Passed
Push — develop ( b53f60...9961ae )
by Greg
14:15
created

I18N::direction()   A

Complexity

Conditions 1
Paths 1

Size

Total Lines 3
Code Lines 1

Duplication

Lines 0
Ratio 0 %

Importance

Changes 0
Metric Value
cc 1
eloc 1
nc 1
nop 0
dl 0
loc 3
rs 10
c 0
b 0
f 0
1
<?php
2
3
/**
4
 * webtrees: online genealogy
5
 * Copyright (C) 2022 webtrees development team
6
 * This program is free software: you can redistribute it and/or modify
7
 * it under the terms of the GNU General Public License as published by
8
 * the Free Software Foundation, either version 3 of the License, or
9
 * (at your option) any later version.
10
 * This program is distributed in the hope that it will be useful,
11
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
12
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
13
 * GNU General Public License for more details.
14
 * You should have received a copy of the GNU General Public License
15
 * along with this program. If not, see <https://www.gnu.org/licenses/>.
16
 */
17
18
declare(strict_types=1);
19
20
namespace Fisharebest\Webtrees;
21
22
use Closure;
23
use Collator;
24
use Exception;
25
use Fisharebest\Localization\Locale;
26
use Fisharebest\Localization\Locale\LocaleEnUs;
27
use Fisharebest\Localization\Locale\LocaleInterface;
28
use Fisharebest\Localization\Translation;
29
use Fisharebest\Localization\Translator;
30
use Fisharebest\Webtrees\Module\ModuleCustomInterface;
31
use Fisharebest\Webtrees\Module\ModuleLanguageInterface;
32
use Fisharebest\Webtrees\Services\ModuleService;
33
34
use function array_merge;
35
use function class_exists;
36
use function html_entity_decode;
37
use function in_array;
38
use function mb_strtolower;
39
use function mb_strtoupper;
40
use function mb_substr;
41
use function ord;
42
use function sprintf;
43
use function str_contains;
44
use function str_replace;
45
use function strcmp;
46
use function strip_tags;
47
use function strlen;
48
use function strtr;
49
use function var_export;
50
51
/**
52
 * Internationalization (i18n) and localization (l10n).
53
 */
54
class I18N
55
{
56
    // MO files use special characters for plurals and context.
57
    public const PLURAL  = "\x00";
58
    public const CONTEXT = "\x04";
59
60
    // Digits are always rendered LTR, even in RTL text.
61
    private const DIGITS = '0123456789٠١٢٣٤٥٦٧٨٩۰۱۲۳۴۵۶۷۸۹';
62
63
    // These locales need special handling for the dotless letter I.
64
    private const DOTLESS_I_LOCALES = [
65
        'az',
66
        'tr',
67
    ];
68
69
    private const DOTLESS_I_TOLOWER = [
70
        'I' => 'ı',
71
        'İ' => 'i',
72
    ];
73
74
    private const DOTLESS_I_TOUPPER = [
75
        'ı' => 'I',
76
        'i' => 'İ',
77
    ];
78
79
    // The ranges of characters used by each script.
80
    private const SCRIPT_CHARACTER_RANGES = [
81
        [
82
            'Latn',
83
            0x0041,
84
            0x005A,
85
        ],
86
        [
87
            'Latn',
88
            0x0061,
89
            0x007A,
90
        ],
91
        [
92
            'Latn',
93
            0x0100,
94
            0x02AF,
95
        ],
96
        [
97
            'Grek',
98
            0x0370,
99
            0x03FF,
100
        ],
101
        [
102
            'Cyrl',
103
            0x0400,
104
            0x052F,
105
        ],
106
        [
107
            'Hebr',
108
            0x0590,
109
            0x05FF,
110
        ],
111
        [
112
            'Arab',
113
            0x0600,
114
            0x06FF,
115
        ],
116
        [
117
            'Arab',
118
            0x0750,
119
            0x077F,
120
        ],
121
        [
122
            'Arab',
123
            0x08A0,
124
            0x08FF,
125
        ],
126
        [
127
            'Deva',
128
            0x0900,
129
            0x097F,
130
        ],
131
        [
132
            'Taml',
133
            0x0B80,
134
            0x0BFF,
135
        ],
136
        [
137
            'Sinh',
138
            0x0D80,
139
            0x0DFF,
140
        ],
141
        [
142
            'Thai',
143
            0x0E00,
144
            0x0E7F,
145
        ],
146
        [
147
            'Geor',
148
            0x10A0,
149
            0x10FF,
150
        ],
151
        [
152
            'Grek',
153
            0x1F00,
154
            0x1FFF,
155
        ],
156
        [
157
            'Deva',
158
            0xA8E0,
159
            0xA8FF,
160
        ],
161
        [
162
            'Hans',
163
            0x3000,
164
            0x303F,
165
        ],
166
        // Mixed CJK, not just Hans
167
        [
168
            'Hans',
169
            0x3400,
170
            0xFAFF,
171
        ],
172
        // Mixed CJK, not just Hans
173
        [
174
            'Hans',
175
            0x20000,
176
            0x2FA1F,
177
        ],
178
        // Mixed CJK, not just Hans
179
    ];
180
181
    // Characters that are displayed in mirror form in RTL text.
182
    private const MIRROR_CHARACTERS = [
183
        '('  => ')',
184
        ')'  => '(',
185
        '['  => ']',
186
