2 * Copyright (C) 2008 by INdT
4 * This program is free software; you can redistribute it and/or
5 * modify it under the terms of the GNU Lesser General Public License
6 * as published by the Free Software Foundation; either version 2
7 * of the License, or (at your option) any later version.
9 * This program is distributed in the hope that it will be useful,
10 * but WITHOUT ANY WARRANTY; without even the implied warranty of
11 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
12 * GNU General Public License for more details.
14 * You should have received a copy of the GNU Lesser General Public License
15 * along with this program; if not, write to the Free Software
16 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
18 * @author Andre Moreira Magalhaes <andre.magalhaes@openbossa.org>
24 * asf/wma file parser.
31 #define _XOPEN_SOURCE 600
33 #include <lightmediascanner_plugin.h>
34 #include <lightmediascanner_db.h>
36 #include <sys/types.h>
44 #define NSEC100_PER_SEC 10000000ULL
45 #define MSEC_PER_SEC 1000ULL
48 STREAM_TYPE_UNKNOWN = 0,
54 ATTR_TYPE_UNICODE = 0,
64 struct lms_string_size title;
65 struct lms_string_size artist;
66 struct lms_string_size album;
67 struct lms_string_size genre;
68 unsigned char trackno;
72 struct lms_plugin plugin;
73 lms_db_audio_t *audio_db;
74 lms_db_video_t *video_db;
75 lms_charset_conv_t *cs_conv;
78 static const char _name[] = "asf";
79 static const struct lms_string_size _container = LMS_STATIC_STRING_SIZE("asf");
80 static const struct lms_string_size _exts[] = {
81 LMS_STATIC_STRING_SIZE(".wma"),
82 LMS_STATIC_STRING_SIZE(".wmv"),
83 LMS_STATIC_STRING_SIZE(".asf")
85 static const char *_cats[] = {
90 static const char *_authors[] = {
91 "Andre Moreira Magalhaes",
97 * Microsoft defines these 16-byte (128-bit) GUIDs as:
98 * first 8 bytes are in little-endian order
99 * next 8 bytes are in big-endian order
101 * Eg.: AaBbCcDd-EeFf-GgHh-IiJj-KkLlMmNnOoPp:
103 * to convert to byte string do as follow:
105 * $Dd $Cc $Bb $Aa $Ff $Ee $Hh $Gg $Ii $Jj $Kk $Ll $Mm $Nn $Oo $Pp
107 * See http://www.microsoft.com/windows/windowsmedia/forpros/format/asfspec.aspx
109 static const char header_guid[16] = "\x30\x26\xB2\x75\x8E\x66\xCF\x11\xA6\xD9\x00\xAA\x00\x62\xCE\x6C";
110 static const char file_properties_guid[16] = "\xA1\xDC\xAB\x8C\x47\xA9\xCF\x11\x8E\xE4\x00\xC0\x0C\x20\x53\x65";
111 static const char stream_properties_guid[16] = "\x91\x07\xDC\xB7\xB7\xA9\xCF\x11\x8E\xE6\x00\xC0\x0C\x20\x53\x65";
112 static const char stream_type_audio_guid[16] = "\x40\x9E\x69\xF8\x4D\x5B\xCF\x11\xA8\xFD\x00\x80\x5F\x5C\x44\x2B";
113 static const char stream_type_video_guid[16] = "\xC0\xEF\x19\xBC\x4D\x5B\xCF\x11\xA8\xFD\x00\x80\x5F\x5C\x44\x2B";
114 static const char content_description_guid[16] = "\x33\x26\xB2\x75\x8E\x66\xCF\x11\xA6\xD9\x00\xAA\x00\x62\xCE\x6C";
115 static const char extended_content_description_guid[16] = "\x40\xA4\xD0\xD2\x07\xE3\xD2\x11\x97\xF0\x00\xA0\xC9\x5E\xA8\x50";
116 static const char header_extension_guid[16] = "\xb5\x03\xbf_.\xa9\xcf\x11\x8e\xe3\x00\xc0\x0c Se";
117 static const char metadata_guid[16] = "\xEA\xCB\xF8\xC5\xAF[wH\204g\xAA\214D\xFAL\xCA";
118 static const char metadata_library_guid[16] = "\224\034#D\230\224\321I\241A\x1d\x13NEpT";
119 static const char content_encryption_object_guid[16] = "\xFB\xB3\x11\x22\x23\xBD\xD2\x11\xB4\xB7\x00\xA0\xC9\x55\xFC\x6E";
