privatebin-mirror/tst/I18nTest.php

<?php declare(strict_types=1);

use PHPUnit\Framework\TestCase;
use PrivateBin\I18n;
use PrivateBin\Json;

class I18nMock extends I18n
{
    public static function resetAvailableLanguages()
    {
        self::$_availableLanguages = [];
    }

    public static function resetPath($path = '')
    {
        self::$_path = $path;
    }

    public static function getPath($file = '')
    {
        return self::_getPath($file);
    }
}

class I18nTest extends TestCase
{
    private $_translations = [];

    public static function setUpBeforeClass(): void
    {
        parent::setUpBeforeClass();
        fwrite(
            STDERR,
            sprintf(
                "I18n tests: Intl extension loaded=%s; locale_accept_from_http exists=%s; ICU version=%s\n",
                extension_loaded('intl') ? 'yes' : 'no',
                function_exists('locale_accept_from_http') ? 'yes' : 'no',
                defined('INTL_ICU_VERSION') ? INTL_ICU_VERSION : 'none'
            )
        );
    }

    public function setUp(): void
    {
        /* Setup Routine */
        $this->_translations = json_decode(
            file_get_contents(PATH . 'i18n' . DIRECTORY_SEPARATOR . 'de.json'),
            true
        );
    }

    public function tearDown(): void
    {
        unset($_COOKIE['lang'], $_SERVER['HTTP_ACCEPT_LANGUAGE']);
        I18n::loadTranslations();
    }

    public function testTranslationFallback()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'foobar';
        $messageId = 'It does not matter if the message ID exists';
        I18n::loadTranslations();
        $this->assertEquals($messageId, I18n::_($messageId), 'fallback to en');
        I18n::getLanguageLabels();
    }

    public function testCookieLanguageDeDetection()
    {
        $_COOKIE['lang'] = 'de';
        I18n::loadTranslations();
        $this->assertEquals($_COOKIE['lang'], I18n::getLanguage(), 'browser language de');
        $this->assertEquals('0 Stunden', I18n::_('%d hours', 0), '0 hours in German');
        $this->assertEquals('1 Stunde',  I18n::_('%d hours', 1), '1 hour in German');
        $this->assertEquals('2 Stunden', I18n::_('%d hours', 2), '2 hours in German');
    }

    public function testBrowserLanguageDeDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'de-CH,de;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2,fr;q=0.0';
        I18n::loadTranslations();
        $this->assertEquals('de', I18n::getLanguage(), 'browser language de');
        $this->assertEquals('0 Stunden', I18n::_('%d hours', 0), '0 hours in German');
        $this->assertEquals('1 Stunde',  I18n::_('%d hours', 1), '1 hour in German');
        $this->assertEquals('2 Stunden', I18n::_('%d hours', 2), '2 hours in German');
    }

    public function testBrowserLanguageFrDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'fr-CH,fr;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2,de;q=0.0';
        I18n::loadTranslations();
        $this->assertEquals('fr', I18n::getLanguage(), 'browser language fr');
        $this->assertEquals('0 heure',  I18n::_('%d hours', 0), '0 hours in French');
        $this->assertEquals('1 heure',  I18n::_('%d hours', 1), '1 hour in French');
        $this->assertEquals('2 heures', I18n::_('%d hours', 2), '2 hours in French');
    }

    /**
     * @dataProvider provideEqualQualityHeaders
     */
    public function testEqualQualityPreservesBrowserOrder($acceptedLanguage, $language)
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = $acceptedLanguage;
        I18n::loadTranslations();
        $this->assertEquals($language, I18n::getLanguage(), 'browser language ' . $acceptedLanguage);
    }

    public function provideEqualQualityHeaders()
    {
        // ICU before version 67 sorts languages of equal quality alphabetically
        return [
            'fr;q=0.8,de;q=0.8' => ['fr;q=0.8,de;q=0.8', 'fr'],
            'fr,de' => ['fr,de', 'fr'],
        ];
    }