        ']'  => '[',
187
        '{'  => '}',
188
        '}'  => '{',
189
        '<'  => '>',
190
        '>'  => '<',
191
        '‹ ' => '›',
192
        '› ' => '‹',
193
        '«'  => '»',
194
        '»'  => '«',
195
        '﴾ ' => '﴿',
196
        '﴿ ' => '﴾',
197
        '“ ' => '”',
198
        '” ' => '“',
199
        '‘ ' => '’',
200
        '’ ' => '‘',
201
    ];
202
203
    // Punctuation used to separate list items, typically a comma
204
    public static string $list_separator;
205
206
    private static ?ModuleLanguageInterface $language;
207
208
    private static LocaleInterface $locale;
209
210
    private static Translator $translator;
211
212
    private static ?Collator $collator = null;
213
214
    /**
215
     * The preferred locales for this site, or a default list if no preference.
216
     *
217
     * @return array<LocaleInterface>
218
     */
219
    public static function activeLocales(): array
220
    {
221
        $locales = app(ModuleService::class)
222
            ->findByInterface(ModuleLanguageInterface::class, false, true)
223
            ->map(static function (ModuleLanguageInterface $module): LocaleInterface {
224
                return $module->locale();
225
            });
226
227
        if ($locales->isEmpty()) {
228
            return [new LocaleEnUs()];
229
        }
230
231
        return $locales->all();
232
    }
233
234
    /**
235
     * Which MySQL collation should be used for this locale?
236
     *
237
     * @return string
238
     */
239
    public static function collation(): string
240
    {
241
        $collation = self::$locale->collation();
242
        switch ($collation) {
243
            case 'croatian_ci':
244
            case 'german2_ci':
245
            case 'vietnamese_ci':
246
                // Only available in MySQL 5.6
247
                return 'utf8_unicode_ci';
248
            default:
249
                return 'utf8_' . $collation;
250
        }
251
    }
252
253
    /**
254
     * What format is used to display dates in the current locale?
255
     *
256
     * @return string
257
     */
258
    public static function dateFormat(): string
259
    {
260
        /* I18N: This is the format string for full dates. See https://php.net/date for codes */
261
        return self::$translator->translate('%j %F %Y');
262
    }
263
264
    /**
265
     * Convert the digits 0-9 into the local script
266
     * Used for years, etc., where we do not want thousands-separators, decimals, etc.
267
     *
268
     * @param string|int $n
269
     *
270
     * @return string
271
     */
272
    public static function digits(string|int $n): string
273
    {
274
        return self::$locale->digits((string) $n);
275
    }
276
277
    /**
278
     * What is the direction of the current locale
279
     *
280
     * @return string "ltr" or "rtl"
281
     */
282
    public static function direction(): string
283
    {
284
        return self::$locale->direction();
285
    }
286
287
    /**
288
     * Initialise the translation adapter with a locale setting.
289
     *
290
     * @param string $code
291
     * @param bool   $setup
292
     *
293
     * @return void
294
     */
295
    public static function init(string $code, bool $setup = false): void
296
    {
297
        self::$locale = Locale::create($code);
298
299
        // Load the translation file
300
        $translation_file = __DIR__ . '/../resources/lang/' . self::$locale->languageTag() . '/messages.php';
301
302
        try {
303
            $translation  = new Translation($translation_file);
304
            $translations = $translation->asArray();
305
        } catch (Exception $ex) {
306
            // The translations files are created during the build process, and are
307
            // not included in the source code.
308
            // Assuming we are using dev code, and build (or rebuild) the files.
309
            $po_file      = Webtrees::ROOT_DIR . 'resources/lang/' . self::$locale->languageTag() . '/messages.po';
310
            $translation  = new Translation($po_file);
311
            $translations = $translation->asArray();
312
            file_put_contents($translation_file, "<?php\n\nreturn " . var_export($translations, true) . ";\n");
313
        }
314
315
        // Add translations from custom modules (but not during setup, as we have no database/modules)
316
        if (!$setup) {
317
            $module_service = app(ModuleService::class);
318
319
            $translations = $module_service
320
                ->findByInterface(ModuleCustomInterface::class)
321
                ->reduce(static function (array $carry, ModuleCustomInterface $item): array {
322
                    return array_merge($carry, $item->customTranslations(self::$locale->languageTag()));
323
                }, $translations);
0 ignored issues
show
Bug introduced by
$translations of type array is incompatible with the type Illuminate\Support\Traits\TReduceInitial expected by parameter $initial of Illuminate\Support\Collection::reduce(). ( Ignorable by Annotation )