120 static const char extended_content_encryption_object_guid[16] = "\x14\xE6\x8A\x29\x22\x26\x17\x4C\xB9\x35\xDA\xE0\x7E\xE9\x28\x9C";
122 static const char attr_name_wm_album_artist[28] = "\x57\x00\x4d\x00\x2f\x00\x41\x00\x6c\x00\x62\x00\x75\x00\x6d\x00\x41\x00\x72\x00\x74\x00\x69\x00\x73\x00\x74\x00";
123 static const char attr_name_wm_album_title[26] = "\x57\x00\x4d\x00\x2f\x00\x41\x00\x6c\x00\x62\x00\x75\x00\x6d\x00\x54\x00\x69\x00\x74\x00\x6c\x00\x65\x00";
124 static const char attr_name_wm_genre[16] = "\x57\x00\x4d\x00\x2f\x00\x47\x00\x65\x00\x6e\x00\x72\x00\x65\x00";
125 static const char attr_name_wm_track_number[28] = "\x57\x00\x4d\x00\x2f\x00\x54\x00\x72\x00\x61\x00\x63\x00\x6b\x00\x4e\x00\x75\x00\x6d\x00\x62\x00\x65\x00\x72\x00";
128 _to_number(const char *data, unsigned int type_size, unsigned int data_size)
131 unsigned int last, i;
133 last = data_size > type_size ? type_size : data_size;
135 for (i = 0; i < last; i++)
136 sum |= (unsigned char) (data[i]) << (i * 8);
145 if (read(fd, &v, 2) != 2)
147 return (short) _to_number(v, sizeof(unsigned short), 2);
154 if (read(fd, &v, 4) != 4)
156 return (unsigned int) _to_number(v, sizeof(unsigned int), 4);
163 if (read(fd, &v, 8) != 8)
165 return _to_number(v, sizeof(unsigned long long), 8);
169 _read_string(int fd, size_t count, char **str, unsigned int *len)
172 ssize_t data_size, size;
174 data = malloc(sizeof(char) * count);
175 data_size = read(fd, data, count);
176 if (data_size == -1) {
183 if (data[size - 1] != '\0' || data[size - 2] != '\0')
195 _parse_file_properties(lms_charset_conv_t *cs_conv, int fd,
196 struct lms_audio_info *info)
201 uint64_t creation_date;
202 uint64_t data_packets_count;
203 uint64_t play_duration;
204 uint64_t send_duration;
207 uint32_t min_data_packet_size;
208 uint32_t max_data_packet_size;
209 uint32_t max_bitrate;
210 } __attribute__((packed)) props;
213 r = read(fd, &props, sizeof(props));
214 if (r != sizeof(props))
218 if (le32toh(props.flags) & 0x1)
221 /* ASF spec 01.20.06 sec. 3.2: we need to subtract the preroll value from
222 * the duration in order to obtain the real duration */
223 info->length = (unsigned int)((le64toh(props.play_duration) / NSEC100_PER_SEC)
224 - le64toh(props.preroll) / MSEC_PER_SEC);
230 _parse_content_description(lms_charset_conv_t *cs_conv, int fd, struct asf_info *info)
232 int title_length = _read_word(fd);
233 int artist_length = _read_word(fd);
234 int copyright_length = _read_word(fd);
235 int comment_length = _read_word(fd);
236 int rating_length = _read_word(fd);
238 _read_string(fd, title_length, &info->title.str, &info->title.len);
239 lms_charset_conv_force(cs_conv, &info->title.str, &info->title.len);
240 _read_string(fd, artist_length, &info->artist.str, &info->artist.len);
241 lms_charset_conv_force(cs_conv, &info->artist.str, &info->artist.len);
242 /* ignore copyright, comment and rating */
243 lseek(fd, copyright_length + comment_length + rating_length, SEEK_CUR);
247 _parse_attribute_name(int fd,
249 unsigned int *attr_name_len,
253 int attr_name_length;
255 attr_name_length = _read_word(fd);
256 _read_string(fd, attr_name_length, attr_name, attr_name_len);
257 *attr_type = _read_word(fd);
258 *attr_size = _read_word(fd);
262 _parse_attribute_string_data(lms_charset_conv_t *cs_conv,
