3 * Copyright (c) 2003 Fabrice Bellard
5 * This file is part of FFmpeg.
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.
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.
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
24 #include "libavutil/avstring.h"
25 #include "libavutil/intreadwrite.h"
27 #include "avio_internal.h"
29 int ff_id3v2_match(const uint8_t *buf, const char * magic)
31 return buf[0] == magic[0] &&
36 (buf[6] & 0x80) == 0 &&
37 (buf[7] & 0x80) == 0 &&
38 (buf[8] & 0x80) == 0 &&
42 int ff_id3v2_tag_len(const uint8_t * buf)
44 int len = ((buf[6] & 0x7f) << 21) +
45 ((buf[7] & 0x7f) << 14) +
46 ((buf[8] & 0x7f) << 7) +
50 len += ID3v2_HEADER_SIZE;
54 static unsigned int get_size(AVIOContext *s, int len)
58 v = (v << 7) + (avio_r8(s) & 0x7F);
62 static void read_ttag(AVFormatContext *s, AVIOContext *pb, int taglen, const char *key)
65 const char *val = NULL;
66 int len, dstlen = sizeof(dst) - 1;
68 unsigned int (*get)(AVIOContext*) = avio_rb16;
74 taglen--; /* account for encoding type byte */
76 switch (avio_r8(pb)) { /* encoding type */
78 case ID3v2_ENCODING_ISO8859:
80 while (taglen-- && q - dst < dstlen - 7) {
82 PUT_UTF8(avio_r8(pb), tmp, *q++ = tmp;)
87 case ID3v2_ENCODING_UTF16BOM:
89 switch (avio_rb16(pb)) {
95 av_log(s, AV_LOG_ERROR, "Incorrect BOM value in tag %s.\n", key);
100 case ID3v2_ENCODING_UTF16BE:
102 while (taglen > 1 && q - dst < dstlen - 7) {
106 GET_UTF16(ch, ((taglen -= 2) >= 0 ? get(pb) : 0), break;)
107 PUT_UTF8(ch, tmp, *q++ = tmp;)
112 case ID3v2_ENCODING_UTF8:
113 len = FFMIN(taglen, dstlen);
114 avio_read(pb, dst, len);
118 av_log(s, AV_LOG_WARNING, "Unknown encoding in tag %s\n.", key);
121 if (!(strcmp(key, "TCON") && strcmp(key, "TCO"))
122 && (sscanf(dst, "(%d)", &genre) == 1 || sscanf(dst, "%d", &genre) == 1)
123 && genre <= ID3v1_GENRE_MAX)
124 val = ff_id3v1_genre_str[genre];
125 else if (!(strcmp(key, "TXXX") && strcmp(key, "TXX"))) {
126 /* dst now contains two 0-terminated strings */
130 val = dst + FFMIN(len + 1, dstlen);
136 av_metadata_set2(&s->metadata, key, val, AV_METADATA_DONT_OVERWRITE);
139 static void ff_id3v2_parse(AVFormatContext *s, int len, uint8_t version, uint8_t flags)
141 int isv34, tlen, unsync;
147 unsigned char *buffer = NULL;
153 reason = "compression";
171 unsync = flags & 0x80;
173 if (isv34 && flags & 0x40) /* Extended header present, just skip over it */
174 url_fskip(s->pb, get_size(s->pb, 4));
176 while (len >= taghdrlen) {
181 avio_read(s->pb, tag, 4);
184 tlen = avio_rb32(s->pb);
186 tlen = get_size(s->pb, 4);
187 tflags = avio_rb16(s->pb);
188 tunsync = tflags & ID3v2_FLAG_UNSYNCH;
190 avio_read(s->pb, tag, 3);
192 tlen = avio_rb24(s->pb);
194 len -= taghdrlen + tlen;
199 next = url_ftell(s->pb) + tlen;
201 if (tflags & ID3v2_FLAG_DATALEN) {
206 if (tflags & (ID3v2_FLAG_ENCRYPTION | ID3v2_FLAG_COMPRESSION)) {