If this is a false-positive, you can also ignore this issue in your code via the ignore-type  annotation

323
                }, /** @scrutinizer ignore-type */ $translations);
Loading history...
324
325
            self::$language = $module_service
326
                ->findByInterface(ModuleLanguageInterface::class)
327
                ->first(fn (ModuleLanguageInterface $module): bool => $module->locale()->languageTag() === $code);
328
        }
329
330
        // Create a translator
331
        self::$translator = new Translator($translations, self::$locale->pluralRule());
332
333
        /* I18N: This punctuation is used to separate lists of items */
334
        self::$list_separator = self::translate(', ');
335
336
        // Create a collator
337
        try {
338
            // Symfony provides a very incomplete polyfill - which cannot be used.
339
            if (class_exists('Collator')) {
340
                // Need phonebook collation rules for German Ä, Ö and Ü.
341
                if (str_contains(self::$locale->code(), '@')) {
342
                    self::$collator = new Collator(self::$locale->code() . ';collation=phonebook');
343
                } else {
344
                    self::$collator = new Collator(self::$locale->code() . '@collation=phonebook');
345
                }
346
                // Ignore upper/lower case differences
347
                self::$collator->setStrength(Collator::SECONDARY);
348
            }
349
        } catch (Exception $ex) {
350
            // PHP-INTL is not installed?  We'll use a fallback later.
351
        }
352
    }
353
354
    /**
355
     * Translate a string, and then substitute placeholders
356
     * echo I18N::translate('Hello World!');
357
     * echo I18N::translate('The %s sat on the mat', 'cat');
358
     *
359
     * @param string $message
360
     * @param string ...$args
361
     *
362
     * @return string
363
     */
364
    public static function translate(string $message, ...$args): string
365
    {
366
        $message = self::$translator->translate($message);
367
368
        return sprintf($message, ...$args);
369
    }
370
371
    /**
372
     * @return string
373
     */
374
    public static function languageTag(): string
375
    {
376
        return self::$locale->languageTag();
377
    }
378
379
    /**
380
     * @return LocaleInterface
381
     */
382
    public static function locale(): LocaleInterface
383
    {
384
        return self::$locale;
385
    }
386
387
    /**
388
     * @return ModuleLanguageInterface
389
     */
390
    public static function language(): ModuleLanguageInterface
391
    {
392
        return self::$language;
393
    }
394
395
    /**
396
     * Translate a number into the local representation.
397
     * e.g. 12345.67 becomes
398
     * en: 12,345.67
399
     * fr: 12 345,67
400
     * de: 12.345,67
401
     *
402
     * @param float $n
403
     * @param int   $precision
404
     *
405
     * @return string
406
     */
407
    public static function number(float $n, int $precision = 0): string
408
    {
409
        return self::$locale->number(round($n, $precision));
410
    }
411
412
    /**
413
     * Translate a fraction into a percentage.
414
     * e.g. 0.123 becomes
415
     * en: 12.3%
416
     * fr: 12,3 %
417
     * de: 12,3%
418
     *
419
     * @param float $n
420
     * @param int   $precision
421
     *
422
     * @return string
423
     */
424
    public static function percentage(float $n, int $precision = 0): string
425
    {
426
        return self::$locale->percent(round($n, $precision + 2));
427
    }
428
429
    /**
430
     * Translate a plural string
431
     * echo self::plural('There is an error', 'There are errors', $num_errors);
