uint8_t tmp;
199 PUT_UTF8(get_byte(&s->pb), tmp, if (q - dst < dstlen - 1) *q++ = tmp;)
200 }
201 *q = '\0';
202 break;
203
204 case 3: /* UTF-8 */
205 len = FFMIN(taglen, dstlen);
206 get_buffer(&s->pb, dst, len);
207 dst[len] = 0;
208 break;
209 }
210 }
211
212 /**
213 * ID3v2 parser
214 *
215 * Handles ID3v2.2, 2.3 and 2.4.
216 *
217 */
218
219 static void id3v2_parse(AVFormatContext *s, int len, uint8_t version, uint8_t flags)
220 {
221 int isv34, tlen;
222 uint32_t tag;
223 offset_t next;
224 char tmp[16];
225 int taghdrlen;
226 const char *reason;
227
228 switch(version) {
229 case 2:
230 if(flags & 0x40) {
231 reason = "compression";
232 goto error;
233 }
234 isv34 = 0;
235 taghdrlen = 6;
236 break;
237
238 case 3:
239 case 4:
240 isv34 = 1;
241 taghdrlen = 10;
242 break;
243
244 default:
245 reason = "version";
246 goto error;
247 }
248
249 if(flags & 0x80) {
250 reason = "unsynchronization";
251 goto error;
252 }
253
254 if(isv34 && flags & 0x40) /* Extended header present, just skip over it */
255 url_fskip(&s->pb, id3v2_get_size(&s->pb, 4));
256
257 while(len >= taghdrlen) {
258 if(isv34) {
259 tag = get_be32(&s->pb);
260 tlen = id3v2_get_size(&s->pb, 4);
261 get_be16(&s->pb); /* flags */
262 } else {
263 tag = get_be24(&s->pb);
264 tlen = id3v2_get_size(&s->pb, 3);
265 }
266 len -= taghdrlen + tlen;
267
268 if(len < 0)
269 break;
270
271 next = url_ftell(&s->pb) + tlen;
272
273 switch(tag) {
274 case MKBETAG('T', 'I', 'T', '2'):
275 case MKBETAG(0, 'T', 'T', '2'):
276 id3v2_read_ttag(s, tlen, s->title, sizeof(s->title));
277 break;
278 case MKBETAG('T', 'P', 'E', '1'):
279 case MKBETAG(0, 'T', 'P', '1'):
280 id3v2_read_ttag(s, tlen, s->author, sizeof(s->author));
281 break;
282 case MKBETAG('T', 'A', 'L', 'B'):
283 case MKBETAG(0, 'T', 'A', 'L'):
284 id3v2_read_ttag(s, tlen, s->album, sizeof(s->album));
285 break;
286 case MKBETAG('T', 'C', 'O', 'N'):
287 case MKBETAG(0, 'T', 'C', 'O'):
288 id3v2_read_ttag(s, tlen, s->genre, sizeof(s->genre));
289 break;
290 case MKBETAG('T', 'C', 'O', 'P'):
291 case MKBETAG(0, 'T', 'C', 'R'):
292 id3v2_read_ttag(s, tlen, s->copyright, sizeof(s->copyright));
293 break;
294 case MKBETAG('T', 'R', 'C', 'K'):
295 case MKBETAG(0, 'T', 'R', 'K'):
296 id3v2_read_ttag(s, tlen, tmp, sizeof(tmp));
297 s->track = atoi(tmp);
298 break;
299 case 0:
300 /* padding, skip to end */
301 url_fskip(&s->pb, len);
302 len = 0;
303 continue;
304 }
305 /* Skip to end of tag */
306 url_fseek(&s->pb, next, SEEK_SET);
307 }
308
309 if(version == 4 && flags & 0x10) /* Footer preset, always 10 bytes, skip over it */
310 url_fskip(&s->pb, 10);
311 return;
312
313 error:
314 av_log(s, AV_LOG_INFO, "ID3v2.%d tag skipped, cannot handle %s\n", version, reason);
315 url_fskip(&s->pb, len);
316 }
317
318 static void id3v1_get_string(char *str, int str_size,
319 const uint8_t *buf, int buf_size)
320 {
321 int i, c;
322 char *q;
323
324 q = str;
325 for(i = 0; i < buf_size; i++) {
326 c = buf[i];
327 if (c == '\0')
328 break;
329 if ((q - str) >= str_size - 1)
330 break;
331 *q++ = c;
332 }
333 *q = '\0';
334 }
335
336 /* 'buf' must be ID3v1_TAG_SIZE byte long */