    /**
     * @dataProvider provideLanguagesUnknownToIcuHeaders
     */
    public function testBrowserLanguageUnknownToIcuDetection($acceptedLanguage, $language)
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = $acceptedLanguage;
        I18n::loadTranslations();
        $this->assertEquals($language, I18n::getLanguage(), 'browser language ' . $acceptedLanguage);
    }

    public function provideLanguagesUnknownToIcuHeaders()
    {
        // ICU has no locale data for these languages, so it can only match the next one in the header
        return [
            'co,fr;q=0.8' => ['co,fr;q=0.8', 'co'],
            'jbo,de;q=0.8' => ['jbo,de;q=0.8', 'jbo'],
            'la,it;q=0.8' => ['la,it;q=0.8', 'la'],
        ];
    }

    public function testBulgarianMonthExpirationTranslation()
    {
        $_COOKIE['lang'] = 'bg';
        I18n::loadTranslations();

        $this->assertEquals(
            'Този документ изтича след един месец.',
            I18n::_('This document will expire in %d months.', 1)
        );
        $this->assertEquals(
            'Този документ изтича след 2 месеца.',
            I18n::_('This document will expire in %d months.', 2)
        );
    }

    public function testBrowserLanguageLtDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'lt;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('lt', I18n::getLanguage(), 'browser language lt');
        $this->assertEquals('1 sekundę', I18n::_('%d seconds', 1), '1 second in Lithuanian');
        $this->assertEquals('2 sekundės', I18n::_('%d seconds', 2), '2 seconds in Lithuanian');
        $this->assertEquals('10 sekundžių', I18n::_('%d seconds', 10), '10 seconds in Lithuanian');
        $this->assertEquals('22 sekundės', I18n::_('%d seconds', 22), '22 seconds in Lithuanian');
        $this->assertEquals('110 sekundžių', I18n::_('%d seconds', 110), '110 seconds in Lithuanian');
        $this->assertEquals('111 sekundžių', I18n::_('%d seconds', 111), '111 seconds in Lithuanian');
        $this->assertEquals('120 sekundžių', I18n::_('%d seconds', 120), '120 seconds in Lithuanian');
    }

    public function testBrowserLanguageNoDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'no;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('no', I18n::getLanguage(), 'browser language no');
        $this->assertEquals('0 timer',  I18n::_('%d hours', 0), '0 hours in Norwegian');
        $this->assertEquals('1 time',  I18n::_('%d hours', 1), '1 hour in Norwegian');
        $this->assertEquals('2 timer', I18n::_('%d hours', 2), '2 hours in Norwegian');
    }

    public function testBrowserLanguageOcDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'oc;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('oc', I18n::getLanguage(), 'browser language oc');
        $this->assertEquals('0 ora',  I18n::_('%d hours', 0), '0 hours in Occitan');
        $this->assertEquals('1 ora',  I18n::_('%d hours', 1), '1 hour in Occitan');
        $this->assertEquals('2 oras', I18n::_('%d hours', 2), '2 hours in Occitan');
    }

    public function testBrowserLanguageZhDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'zh;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('zh', I18n::getLanguage(), 'browser language zh');
        $this->assertEquals('0 小时',  I18n::_('%d hours', 0), '0 hours in Chinese');
        $this->assertEquals('1 小时',  I18n::_('%d hours', 1), '1 hour in Chinese');
        $this->assertEquals('2 小时', I18n::_('%d hours', 2), '2 hours in Chinese');
    }

    public function testBrowserLanguageZhHantDetection()
    {
        foreach ([
            'zh-Hant,zh;q=0.8,en;q=0.2'            => 'zh-tw',
            'zh-Hant-TW,zh;q=0.8,en;q=0.2'         => 'zh-tw',
            'zh-TW,en;q=0.2'                       => 'zh-tw',
            'zh-TW-u-nu-hanidec,en;q=0.2'          => 'zh-tw',
            'zh-HK,zh;q=0.8,en;q=0.2'              => 'zh-tw',
            'zh-MO,zh;q=0.8,en;q=0.2'              => 'zh-tw',
            'zh-Hant,zh-Hans;q=0.8,en;q=0.2'       => 'zh-tw',
            'zh-CN,zh-Hant;q=0.8,en;q=0.2'         => 'zh',
            'zh-Hans,zh-Hant;q=0.8,en;q=0.2'       => 'zh',
            'zh-Hans-CN,en;q=0.2'                  => 'zh',
            'zh;q=0.9,zh-Hant;q=0.8,en;q=0.2'      => 'zh',
        ] as $acceptedLanguage => $language) {
            $_SERVER['HTTP_ACCEPT_LANGUAGE'] = $acceptedLanguage;
            I18n::loadTranslations();
            $this->assertEquals($language, I18n::getLanguage(), 'browser language ' . $acceptedLanguage);
        }
    }

