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., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
18 * @author Andre Moreira Magalhaes <andre.magalhaes@openbossa.org>
24 * asf/wma file parser.
31 #include <lightmediascanner_plugin.h>
32 #include <lightmediascanner_db.h>
33 #include <sys/types.h>
41 STREAM_TYPE_UNKNOWN = 0,
47 ATTR_TYPE_UNICODE = 0,
57 struct lms_string_size title;
58 struct lms_string_size artist;
59 struct lms_string_size album;
60 struct lms_string_size genre;
61 unsigned char trackno;
65 struct lms_plugin plugin;
66 lms_db_audio_t *audio_db;
67 lms_db_video_t *video_db;
68 lms_charset_conv_t *cs_conv;
71 static const char _name[] = "asf";
72 static const struct lms_string_size _exts[] = {
73 LMS_STATIC_STRING_SIZE(".wma"),
74 LMS_STATIC_STRING_SIZE(".wmv"),
75 LMS_STATIC_STRING_SIZE(".asf")
78 static const char header_guid[16] = "\x30\x26\xB2\x75\x8E\x66\xCF\x11\xA6\xD9\x00\xAA\x00\x62\xCE\x6C";
79 static const char file_properties_guid[16] = "\xA1\xDC\xAB\x8C\x47\xA9\xCF\x11\x8E\xE4\x00\xC0\x0C\x20\x53\x65";
80 static const char stream_properties_guid[16] = "\x91\x07\xDC\xB7\xB7\xA9\xCF\x11\x8E\xE6\x00\xC0\x0C\x20\x53\x65";
81 static const char stream_type_audio_guid[16] = "\x40\x9E\x69\xF8\x4D\x5B\xCF\x11\xA8\xFD\x00\x80\x5F\x5C\x44\x2B";
82 static const char stream_type_video_guid[16] = "\xC0\xEF\x19\xBC\x4D\x5B\xCF\x11\xA8\xFD\x00\x80\x5F\x5C\x44\x2B";
83 static const char content_description_guid[16] = "\x33\x26\xB2\x75\x8E\x66\xCF\x11\xA6\xD9\x00\xAA\x00\x62\xCE\x6C";
84 static const char extended_content_description_guid[16] = "\x40\xA4\xD0\xD2\x07\xE3\xD2\x11\x97\xF0\x00\xA0\xC9\x5E\xA8\x50";
85 static const char header_extension_guid[16] = "\xb5\x03\xbf_.\xa9\xcf\x11\x8e\xe3\x00\xc0\x0c Se";
86 static const char metadata_guid[16] = "\xEA\xCB\xF8\xC5\xAF[wH\204g\xAA\214D\xFAL\xCA";
87 static const char metadata_library_guid[16] = "\224\034#D\230\224\321I\241A\x1d\x13NEpT";
88 static const char content_encryption_object_guid[16] = "\xFB\xB3\x11\x22\x23\xBD\xD2\x11\xB4\xB7\x00\xA0\xC9\x55\xFC\x6E";
89 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";
92 _to_number(const char *data, unsigned int type_size, unsigned int data_size)
97 last = data_size > type_size ? type_size - 1 : data_size - 1;
99 for (i = 0; i <= last; i++)
100 sum |= (unsigned char) (data[i]) << (i * 8);
109 if (read(fd, &v, 2) != 2) {
112 return (unsigned int) _to_number(v, sizeof(unsigned short), 2);
119 if (read(fd, &v, 4) != 4) {
122 return (unsigned int) _to_number(v, sizeof(unsigned int), 4);
129 if (read(fd, &v, 8) != 8) {
132 return (unsigned int) _to_number(v, sizeof(unsigned long long), 8);
136 _read_string(int fd, size_t count, char **str, size_t *len)
139 size_t data_size, size;
141 data = (char *) malloc(sizeof(char) * count);
142 data_size = read(fd, data, count);
143 if (data_size == -1) {
150 if (data[size - 1] != '\0' || data[size - 2] != '\0') {
163 _parse_content_description(lms_charset_conv_t *cs_conv, int fd, struct asf_info *info)
165 int title_length = _read_word(fd);
166 int artist_length = _read_word(fd);
167 int copyright_length = _read_word(fd);
168 int comment_length = _read_word(fd);
169 int rating_length = _read_word(fd);
171 char *copyright, *comment, *rating;
173 _read_string(fd, title_length, &info->title.str, &info->title.len);
174 lms_charset_conv_force(cs_conv, &info->title.str, &info->title.len);
175 _read_string(fd, artist_length, &info->artist.str, &info->artist.len);
176 lms_charset_conv_force(cs_conv, &info->artist.str, &info->artist.len);
177 _read_string(fd, copyright_length, ©right, &len);
178 _read_string(fd, comment_length, &comment, &len);
179 _read_string(fd, rating_length, &rating, &len);
183 _parse_attribute_name(int fd,
186 size_t *attr_name_len,
190 int attr_name_length;
192 /* extended content descriptor */
194 attr_name_length = _read_word(fd);