337 static int id3v1_parse_tag(AVFormatContext *s, const uint8_t *buf)
338 {
339 char str[5];
340 int genre;
341
342 if (!(buf[0] == 'T' &&
343 buf[1] == 'A' &&
344 buf[2] == 'G'))
345 return -1;
346 id3v1_get_string(s->title, sizeof(s->title), buf + 3, 30);
347 id3v1_get_string(s->author, sizeof(s->author), buf + 33, 30);
348 id3v1_get_string(s->album, sizeof(s->album), buf + 63, 30);
349 id3v1_get_string(str, sizeof(str), buf + 93, 4);
350 s->year = atoi(str);
351 id3v1_get_string(s->comment, sizeof(s->comment), buf + 97, 30);
352 if (buf[125] == 0 && buf[126] != 0)
353 s->track = buf[126];
354 genre = buf[127];
355 if (genre <= ID3v1_GENRE_MAX)
356 av_strlcpy(s->genre, id3v1_genre_str[genre], sizeof(s->genre));
357 return 0;
358 }
359
360 static void id3v1_create_tag(AVFormatContext *s, uint8_t *buf)
361 {
362 int v, i;
363
364 memset(buf, 0, ID3v1_TAG_SIZE); /* fail safe */
365 buf[0] = 'T';
366 buf[1] = 'A';
367 buf[2] = 'G';
368 strncpy(buf + 3, s->title, 30);
369 strncpy(buf + 33, s->author, 30);
370 strncpy(buf + 63, s->album, 30);
371 v = s->year;
372 if (v > 0) {
373 for(i = 0;i < 4; i++) {
374 buf[96 - i] = '0' + (v % 10);
375 v = v / 10;
376 }
377 }
378 strncpy(buf + 97, s->comment, 30);
379 if (s->track != 0) {
380 buf[125] = 0;
381 buf[126] = s->track;
382 }
383 for(i = 0; i <= ID3v1_GENRE_MAX; i++) {
384 if (!strcasecmp(s->genre, id3v1_genre_str[i])) {
385 buf[127] = i;
386 break;
387 }
388 }
389 }
390
391 /* mp3 read */
392
393 static int mp3_read_probe(AVProbeData *p)
394 {
395 int max_frames, first_frames = 0;
396 int fsize, frames, sample_rate;
397 uint32_t header;
398 uint8_t *buf, *buf2, *end;
399 AVCodecContext avctx;
400
401 if(id3v2_match(p->buf))
402 return AVPROBE_SCORE_MAX/2+1; // this must be less than mpeg-ps because some retards put id3v2 tags before mpeg-ps files
403
404 max_frames = 0;
405 buf = p->buf;
406 end = buf + FFMIN(4096, p->buf_size - sizeof(uint32_t));
407
408 for(; buf < end; buf++) {
409 buf2 = buf;
410
411 for(frames = 0; buf2 < end; frames++) {
412 header = AV_RB32(buf2);
413 fsize = ff_mpa_decode_header(&avctx, header, &sample_rate);
414 if(fsize < 0)
415 break;
416 buf2 += fsize;
417 }
418 max_frames = FFMAX(max_frames, frames);
419 if(buf == p->buf)
420 first_frames= frames;
421 }
422 if (first_frames>=3) return AVPROBE_SCORE_MAX/2+1;
423 else if(max_frames>=3) return AVPROBE_SCORE_MAX/4;
424 else if(max_frames>=1) return 1;
425 else return 0;
426 }
427
428 /**
429 * Try to find Xing/Info/VBRI tags and compute duration from info therein
430 */
431 static void mp3_parse_vbr_tags(AVFormatContext *s, AVStream *st, offset_t base)
432 {
433 uint32_t v, spf;
434 int frames = -1; /* Total number of frames in file */
435 const offset_t xing_offtbl[2][2] = {{32, 17}, {17,9}};
436 MPADecodeContext c;
437
438 v = get_be32(&s->pb);
439 if(ff_mpa_check_header(v) < 0)
440 return;
441
442 ff_mpegaudio_decode_header(&c, v);
443 if(c.layer != 3)
444 return;
445
446 /* Check for Xing / Info tag */
447 url_fseek(&s->pb, xing_offtbl[c.lsf == 1][c.nb_channels == 1], SEEK_CUR);
448 v = get_be32(&s->pb);
449 if(v == MKBETAG('X', 'i', 'n', 'g') || v == MKBETAG('I', 'n', 'f', 'o')) {