    /**
     * @dataProvider provideTraditionalChineseWithZhFallbackHeaders
     */
    public function testBrowserLanguageZhHantWithZhFallbackDetection($acceptedLanguage, $language)
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = $acceptedLanguage;
        I18n::loadTranslations();
        $this->assertEquals($language, I18n::getLanguage(), 'browser language ' . $acceptedLanguage);
    }

    public function provideTraditionalChineseWithZhFallbackHeaders()
    {
        // typical Chrome headers from Taiwan and Hong Kong, ICU before version 67
        // has zh_TW and zh_HK only as aliases, skips them and returns zh instead
        return [
            'zh-TW,zh;q=0.9,en-US;q=0.8,en;q=0.7' => ['zh-TW,zh;q=0.9,en-US;q=0.8,en;q=0.7', 'zh-tw'],
            'zh-HK,zh;q=0.9,en;q=0.8' => ['zh-HK,zh;q=0.9,en;q=0.8', 'zh-tw'],
        ];
    }

    public function testBrowserLanguagePlDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'pl;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('pl', I18n::getLanguage(), 'browser language pl');
        $this->assertEquals('1 godzina', I18n::_('%d hours', 1), '1 hour in Polish');
        $this->assertEquals('2 godziny', I18n::_('%d hours', 2), '2 hours in Polish');
        $this->assertEquals('12 godzin', I18n::_('%d hours', 12), '12 hours in Polish');
        $this->assertEquals('22 godziny', I18n::_('%d hours', 22), '22 hours in Polish');
        $this->assertEquals('1 minuta',  I18n::_('%d minutes', 1), '1 minute in Polish');
        $this->assertEquals('3 minuty',  I18n::_('%d minutes', 3), '3 minutes in Polish');
        $this->assertEquals('13 minut',  I18n::_('%d minutes', 13), '13 minutes in Polish');
        $this->assertEquals('23 minuty',  I18n::_('%d minutes', 23), '23 minutes in Polish');
    }

    public function testBrowserLanguageRuDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'ru;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('ru', I18n::getLanguage(), 'browser language ru');
        $this->assertEquals('1 минуту',  I18n::_('%d minutes', 1), '1 minute in Russian');
        $this->assertEquals('3 минуты',  I18n::_('%d minutes', 3), '3 minutes in Russian');
        $this->assertEquals('10 минут',  I18n::_('%d minutes', 10), '10 minutes in Russian');
        $this->assertEquals('21 минуту',  I18n::_('%d minutes', 21), '21 minutes in Russian');
    }

    public function testBrowserLanguageSlDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'sl;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('sl', I18n::getLanguage(), 'browser language sl');
        $this->assertEquals('0 ur',  I18n::_('%d hours', 0), '0 hours in Slovene');
        $this->assertEquals('1 ura',  I18n::_('%d hours', 1), '1 hour in Slovene');
        $this->assertEquals('2 uri', I18n::_('%d hours', 2), '2 hours in Slovene');
        $this->assertEquals('3 ure',  I18n::_('%d hours', 3), '3 hours in Slovene');
        $this->assertEquals('11 ur',  I18n::_('%d hours', 11), '11 hours in Slovene');
        $this->assertEquals('101 ura',  I18n::_('%d hours', 101), '101 hours in Slovene');
        $this->assertEquals('102 uri', I18n::_('%d hours', 102), '102 hours in Slovene');
        $this->assertEquals('104 ure',  I18n::_('%d hours', 104), '104 hours in Slovene');
    }