432
     * echo self::plural('There is one error', 'There are %s errors', $num_errors);
433
     * echo self::plural('There is %1$s %2$s cat', 'There are %1$s %2$s cats', $num, $num, $colour);
434
     *
435
     * @param string $singular
436
     * @param string $plural
437
     * @param int    $count
438
     * @param string ...$args
439
     *
440
     * @return string
441
     */
442
    public static function plural(string $singular, string $plural, int $count, ...$args): string
443
    {
444
        $message = self::$translator->translatePlural($singular, $plural, $count);
445
446
        return sprintf($message, ...$args);
447
    }
448
449
    /**
450
     * UTF8 version of PHP::strrev()
451
     * Reverse RTL text for third-party libraries such as GD2 and googlechart.
452
     * These do not support UTF8 text direction, so we must mimic it for them.
453
     * Numbers are always rendered LTR, even in RTL text.
454
     * The visual direction of characters such as parentheses should be reversed.
455
     *
456
     * @param string $text Text to be reversed
457
     *
458
     * @return string
459
     */
460
    public static function reverseText(string $text): string
461
    {
462
        // Remove HTML markup - we can't display it and it is LTR.
463
        $text = strip_tags($text);
464
        // Remove HTML entities.
465
        $text = html_entity_decode($text, ENT_QUOTES, 'UTF-8');
466
467
        // LTR text doesn't need reversing
468
        if (self::scriptDirection(self::textScript($text)) === 'ltr') {
469
            return $text;
470
        }
471
472
        // Mirrored characters
473
        $text = strtr($text, self::MIRROR_CHARACTERS);
474
475
        $reversed = '';
476
        $digits   = '';
477
        while ($text !== '') {
478
            $letter = mb_substr($text, 0, 1);
479
            $text   = mb_substr($text, 1);
480
            if (str_contains(self::DIGITS, $letter)) {
481
                $digits .= $letter;
482
            } else {
483
                $reversed = $letter . $digits . $reversed;
484
                $digits   = '';
485
            }
486
        }
487
488
        return $digits . $reversed;
489
    }
490
491
    /**
492
     * Return the direction (ltr or rtl) for a given script
493
     * The PHP/intl library does not provde this information, so we need
494
     * our own lookup table.
495
     *
496
     * @param string $script
497
     *
498
     * @return string
499
     */
500
    public static function scriptDirection(string $script): string
501
    {
502
        switch ($script) {
503
            case 'Arab':
504
            case 'Hebr':
505
            case 'Mong':
506
            case 'Thaa':
507
                return 'rtl';
508
            default:
509
                return 'ltr';
510
        }
511
    }
512
513
    /**
514
     * Identify the script used for a piece of text
515
     *
516
     * @param string $string
517
     *
518
     * @return string
519
     */
520
    public static function textScript(string $string): string
521
    {
522
        $string = strip_tags($string); // otherwise HTML tags show up as latin
523
        $string = html_entity_decode($string, ENT_QUOTES, 'UTF-8'); // otherwise HTML entities show up as latin
524
        $string = str_replace([
525
            Individual::NOMEN_NESCIO,
526
            Individual::PRAENOMEN_NESCIO,
527
        ], '', $string);
528
        $pos    = 0;
529
        $strlen = strlen($string);
530
        while ($pos < $strlen) {
531
            // get the Unicode Code Point for the character at position $pos
532
            $byte1 = ord($string[$pos]);
533
            if ($byte1 < 0x80) {
534
                $code_point = $byte1;
535
                $chrlen     = 1;