266 unsigned int *attr_data_len)
268 _read_string(fd, attr_size, attr_data, attr_data_len);
269 lms_charset_conv_force(cs_conv, attr_data, attr_data_len);
273 _skip_attribute_data(int fd, int kind, int attr_type, int attr_size)
277 lseek(fd, 2, SEEK_CUR);
282 lseek(fd, 4, SEEK_CUR);
284 lseek(fd, 2, SEEK_CUR);
287 case ATTR_TYPE_DWORD:
288 lseek(fd, 4, SEEK_CUR);
291 case ATTR_TYPE_QWORD:
292 lseek(fd, 8, SEEK_CUR);
295 case ATTR_TYPE_UNICODE:
296 case ATTR_TYPE_BYTES:
298 lseek(fd, attr_size, SEEK_CUR);
307 _parse_extended_content_description_object(lms_charset_conv_t *cs_conv, int fd, struct asf_info *info)
309 int count = _read_word(fd);
311 unsigned int attr_name_len;
312 int attr_type, attr_size;
315 _parse_attribute_name(fd,
316 &attr_name, &attr_name_len,
317 &attr_type, &attr_size);
318 if (attr_type == ATTR_TYPE_UNICODE) {
319 if (memcmp(attr_name, attr_name_wm_album_title, attr_name_len) == 0)
320 _parse_attribute_string_data(cs_conv,
324 else if (memcmp(attr_name, attr_name_wm_genre, attr_name_len) == 0)
325 _parse_attribute_string_data(cs_conv,
329 else if (memcmp(attr_name, attr_name_wm_album_artist, attr_name_len) == 0)
330 _parse_attribute_string_data(cs_conv,
334 else if (memcmp(attr_name, attr_name_wm_track_number, attr_name_len) == 0) {
336 unsigned int trackno_len;
337 _parse_attribute_string_data(cs_conv,
342 info->trackno = atoi(trackno);
347 _skip_attribute_data(fd, 0, attr_type, attr_size);
350 _skip_attribute_data(fd, 0, attr_type, attr_size);
357 _match(struct plugin *p, const char *path, int len, int base)
361 i = lms_which_extension(path, len, _exts, LMS_ARRAY_SIZE(_exts));
365 return (void*)(i + 1);
369 _parse(struct plugin *plugin, struct lms_context *ctxt, const struct lms_file_info *finfo, void *match)
371 struct asf_info info = { };
372 struct lms_audio_info audio_info = { };
373 struct lms_video_info video_info = { };
374 int r, fd, num_objects, i;
377 int stream_type = STREAM_TYPE_UNKNOWN;
379 fd = open(finfo->path, O_RDONLY);
385 if (read(fd, &guid, 16) != 16) {
390 if (memcmp(guid, header_guid, 16) != 0) {
391 fprintf(stderr, "ERROR: invalid header (%s).\n", finfo->path);
396 size = _read_qword(fd);
397 num_objects = _read_dword(fd);
399 lseek(fd, 2, SEEK_CUR);
401 for (i = 0; i < num_objects; ++i) {
403 size = _read_qword(fd);
405 if (memcmp(guid, file_properties_guid, 16) == 0) {
406 r = _parse_file_properties(plugin->cs_conv, fd, &audio_info);
409 lseek(fd, size - (24 + r), SEEK_CUR);
410 } else if (memcmp(guid, stream_properties_guid, 16) == 0) {
412 if (memcmp(guid, stream_type_audio_guid, 16) == 0)
413 stream_type = STREAM_TYPE_AUDIO;
414 else if (memcmp(guid, stream_type_video_guid, 16) == 0)
415 stream_type = STREAM_TYPE_VIDEO;
416 lseek(fd, size - 40, SEEK_CUR);
418 else if (memcmp(guid, content_description_guid, 16) == 0)
419 _parse_content_description(plugin->cs_conv, fd, &info);
420 else if (memcmp(guid, extended_content_description_guid, 16) == 0)
421 _parse_extended_content_description_object(plugin->cs_conv, fd, &info);
422 else if (memcmp(guid, content_encryption_object_guid, 16) == 0 ||
423 memcmp(guid, extended_content_encryption_object_guid, 16) == 0) {
424 /* ignore DRM'd files */
425 fprintf(stderr, "ERROR: ignoring DRM'd file %s\n", finfo->path);
429 lseek(fd, size - 24, SEEK_CUR);
432 /* try to define stream type by extension */
