|
1
|
|
|
<?php declare(strict_types=1); |
|
2
|
|
|
|
|
3
|
|
|
/** |
|
4
|
|
|
* ownCloud - Music app |
|
5
|
|
|
* |
|
6
|
|
|
* This file is licensed under the Affero General Public License version 3 or |
|
7
|
|
|
* later. See the COPYING file. |
|
8
|
|
|
* |
|
9
|
|
|
* @author Pauli Järvinen <[email protected]> |
|
10
|
|
|
* @copyright Pauli Järvinen 2018 - 2025 |
|
11
|
|
|
*/ |
|
12
|
|
|
|
|
13
|
|
|
namespace OCA\Music\Utility; |
|
14
|
|
|
|
|
15
|
|
|
use OCP\Files\File; |
|
16
|
|
|
use OCP\Files\Folder; |
|
17
|
|
|
|
|
18
|
|
|
use OCA\Music\AppFramework\Core\Logger; |
|
19
|
|
|
|
|
20
|
|
|
class DetailsHelper { |
|
21
|
|
|
private Extractor $extractor; |
|
22
|
|
|
private Logger $logger; |
|
23
|
|
|
|
|
24
|
|
|
public function __construct( |
|
25
|
|
|
Extractor $extractor, |
|
26
|
|
|
Logger $logger) { |
|
27
|
|
|
$this->extractor = $extractor; |
|
28
|
|
|
$this->logger = $logger; |
|
29
|
|
|
} |
|
30
|
|
|
|
|
31
|
|
|
public function getDetails(int $fileId, Folder $userFolder) : ?array { |
|
32
|
|
|
$file = $userFolder->getById($fileId)[0] ?? null; |
|
33
|
|
|
if ($file instanceof File) { |
|
34
|
|
|
$data = $this->extractor->extract($file); |
|
35
|
|
|
$audio = $data['audio'] ?? []; |
|
36
|
|
|
$comments = $data['comments'] ?? []; |
|
37
|
|
|
|
|
38
|
|
|
// remove intermediate arrays |
|
39
|
|
|
$comments = self::flattenComments($comments); |
|
40
|
|
|
|
|
41
|
|
|
// cleanup strings from invalid characters |
|
42
|
|
|
\array_walk($audio, [$this, 'sanitizeString']); |
|
43
|
|
|
\array_walk($comments, [$this, 'sanitizeString']); |
|
44
|
|
|
|
|
45
|
|
|
$result = [ |
|
46
|
|
|
'fileinfo' => $audio, |
|
47
|
|
|
'tags' => $comments |
|
48
|
|
|
]; |
|
49
|
|
|
|
|
50
|
|
|
// binary data has to be encoded |
|
51
|
|
|
if (\array_key_exists('picture', $result['tags'])) { |
|
52
|
|
|
$result['tags']['picture'] = self::encodePictureTag($result['tags']['picture']); |
|
53
|
|
|
} |
|
54
|
|
|
|
|
55
|
|
|
// 'streams' contains duplicate data |
|
56
|
|
|
unset($result['fileinfo']['streams']); |
|
57
|
|
|
|
|
58
|
|
|
// one track number is enough |
|
59
|
|
|
if (\array_key_exists('track', $result['tags']) |
|
60
|
|
|
&& \array_key_exists('track_number', $result['tags'])) { |
|
61
|
|
|
unset($result['tags']['track']); |
|
62
|
|
|
} |
|
63
|
|
|
|
|
64
|
|
|
// special handling for lyrics tags |
|
65
|
|
|
$lyricsNode = self::transformLyrics($result['tags'], self::getLyricsFileContent($file)); |
|
66
|
|
|
if ($lyricsNode !== null) { |
|
67
|
|
|
$result['lyrics'] = $lyricsNode; |
|
68
|
|
|
unset( |
|
69
|
|
|
$result['tags']['LYRICS'], |
|
70
|
|
|
$result['tags']['lyrics'], |
|
71
|
|
|
$result['tags']['unsynchronised_lyric'], |
|
72
|
|
|
$result['tags']['unsynced lyrics'], |
|
73
|
|
|
$result['tags']['unsynced_lyrics'], |
|
74
|
|
|
$result['tags']['unsyncedlyrics'] |
|
75
|
|
|
); |
|
76
|
|
|
} |
|
77
|
|
|
|
|
78
|
|
|
// add track length |
|
79
|
|
|
$result['length'] = $data['playtime_seconds'] ?? null; |
|
80
|
|
|
|
|
81
|
|
|
// add file path |
|
82
|
|
|
$result['path'] = $userFolder->getRelativePath($file->getPath()); |