450 v = get_be32(&s->pb);
451 if(v & 0x1)
452 frames = get_be32(&s->pb);
453 }
454
455 /* Check for VBRI tag (always 32 bytes after end of mpegaudio header) */
456 url_fseek(&s->pb, base + 4 + 32, SEEK_SET);
457 v = get_be32(&s->pb);
458 if(v == MKBETAG('V', 'B', 'R', 'I')) {
459 /* Check tag version */
460 if(get_be16(&s->pb) == 1) {
461 /* skip delay, quality and total bytes */
462 url_fseek(&s->pb, 8, SEEK_CUR);
463 frames = get_be32(&s->pb);
464 }
465 }
466
467 if(frames < 0)
468 return;
469
470 spf = c.lsf ? 576 : 1152; /* Samples per frame, layer 3 */
471 st->duration = av_rescale_q(frames, (AVRational){spf, c.sample_rate},
472 st->time_base);
473 }
474
475 static int mp3_read_header(AVFormatContext *s,
476 AVFormatParameters *ap)
477 {
478 AVStream *st;
479 uint8_t buf[ID3v1_TAG_SIZE];
480 int len, ret, filesize;
481 offset_t off;
482
483 st = av_new_stream(s, 0);
484 if (!st)
485 return AVERROR(ENOMEM);
486
487 st->codec->codec_type = CODEC_TYPE_AUDIO;
488 st->codec->codec_id = CODEC_ID_MP3;
489 st->need_parsing = AVSTREAM_PARSE_FULL;
490 st->start_time = 0;
491
492 /* try to get the TAG */
493 if (!url_is_streamed(&s->pb)) {
494 /* XXX: change that */
495 filesize = url_fsize(&s->pb);
496 if (filesize > 128) {
497 url_fseek(&s->pb, filesize - 128, SEEK_SET);
498 ret = get_buffer(&s->pb, buf, ID3v1_TAG_SIZE);
499 if (ret == ID3v1_TAG_SIZE) {
500 id3v1_parse_tag(s, buf);
501 }
502 url_fseek(&s->pb, 0, SEEK_SET);
503 }
504 }
505
506 /* if ID3v2 header found, skip it */
507 ret = get_buffer(&s->pb, buf, ID3v2_HEADER_SIZE);
508 if (ret != ID3v2_HEADER_SIZE)
509 return -1;
510 if (id3v2_match(buf)) {
511 /* parse ID3v2 header */
512 len = ((buf[6] & 0x7f) << 21) |
513 ((buf[7] & 0x7f) << 14) |
514 ((buf[8] & 0x7f) << 7) |
515 (buf[9] & 0x7f);
516 id3v2_parse(s, len, buf[3], buf[5]);
517 } else {
518 url_fseek(&s->pb, 0, SEEK_SET);
519 }
520
521 off = url_ftell(&s->pb);
522 mp3_parse_vbr_tags(s, st, off);
523 url_fseek(&s->pb, off, SEEK_SET);
524
525 /* the parameters will be extracted from the compressed bitstream */
526 return 0;
527 }
528
529 #define MP3_PACKET_SIZE 1024
530
531 static int mp3_read_packet(AVFormatContext *s, AVPacket *pkt)
532 {
533 int ret, size;
534 // AVStream *st = s->streams[0];
535
536 size= MP3_PACKET_SIZE;
537
538 ret= av_get_packet(&s->pb, pkt, size);
539
540 pkt->stream_index = 0;
541 if (ret <= 0) {
542 return AVERROR(EIO);
543 }
544 /* note: we need to modify the packet size here to handle the last
545 packet */
546 pkt->size = ret;
547 return ret;
548 }
549
550 static int mp3_read_close(AVFormatContext *s)
551 {
552 return 0;
553 }
554
555 #ifdef CONFIG_MUXERS
556 /* simple formats */
557
558 static void id3v2_put_size(AVFormatContext *s, int size)
559 {
560 put_byte(&s->pb, size >> 21 & 0x7f);
561 put_byte(&s->pb, size >> 14 & 0x7f);
562 put_byte(&s->pb, size >> 7 & 0x7f);
563 put_byte(&s->pb, size & 0x7f);
564 }
565
566 static void id3v2_put_ttag(AVFormatContext *s, char *string, uint32_t tag)
567 {
568 int len = strlen(string);
569 put_be32(&s->pb, tag);
570 id3v2_put_size(s, len + 1);
571 put_be16(&s->pb, 0);
572 put_byte(&s->pb, 3); /* UTF-8 */