196 _read_string(fd, attr_name_length, attr_name, attr_name_len);
198 lseek(fd, attr_name_length, SEEK_CUR);
199 *attr_type = _read_word(fd);
200 *attr_size = _read_word(fd);
202 /* metadata & metadata library */
204 lseek(fd, 2, SEEK_CUR); /* language */
205 lseek(fd, 2, SEEK_CUR); /* stream */
206 attr_name_length = _read_word(fd);
207 *attr_type = _read_word(fd);
208 *attr_size = _read_dword(fd);
210 _read_string(fd, attr_name_length, attr_name, attr_name_len);
212 lseek(fd, attr_name_length, SEEK_CUR);
217 _parse_attribute_string_data(lms_charset_conv_t *cs_conv,
221 size_t *attr_data_len)
223 _read_string(fd, attr_size, attr_data, attr_data_len);
224 lms_charset_conv_force(cs_conv, attr_data, attr_data_len);
228 _skip_attribute_data(int fd, int kind, int attr_type, int attr_size)
232 lseek(fd, 2, SEEK_CUR);
237 lseek(fd, 4, SEEK_CUR);
240 lseek(fd, 2, SEEK_CUR);
244 case ATTR_TYPE_DWORD:
245 lseek(fd, 4, SEEK_CUR);
248 case ATTR_TYPE_QWORD:
249 lseek(fd, 8, SEEK_CUR);
252 case ATTR_TYPE_UNICODE:
253 case ATTR_TYPE_BYTES:
255 lseek(fd, attr_size, SEEK_CUR);
264 _skip_attribute(int fd, int kind)
266 int attr_type, attr_size;
267 _parse_attribute_name(fd, kind, NULL, NULL, &attr_type, &attr_size);
268 _skip_attribute_data(fd, kind, attr_type, attr_size);
272 _parse_extended_content_description_object(lms_charset_conv_t *cs_conv, int fd, struct asf_info *info)
274 int count = _read_word(fd);
276 size_t attr_name_len;
277 int attr_type, attr_size;
280 _parse_attribute_name(fd, 0,
281 &attr_name, &attr_name_len,
282 &attr_type, &attr_size);
283 if (attr_type == ATTR_TYPE_UNICODE) {
284 lms_charset_conv_force(cs_conv, &attr_name, &attr_name_len);
285 if (strcmp(attr_name, "WM/AlbumTitle") == 0)
286 _parse_attribute_string_data(cs_conv,
290 else if (strcmp(attr_name, "WM/Genre") == 0)
291 _parse_attribute_string_data(cs_conv,
295 else if (strcmp(attr_name, "WM/TrackNumber") == 0) {
298 _parse_attribute_string_data(cs_conv,
303 info->trackno = atoi(trackno);
308 _skip_attribute_data(fd, 0, attr_type, attr_size);
311 _skip_attribute_data(fd, 0, attr_type, attr_size);
318 _skip_metadata(int fd)
320 int count = _read_word(fd);
322 _skip_attribute(fd, 1);
327 _skip_metadata_library(int fd)
329 int count = _read_word(fd);
331 _skip_attribute(fd, 2);
336 _skip_header_extension(int fd)
339 long long size, data_size, data_pos;
341 lseek(fd, 18, SEEK_CUR);
342 data_size = _read_dword(fd);
344 while (data_pos < data_size) {
346 size = _read_qword(fd);
347 if (memcmp(guid, metadata_guid, 16) == 0) {
350 else if (memcmp(guid, metadata_library_guid, 16) == 0) {
351 _skip_metadata_library(fd);
354 lseek(fd, size - 24, SEEK_CUR);
361 _strstrip(char **str, unsigned int *p_len)
364 lms_strstrip(*str, p_len);
366 if (*p_len == 0 && *str) {
373 _match(struct plugin *p, const char *path, int len, int base)
377 i = lms_which_extension(path, len, _exts, LMS_ARRAY_SIZE(_exts));
381 return (void*)(i + 1);
385 _parse(struct plugin *plugin, struct lms_context *ctxt, const struct lms_file_info *finfo, void *match)
387 struct asf_info info = {0};
388 struct lms_audio_info audio_info = {0};
389 struct lms_video_info video_info = {0};
390 int r, fd, num_objects, i;
393 int stream_type = STREAM_TYPE_UNKNOWN;
395 fd = open(finfo->path, O_RDONLY);
401 if (read(fd, &guid, 16) != 16) {
406 if (memcmp(guid, header_guid, 16) != 0) {
407 fprintf(stderr, "ERROR: invalid header (%s).\n", finfo->path);
412 size = _read_qword(fd);
413 num_objects = _read_dword(fd);
415 lseek(fd, 2, SEEK_CUR);
417 for (i = 0; i < num_objects; ++i) {
419 size = _read_qword(fd);
421 if (memcmp(guid, file_properties_guid, 16) == 0)
422 lseek(fd, size - 24, SEEK_CUR);
423 else if (memcmp(guid, stream_properties_guid, 16) == 0) {
425 if (memcmp(guid, stream_type_audio_guid, 16) == 0)
426 stream_type = STREAM_TYPE_AUDIO;
427 else if (memcmp(guid, stream_type_video_guid, 16) == 0)
428 stream_type = STREAM_TYPE_VIDEO;