|
83
|
|
|
|
|
84
|
|
|
return $result; |
|
85
|
|
|
} |
|
86
|
|
|
return null; |
|
87
|
|
|
} |
|
88
|
|
|
|
|
89
|
|
|
/** |
|
90
|
|
|
* Check if a file has embedded lyrics without parsing them |
|
91
|
|
|
*/ |
|
92
|
|
|
public function hasLyrics(int $fileId, Folder $userFolder) : bool { |
|
93
|
|
|
$fileNode = $userFolder->getById($fileId)[0] ?? null; |
|
94
|
|
|
if ($fileNode instanceof File) { |
|
95
|
|
|
$data = $this->extractor->extract($fileNode); |
|
96
|
|
|
$lyrics = ExtractorGetID3::getFirstOfTags($data, ['unsynchronised_lyric', 'unsynced lyrics', 'unsynced_lyrics', 'unsyncedlyrics']) |
|
97
|
|
|
?? self::getLyricsFileContent($fileNode); |
|
98
|
|
|
return ($lyrics !== null); |
|
99
|
|
|
} |
|
100
|
|
|
return false; |
|
101
|
|
|
} |
|
102
|
|
|
|
|
103
|
|
|
/** |
|
104
|
|
|
* Get lyrics from the file metadata in plain-text format. If there's no unsynchronised lyrics available |
|
105
|
|
|
* but there is synchronised lyrics, then the plain-text format is converted from the synchronised lyrics. |
|
106
|
|
|
*/ |
|
107
|
|
|
public function getLyricsAsPlainText(int $fileId, Folder $userFolder) : ?string { |
|
108
|
|
|
$lyrics = null; |
|
109
|
|
|
$fileNode = $userFolder->getById($fileId)[0] ?? null; |
|
110
|
|
|
if ($fileNode instanceof File) { |
|
111
|
|
|
$data = $this->extractor->extract($fileNode); |
|
112
|
|
|
$lyrics = ExtractorGetID3::getFirstOfTags($data, ['unsynchronised_lyric', 'unsynced lyrics', 'unsynced_lyrics', 'unsyncedlyrics']); |
|
113
|
|
|
self::sanitizeString($lyrics); |
|
114
|
|
|
|
|
115
|
|
|
if ($lyrics === null) { |
|
116
|
|
|
// no unsynchronized lyrics, try to get and convert the potentially synchronized lyrics |
|
117
|
|
|
$lyrics = ExtractorGetID3::getFirstOfTags($data, ['LYRICS', 'lyrics']) ?? self::getLyricsFileContent($fileNode); |
|
118
|
|
|
self::sanitizeString($lyrics); |
|
119
|
|
|
$parsed = LyricsParser::parseSyncedLyrics($lyrics); |
|
120
|
|
|
if ($parsed) { |
|
121
|
|
|
// the lyrics were indeed time-synced, convert the parsed array to a plain string |
|
122
|
|
|
$lyrics = LyricsParser::syncedToUnsynced($parsed); |
|
123
|
|
|
} |
|
124
|
|
|
} |
|
125
|
|
|
} |
|
126
|
|
|
return $lyrics; |
|
127
|
|
|
} |
|
128
|
|
|
|
|
129
|
|
|
/** |
|
130
|
|
|
* Get all lyrics from the file metadata, both synced and unsynced. For both lyrics types, a single instance |
|
131
|
|
|
* of lyrics contains an array of lyrics lines. In case of synced lyrics, the key of this array is the offset |
|
132
|
|
|
* in milliseconds. |
|
133
|
|
|
* @return array of items like ['synced' => boolean, 'lines' => array] |
|
134
|
|
|
*/ |
|
135
|
|
|
public function getLyricsAsStructured(int $fileId, Folder $userFolder) : array { |
|
136
|
|
|
$result = []; |
|
137
|
|
|
$fileNode = $userFolder->getById($fileId)[0] ?? null; |
|
138
|
|
|
if ($fileNode instanceof File) { |
|
139
|
|
|
$data = $this->extractor->extract($fileNode); |
|
140
|
|
|
$lyricsTags = ExtractorGetID3::getTags($data, ['LYRICS', 'lyrics', 'unsynchronised_lyric', 'unsynced lyrics', 'unsynced_lyrics', 'unsyncedlyrics']); |