573 put_buffer(&s->pb, string, len);
574 }
575
576
577 /**
578 * Write an ID3v2.4 header at beginning of stream
579 */
580
581 static int mp3_write_header(struct AVFormatContext *s)
582 {
583 int totlen = 0;
584 char tracktxt[10];
585 char yeartxt[10];
586
587 if(s->track)
588 snprintf(tracktxt, sizeof(tracktxt) - 1, "%d", s->track);
589 if(s->year)
590 snprintf( yeartxt, sizeof(yeartxt) , "%d", s->year );
591
592 if(s->title[0]) totlen += 11 + strlen(s->title);
593 if(s->author[0]) totlen += 11 + strlen(s->author);
594 if(s->album[0]) totlen += 11 + strlen(s->album);
595 if(s->genre[0]) totlen += 11 + strlen(s->genre);
596 if(s->copyright[0]) totlen += 11 + strlen(s->copyright);
597 if(s->track) totlen += 11 + strlen(tracktxt);
598 if(s->year) totlen += 11 + strlen(yeartxt);
599 if(!(s->streams[0]->codec->flags & CODEC_FLAG_BITEXACT))
600 totlen += strlen(LIBAVFORMAT_IDENT) + 11;
601
602 if(totlen == 0)
603 return 0;
604
605 put_be32(&s->pb, MKBETAG('I', 'D', '3', 0x04)); /* ID3v2.4 */
606 put_byte(&s->pb, 0);
607 put_byte(&s->pb, 0); /* flags */
608
609 id3v2_put_size(s, totlen);
610
611 if(s->title[0]) id3v2_put_ttag(s, s->title, MKBETAG('T', 'I', 'T', '2'));
612 if(s->author[0]) id3v2_put_ttag(s, s->author, MKBETAG('T', 'P', 'E', '1'));
613 if(s->album[0]) id3v2_put_ttag(s, s->album, MKBETAG('T', 'A', 'L', 'B'));
614 if(s->genre[0]) id3v2_put_ttag(s, s->genre, MKBETAG('T', 'C', 'O', 'N'));
615 if(s->copyright[0]) id3v2_put_ttag(s, s->copyright, MKBETAG('T', 'C', 'O', 'P'));
616 if(s->track) id3v2_put_ttag(s, tracktxt, MKBETAG('T', 'R', 'C', 'K'));
617 if(s->year) id3v2_put_ttag(s, yeartxt, MKBETAG('T', 'Y', 'E', 'R'));
618 if(!(s->streams[0]->codec->flags & CODEC_FLAG_BITEXACT))
619 id3v2_put_ttag(s, LIBAVFORMAT_IDENT, MKBETAG('T', 'E', 'N', 'C'));
620 return 0;
621 }
622
623 static int mp3_write_packet(struct AVFormatContext *s, AVPacket *pkt)
624 {
625 put_buffer(&s->pb, pkt->data, pkt->size);
626 put_flush_packet(&s->pb);
627 return 0;
628 }
629
630 static int mp3_write_trailer(struct AVFormatContext *s)
631 {
632 uint8_t buf[ID3v1_TAG_SIZE];
633
634 /* write the id3v1 tag */
635 if (s->title[0] != '\0') {
636 id3v1_create_tag(s, buf);
637 put_buffer(&s->pb, buf, ID3v1_TAG_SIZE);
638 put_flush_packet(&s->pb);
639 }
640 return 0;
641 }
642 #endif //CONFIG_MUXERS
643
644 #ifdef CONFIG_MP3_DEMUXER
645 AVInputFormat mp3_demuxer = {
646 "mp3",
647 "MPEG audio",
648 0,
649 mp3_read_probe,
650 mp3_read_header,
651 mp3_read_packet,
652 mp3_read_close,
653 .flags= AVFMT_GENERIC_INDEX,
654 .extensions = "mp2,mp3,m2a", /* XXX: use probe */
655 };
656 #endif
657 #ifdef CONFIG_MP2_MUXER
658 AVOutputFormat mp2_muxer = {
659 "mp2",
660 "MPEG audio layer 2",
661 "audio/x-mpeg",
662 #ifdef CONFIG_LIBMP3LAME
663 "mp2,m2a",
664 #else
665 "mp2,mp3,m2a",
666 #endif
667 0,
668 CODEC_ID_MP2,
669 0,
670 NULL,
671 mp3_write_packet,
672 mp3_write_trailer,
673 };
674 #endif
675 #ifdef CONFIG_MP3_MUXER
676 AVOutputFormat mp3_muxer = {
677 "mp3",
678 "MPEG audio layer 3",
679 "audio/x-mpeg",
680 "mp3",
681 0,
682 CODEC_ID_MP3,
683 0,
684 mp3_write_header,
685 mp3_write_packet,
686 mp3_write_trailer,
687 };
688 #endif