429 lseek(fd, size - 40, SEEK_CUR);
431 else if (memcmp(guid, content_description_guid, 16) == 0)
432 _parse_content_description(plugin->cs_conv, fd, &info);
433 else if (memcmp(guid, extended_content_description_guid, 16) == 0)
434 _parse_extended_content_description_object(plugin->cs_conv, fd, &info);
435 else if (memcmp(guid, header_extension_guid, 16) == 0)
436 _skip_header_extension(fd);
437 else if (memcmp(guid, content_encryption_object_guid, 16) == 0 ||
438 memcmp(guid, extended_content_encryption_object_guid, 16) == 0) {
439 /* ignore DRM'd files */
440 fprintf(stderr, "ERROR: ignoring DRM'd file %s\n", finfo->path);
445 lseek(fd, size - 24, SEEK_CUR);
448 /* try to define stream type by extension */
449 if (stream_type == STREAM_TYPE_UNKNOWN) {
450 int ext_idx = ((int)match) - 1;
451 if (strcmp(_exts[ext_idx].str, ".wma") == 0)
452 stream_type = STREAM_TYPE_AUDIO;
453 /* consider wmv and asf as video */
455 stream_type = STREAM_TYPE_VIDEO;
458 _strstrip(&info.title.str, &info.title.len);
459 _strstrip(&info.artist.str, &info.genre.len);
460 _strstrip(&info.album.str, &info.album.len);
461 _strstrip(&info.genre.str, &info.genre.len);
463 if (!info.title.str) {
466 free(info.title.str);
467 ext_idx = ((int)match) - 1;
468 info.title.len = finfo->path_len - finfo->base - _exts[ext_idx].len;
469 info.title.str = malloc((info.title.len + 1) * sizeof(char));
470 memcpy(info.title.str, finfo->path + finfo->base, info.title.len);
471 info.title.str[info.title.len] = '\0';
472 lms_charset_conv(ctxt->cs_conv, &info.title.str, &info.title.len);
476 fprintf(stderr, "file %s info\n", finfo->path);
477 fprintf(stderr, "\ttitle='%s' len=%d\n", info.title.str, info.title.len);
478 fprintf(stderr, "\tartist='%s'\n", info.artist.str);
479 fprintf(stderr, "\talbum='%s'\n", info.album.str);
480 fprintf(stderr, "\tgenre='%s'\n", info.genre.str);
481 fprintf(stderr, "\ttrackno=%d\n", info.trackno);
484 if (stream_type == STREAM_TYPE_AUDIO) {
485 audio_info.id = finfo->id;
486 audio_info.title = info.title;
487 audio_info.artist = info.artist;
488 audio_info.album = info.album;
489 audio_info.genre = info.genre;
490 audio_info.trackno = info.trackno;
491 r = lms_db_audio_add(plugin->audio_db, &audio_info);
494 video_info.id = finfo->id;
495 video_info.title = info.title;
496 video_info.artist = info.artist;
497 r = lms_db_video_add(plugin->video_db, &video_info);
502 free(info.title.str);
504 free(info.artist.str);
506 free(info.album.str);
508 free(info.genre.str);
510 posix_fadvise(fd, 0, 0, POSIX_FADV_DONTNEED);
517 _setup(struct plugin *plugin, struct lms_context *ctxt)
519 plugin->audio_db = lms_db_audio_new(ctxt->db);
520 if (!plugin->audio_db)
522 plugin->video_db = lms_db_video_new(ctxt->db);
523 if (!plugin->video_db)
525 plugin->cs_conv = lms_charset_conv_new();
526 if (!plugin->cs_conv)
528 lms_charset_conv_add(plugin->cs_conv, "UTF-16LE");
534 _start(struct plugin *plugin, struct lms_context *ctxt)
537 r = lms_db_audio_start(plugin->audio_db);
538 r |= lms_db_video_start(plugin->video_db);
543 _finish(struct plugin *plugin, struct lms_context *ctxt)
545 if (plugin->audio_db)
546 lms_db_audio_free(plugin->audio_db);
547 if (plugin->video_db)
548 lms_db_video_free(plugin->video_db);
550 lms_charset_conv_free(plugin->cs_conv);
556 _close(struct plugin *plugin)
562 API struct lms_plugin *
563 lms_plugin_open(void)
565 struct plugin *plugin;
567 plugin = (struct plugin *)malloc(sizeof(*plugin));
568 plugin->plugin.name = _name;
569 plugin->plugin.match = (lms_plugin_match_fn_t)_match;
570 plugin->plugin.parse = (lms_plugin_parse_fn_t)_parse;
571 plugin->plugin.close = (lms_plugin_close_fn_t)_close;
572 plugin->plugin.setup = (lms_plugin_setup_fn_t)_setup;
573 plugin->plugin.start = (lms_plugin_start_fn_t)_start;
574 plugin->plugin.finish = (lms_plugin_finish_fn_t)_finish;
576 return (struct lms_plugin *)plugin;