|
141
|
|
|
|
|
142
|
|
|
$lrcFileContent = self::getLyricsFileContent($fileNode); |
|
143
|
|
|
if ($lrcFileContent !== null) { |
|
144
|
|
|
// prepend the LRC file content to make it the preferred option for the clients |
|
145
|
|
|
$lyricsTags = \array_merge(['lyrics_file' => $lrcFileContent], $lyricsTags); |
|
146
|
|
|
} |
|
147
|
|
|
|
|
148
|
|
|
foreach ($lyricsTags as $tagKey => $tagValue) { |
|
149
|
|
|
self::sanitizeString($tagValue); |
|
150
|
|
|
|
|
151
|
|
|
// Never try to parse synced lyrics from the "unsync*" tags. The "lyrics" tag, on the other hand, |
|
152
|
|
|
// may contain either synced or unsynced lyrics and a parse attempt is needed to find out. |
|
153
|
|
|
$mayBeSynced = !Util::startsWith($tagKey, 'unsync'); |
|
154
|
|
|
$syncedLyrics = $mayBeSynced ? LyricsParser::parseSyncedLyrics($tagValue) : null; |
|
155
|
|
|
|
|
156
|
|
|
if ($syncedLyrics) { |
|
157
|
|
|
$result[] = [ |
|
158
|
|
|
'synced' => true, |
|
159
|
|
|
'lines' => $syncedLyrics |
|
160
|
|
|
]; |
|
161
|
|
|
} else { |
|
162
|
|
|
$result[] = [ |
|
163
|
|
|
'synced' => false, |
|
164
|
|
|
'lines' => \explode("\n", $tagValue) |
|
165
|
|
|
]; |
|
166
|
|
|
} |
|
167
|
|
|
} |
|
168
|
|
|
} |
|
169
|
|
|
return $result; |
|
170
|
|
|
} |
|
171
|
|
|
|
|
172
|
|
|
/** |
|
173
|
|
|
* Read lyrics-related tags, and build a result array containing potentially |
|
174
|
|
|
* both time-synced and unsynced lyrics. If no lyrics tags are found, the result will |
|
175
|
|
|
* be null. In case the result is non-null, there is always at least the key 'unsynced' |
|
176
|
|
|
* in the result which will hold a string representing the lyrics with no timestamps. |
|
177
|
|
|
* If found and successfully parsed, there will be also another key 'synced', which will |
|
178
|
|
|
* hold the time-synced lyrics. These are presented as an array of arrays of form |
|
179
|
|
|
* ['time' => int (ms), 'text' => string]. |
|
180
|
|
|
*/ |
|
181
|
|
|
private static function transformLyrics(array $tags, ?string $lrcFileContent) : ?array { |
|
182
|
|
|
$lyrics = $tags['LYRICS'] ?? $tags['lyrics'] ?? null; // may be synced or unsynced |
|
183
|
|
|
$syncedLyrics = LyricsParser::parseSyncedLyrics($lrcFileContent) ?? LyricsParser::parseSyncedLyrics($lyrics); |
|
184
|
|
|
$unsyncedLyrics = $tags['unsynchronised_lyric'] |
|
185
|
|
|
?? $tags['unsynced lyrics'] |
|
186
|
|
|
?? $tags['unsynced_lyrics'] |
|
187
|
|
|
?? $tags['unsyncedlyrics'] |
|
188
|
|
|
?? LyricsParser::syncedToUnsynced($syncedLyrics) |
|
189
|
|
|
?? $lrcFileContent |
|
190
|
|
|
?? $lyrics; |
|
191
|
|
|
|
|
192
|
|
|
if ($unsyncedLyrics !== null) { |
|
193
|
|
|
$result = ['unsynced' => $unsyncedLyrics]; |
|
194
|
|
|
|
|
195
|
|
|
if ($syncedLyrics !== null) { |
|
196
|
|
|
$result['synced'] = \array_map(fn($timestamp, $text) => [ |
|
197
|
|
|
'time' => \max(0, $timestamp), 'text' => $text |
|
198
|
|
|
], \array_keys($syncedLyrics), $syncedLyrics); |
|
199
|
|
|
} |
|
200
|
|
|
} else { |
|
201
|
|
|
$result = null; |
|
202
|
|
|
} |
|
203
|
|
|
|
|
204
|
|
|
return $result; |
|
205
|