433 if (stream_type == STREAM_TYPE_UNKNOWN) {
434 long ext_idx = ((long)match) - 1;
435 if (strcmp(_exts[ext_idx].str, ".wma") == 0)
436 stream_type = STREAM_TYPE_AUDIO;
437 /* consider wmv and asf as video */
439 stream_type = STREAM_TYPE_VIDEO;
442 lms_string_size_strip_and_free(&info.title);
443 lms_string_size_strip_and_free(&info.artist);
444 lms_string_size_strip_and_free(&info.album);
445 lms_string_size_strip_and_free(&info.genre);
447 if (!info.title.str) {
449 ext_idx = ((long)match) - 1;
450 info.title.len = finfo->path_len - finfo->base - _exts[ext_idx].len;
451 info.title.str = malloc((info.title.len + 1) * sizeof(char));
452 memcpy(info.title.str, finfo->path + finfo->base, info.title.len);
453 info.title.str[info.title.len] = '\0';
454 lms_charset_conv(ctxt->cs_conv, &info.title.str, &info.title.len);
458 fprintf(stderr, "file %s info\n", finfo->path);
459 fprintf(stderr, "\ttitle='%s'\n", info.title.str);
460 fprintf(stderr, "\tartist='%s'\n", info.artist.str);
461 fprintf(stderr, "\talbum='%s'\n", info.album.str);
462 fprintf(stderr, "\tgenre='%s'\n", info.genre.str);
463 fprintf(stderr, "\ttrackno=%d\n", info.trackno);
466 audio_info.container = _container;
468 if (stream_type == STREAM_TYPE_AUDIO) {
469 audio_info.id = finfo->id;
470 audio_info.title = info.title;
471 audio_info.artist = info.artist;
472 audio_info.album = info.album;
473 audio_info.genre = info.genre;
474 audio_info.trackno = info.trackno;
475 r = lms_db_audio_add(plugin->audio_db, &audio_info);
477 video_info.id = finfo->id;
478 video_info.title = info.title;
479 video_info.artist = info.artist;
480 r = lms_db_video_add(plugin->video_db, &video_info);
484 free(info.title.str);
485 free(info.artist.str);
486 free(info.album.str);
487 free(info.genre.str);
489 posix_fadvise(fd, 0, 0, POSIX_FADV_DONTNEED);
496 _setup(struct plugin *plugin, struct lms_context *ctxt)
498 plugin->audio_db = lms_db_audio_new(ctxt->db);
499 if (!plugin->audio_db)
501 plugin->video_db = lms_db_video_new(ctxt->db);
502 if (!plugin->video_db)
504 plugin->cs_conv = lms_charset_conv_new();
505 if (!plugin->cs_conv)
507 lms_charset_conv_add(plugin->cs_conv, "UTF-16LE");
513 _start(struct plugin *plugin, struct lms_context *ctxt)
516 r = lms_db_audio_start(plugin->audio_db);
517 r |= lms_db_video_start(plugin->video_db);
522 _finish(struct plugin *plugin, struct lms_context *ctxt)
524 if (plugin->audio_db)
525 lms_db_audio_free(plugin->audio_db);
526 if (plugin->video_db)
527 lms_db_video_free(plugin->video_db);
529 lms_charset_conv_free(plugin->cs_conv);
535 _close(struct plugin *plugin)
541 API struct lms_plugin *
542 lms_plugin_open(void)
544 struct plugin *plugin;
546 plugin = (struct plugin *)malloc(sizeof(*plugin));
547 plugin->plugin.name = _name;
548 plugin->plugin.match = (lms_plugin_match_fn_t)_match;
549 plugin->plugin.parse = (lms_plugin_parse_fn_t)_parse;
550 plugin->plugin.close = (lms_plugin_close_fn_t)_close;
551 plugin->plugin.setup = (lms_plugin_setup_fn_t)_setup;
552 plugin->plugin.start = (lms_plugin_start_fn_t)_start;
553 plugin->plugin.finish = (lms_plugin_finish_fn_t)_finish;
555 return (struct lms_plugin *)plugin;
558 API const struct lms_plugin_info *
559 lms_plugin_info(void)
561 static struct lms_plugin_info info = {
564 "Microsoft WMA, WMV and ASF",
567 "http://lms.garage.maemo.org"