207 av_log(s, AV_LOG_WARNING, "Skipping encrypted/compressed ID3v2 frame %s.\n", tag);
208 url_fskip(s->pb, tlen);
209 } else if (tag[0] == 'T') {
210 if (unsync || tunsync) {
212 av_fast_malloc(&buffer, &buffer_size, tlen);
213 for (i = 0, j = 0; i < tlen; i++, j++) {
214 buffer[j] = avio_r8(s->pb);
215 if (j > 0 && !buffer[j] && buffer[j - 1] == 0xff) {
216 /* Unsynchronised byte, skip it */
220 ffio_init_context(&pb, buffer, j, 0, NULL, NULL, NULL, NULL);
221 read_ttag(s, &pb, j, tag);
223 read_ttag(s, s->pb, tlen, tag);
228 av_log(s, AV_LOG_WARNING, "invalid frame id, assuming padding");
229 url_fskip(s->pb, tlen);
232 /* Skip to end of tag */
233 url_fseek(s->pb, next, SEEK_SET);
238 url_fskip(s->pb, len);
240 if (version == 4 && flags & 0x10) /* Footer preset, always 10 bytes, skip over it */
241 url_fskip(s->pb, 10);
247 av_log(s, AV_LOG_INFO, "ID3v2.%d tag skipped, cannot handle %s\n", version, reason);
248 url_fskip(s->pb, len);
252 void ff_id3v2_read(AVFormatContext *s, const char *magic)
255 uint8_t buf[ID3v2_HEADER_SIZE];
260 /* save the current offset in case there's nothing to read/skip */
261 off = url_ftell(s->pb);
262 ret = avio_read(s->pb, buf, ID3v2_HEADER_SIZE);
263 if (ret != ID3v2_HEADER_SIZE)
265 found_header = ff_id3v2_match(buf, magic);
267 /* parse ID3v2 header */
268 len = ((buf[6] & 0x7f) << 21) |
269 ((buf[7] & 0x7f) << 14) |
270 ((buf[8] & 0x7f) << 7) |
272 ff_id3v2_parse(s, len, buf[3], buf[5]);
274 url_fseek(s->pb, off, SEEK_SET);
276 } while (found_header);
277 ff_metadata_conv(&s->metadata, NULL, ff_id3v2_34_metadata_conv);
278 ff_metadata_conv(&s->metadata, NULL, ff_id3v2_2_metadata_conv);
279 ff_metadata_conv(&s->metadata, NULL, ff_id3v2_4_metadata_conv);
282 const AVMetadataConv ff_id3v2_34_metadata_conv[] = {
284 { "TCOM", "composer"},
286 { "TCOP", "copyright"},
287 { "TENC", "encoded_by"},
289 { "TLAN", "language"},
291 { "TPE2", "album_artist"},
292 { "TPE3", "performer"},
294 { "TPUB", "publisher"},
296 { "TSSE", "encoder"},
300 const AVMetadataConv ff_id3v2_4_metadata_conv[] = {
303 { "TDEN", "creation_time"},
304 { "TSOA", "album-sort"},
305 { "TSOP", "artist-sort"},
306 { "TSOT", "title-sort"},
310 const AVMetadataConv ff_id3v2_2_metadata_conv[] = {
314 { "TEN", "encoded_by"},
316 { "TP2", "album_artist"},
317 { "TP3", "performer"},
323 const char ff_id3v2_tags[][4] = {
324 "TALB", "TBPM", "TCOM", "TCON", "TCOP", "TDLY", "TENC", "TEXT",
325 "TFLT", "TIT1", "TIT2", "TIT3", "TKEY", "TLAN", "TLEN", "TMED",
326 "TOAL", "TOFN", "TOLY", "TOPE", "TOWN", "TPE1", "TPE2", "TPE3",
327 "TPE4", "TPOS", "TPUB", "TRCK", "TRSN", "TRSO", "TSRC", "TSSE",
331 const char ff_id3v2_4_tags[][4] = {
332 "TDEN", "TDOR", "TDRC", "TDRL", "TDTG", "TIPL", "TMCL", "TMOO",
333 "TPRO", "TSOA", "TSOP", "TSOT", "TSST",
337 const char ff_id3v2_3_tags[][4] = {
338 "TDAT", "TIME", "TORY", "TRDA", "TSIZ", "TYER",