|
|
} |
|
206
|
|
|
|
|
207
|
|
|
private static function getLyricsFileContent(File $audioFile) : ?string { |
|
208
|
|
|
$audioName = $audioFile->getName(); |
|
209
|
|
|
$bareName = \pathinfo($audioName, PATHINFO_FILENAME); |
|
210
|
|
|
$parentDir = $audioFile->getParent(); |
|
211
|
|
|
|
|
212
|
|
|
// if the audio file is named "someSong.mp3", we allow the lyrics file to be named "someSong.lrc" or "someSong.mp3.lrc" |
|
213
|
|
|
$lrcName1 = "$bareName.lrc"; |
|
214
|
|
|
$lrcName2 = "$audioName.lrc"; |
|
215
|
|
|
|
|
216
|
|
|
if ($parentDir->nodeExists($lrcName1)) { |
|
217
|
|
|
$lrcFile = $parentDir->get($lrcName1); |
|
218
|
|
|
} else if ($parentDir->nodeExists($lrcName2)) { |
|
219
|
|
|
$lrcFile = $parentDir->get($lrcName2); |
|
220
|
|
|
} |
|
221
|
|
|
|
|
222
|
|
|
return (isset($lrcFile) && $lrcFile instanceof File) ? $lrcFile->getContent() : null; |
|
223
|
|
|
} |
|
224
|
|
|
|
|
225
|
|
|
/** |
|
226
|
|
|
* Base64 encode the picture binary data and wrap it so that it can be directly used as |
|
227
|
|
|
* src of an HTML img element. |
|
228
|
|
|
*/ |
|
229
|
|
|
private static function encodePictureTag(array $pic) : ?string { |
|
230
|
|
|
if ($pic['data']) { |
|
231
|
|
|
return 'data:' . $pic['image_mime'] . ';base64,' . \base64_encode($pic['data']); |
|
232
|
|
|
} else { |
|
233
|
|
|
return null; |
|
234
|
|
|
} |
|
235
|
|
|
} |
|
236
|
|
|
|
|
237
|
|
|
/** |
|
238
|
|
|
* Remove potentially invalid characters from the string and normalize the line breaks to LF. |
|
239
|
|
|
* @param mixed $item |
|
240
|
|
|
*/ |
|
241
|
|
|
private static function sanitizeString(&$item) : void { |
|
242
|
|
|
if (\is_string($item)) { |
|
243
|
|
|
\mb_substitute_character(0xFFFD); // Use the Unicode REPLACEMENT CHARACTER (U+FFFD) |
|
244
|
|
|
$item = \mb_convert_encoding($item, 'UTF-8', 'UTF-8'); |
|
245
|
|
|
// The tags could contain line breaks in formats LF, CRLF, or CR, but we want the output |
|
246
|
|
|
// to always use the LF style. Note that the order of the next two lines is important! |
|
247
|
|
|
$item = \str_replace("\r\n", "\n", $item); |
|
248
|
|
|
$item = \str_replace("\r", "\n", $item); |
|
249
|
|
|
} |
|
250
|
|
|
} |
|
251
|
|
|
|
|
252
|
|
|
/** |
|
253
|
|
|
* In the 'comments' field from the extractor, the value for each key is a 1-element |
|
254
|
|
|
* array containing the actual tag value. Remove these intermediate arrays. |
|
255
|
|
|
*/ |
|
256
|
|
|
private static function flattenComments(array $array) : array { |
|
257
|
|
|
// key 'text' is an exception, its value is an associative array |
|
258
|
|
|
$textArray = null; |
|
259
|
|
|
|
|
260
|
|
|
foreach ($array as $key => $value) { |
|
261
|
|
|
if ($key === 'text') { |
|
262
|
|
|
$textArray = $value; |
|
263
|
|
|
} else { |
|
264
|
|
|
$array[$key] = $value[0]; |
|
265
|
|
|
} |
|
266
|
|
|
} |
|
267
|
|
|
|
|
268
|
|
|
if (!empty($textArray)) { |
|
269
|
|
|
$array = \array_merge($array, $textArray); |
|
270
|
|
|
unset($array['text']); |
|
271
|
|
|
} |
|
272
|
|
|
|
|
273
|
|
|
return $array; |
|
274
|
|
|
} |
|
275
|
|
|
} |
|
276
|
|
|
|