536
            } elseif ($byte1 < 0xC0) {
537
                // Invalid continuation character
538
                return 'Latn';
539
            } elseif ($byte1 < 0xE0) {
540
                $code_point = (($byte1 & 0x1F) << 6) + (ord($string[$pos + 1]) & 0x3F);
541
                $chrlen     = 2;
542
            } elseif ($byte1 < 0xF0) {
543
                $code_point = (($byte1 & 0x0F) << 12) + ((ord($string[$pos + 1]) & 0x3F) << 6) + (ord($string[$pos + 2]) & 0x3F);
544
                $chrlen     = 3;
545
            } elseif ($byte1 < 0xF8) {
546
                $code_point = (($byte1 & 0x07) << 24) + ((ord($string[$pos + 1]) & 0x3F) << 12) + ((ord($string[$pos + 2]) & 0x3F) << 6) + (ord($string[$pos + 3]) & 0x3F);
547
                $chrlen     = 3;
548
            } else {
549
                // Invalid UTF
550
                return 'Latn';
551
            }
552
553
            foreach (self::SCRIPT_CHARACTER_RANGES as $range) {
554
                if ($code_point >= $range[1] && $code_point <= $range[2]) {
555
                    return $range[0];
556
                }
557
            }
558
            // Not a recognised script. Maybe punctuation, spacing, etc. Keep looking.
559
            $pos += $chrlen;
560
        }
561
562
        return 'Latn';
563
    }
564
565
    /**
566
     * A closure which will compare strings using local collation rules.
567
     *
568
     * @return Closure
569
     */
570
    public static function comparator(): Closure
571
    {
572
        if (self::$collator instanceof Collator) {
573
            return static fn (string $x, string $y): int => (int) self::$collator->compare($x, $y);
574
        }
575
576
        return static fn (string $x, string $y): int => strcmp(self::strtolower($x), self::strtolower($y));
577
    }
578
579
580
581
    /**
582
     * Convert a string to lower case.
583
     *
584
     * @param string $string
585
     *
586
     * @return string
587
     */
588
    public static function strtolower(string $string): string
589
    {
590
        if (in_array(self::$locale->language()->code(), self::DOTLESS_I_LOCALES, true)) {
591
            $string = strtr($string, self::DOTLESS_I_TOLOWER);
592
        }
593
594
        return mb_strtolower($string);
595
    }
596
597
    /**
598
     * Convert a string to upper case.
599
     *
600
     * @param string $string
601
     *
602
     * @return string
603
     */
604
    public static function strtoupper(string $string): string
605
    {
606
        if (in_array(self::$locale->language()->code(), self::DOTLESS_I_LOCALES, true)) {
607
            $string = strtr($string, self::DOTLESS_I_TOUPPER);
608
        }
609
610
        return mb_strtoupper($string);
611
    }
612
613
    /**
614
     * What format is used to display dates in the current locale?
615
     *
616
     * @return string
617
     */
618
    public static function timeFormat(): string
619
    {
620
        /* I18N: This is the format string for the time-of-day. See https://php.net/date for codes */
621
        return self::$translator->translate('%H:%i:%s');
622
    }
623
624
    /**
625
     * Context sensitive version of translate.
626
     * echo I18N::translateContext('NOMINATIVE', 'January');
627
     * echo I18N::translateContext('GENITIVE', 'January');
628
     *
629
     * @param string $context
630
     * @param string $message
631
     * @param string ...$args
632
     *
633
     * @return string
634
     */
635
    public static function translateContext(string $context, string $message, ...$args): string
636
    {
637
        $message = self::$translator->translateContext($context, $message);
638
639
        return sprintf($message, ...$args);
640
    }
641
}
642