    public function testBrowserLanguageCsDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'cs;q=0.8,en-GB;q=0.6,en-US;q=0.4,en;q=0.2';
        I18n::loadTranslations();
        $this->assertEquals('cs', I18n::getLanguage(), 'browser language cs');
        $this->assertEquals('1 hodina', I18n::_('%d hours', 1), '1 hour in Czech');
        $this->assertEquals('2 hodiny', I18n::_('%d hours', 2), '2 hours in Czech');
        $this->assertEquals('5 minut',  I18n::_('%d minutes', 5), '5 minutes in Czech');
        $this->assertEquals('14 minut',  I18n::_('%d minutes', 14), '14 minutes in Czech');
    }

    public function testInvalidBrowserLanguageQualityIsIgnored()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'de;q=2,en;q=0.8';
        $this->assertSame(['0.8' => ['en']], I18n::getBrowserLanguages());
    }

    public function testBrowserLanguageAnyDetection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = '*';
        I18n::loadTranslations();
        $this->assertTrue(strlen(I18n::getLanguage()) >= 2, 'browser language any');
    }

    public function testVariableInjection()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'foobar';
        I18n::loadTranslations();
        $this->assertEquals('some string + 1', I18n::_('some %s + %d', 'string', 1), 'browser language en');
    }

    public function testHtmlEntityEncoding()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'foobar';
        I18n::loadTranslations();
        $input = '&<>"\'/`=';
        $result = htmlspecialchars($input, ENT_QUOTES | ENT_HTML5 | ENT_DISALLOWED, 'UTF-8', false);
        $this->assertEquals($result, I18n::encode($input), 'encodes HTML entities');
        $this->assertEquals('<a>some ' . $result . ' + 1</a>', I18n::_('<a>some %s + %d</a>', $input, 1), 'encodes parameters in translations');
        // Message ID should NOT be encoded (it comes from trusted source), only the parameter should be
        $this->assertEquals($input . $result, I18n::_($input . '%s', $input), 'encodes only parameters, not message ID');
    }

    public function testApostropheEncodngInMessage()
    {
        $_SERVER['HTTP_ACCEPT_LANGUAGE'] = 'fr';
        I18n::loadTranslations();
        // For example, the French translation should not have the apostrophe encoded
        // See https://github.com/PrivateBin/PrivateBin/issues/1712
        $message = I18n::_('Document does not exist, has expired or has been deleted.');
        $this->assertFalse(str_contains($message, '&apos;'), 'French apostrophe should not be encoded in translation message');
        $this->assertTrue(str_contains($message, "n'existe"), 'French apostrophe should be present as literal character');
    }

    public function testFallbackAlwaysPresent()
    {
        $path = sys_get_temp_dir() . DIRECTORY_SEPARATOR . 'privatebin_i18n';
        if (!is_dir($path)) {
            mkdir($path);
        }

        $languageIterator = new AppendIterator();
        $languageIterator->append(new GlobIterator(I18nMock::getPath('??.json')));
        $languageIterator->append(new GlobIterator(I18nMock::getPath('???.json'))); // for jbo
        $languageIterator->append(new GlobIterator(I18nMock::getPath('??-??.json'))); // for regional variants like zh-tw
        $languageCount = 0;
        foreach ($languageIterator as $file) {
            ++$languageCount;
            $this->assertTrue(copy($file->getPathname(), $path . DIRECTORY_SEPARATOR . $file->getBasename()));
        }

        I18nMock::resetPath($path);
        $languagesDevelopment = I18nMock::getAvailableLanguages();
        $this->assertEquals($languageCount, count($languagesDevelopment), 'all copied languages detected');
        $this->assertTrue(in_array('en', $languagesDevelopment), 'English fallback present');

        unlink($path . DIRECTORY_SEPARATOR . 'en.json');
        I18nMock::resetAvailableLanguages();
        $languagesDeployed = I18nMock::getAvailableLanguages();
        $this->assertEquals($languageCount, count($languagesDeployed), 'all copied languages detected, plus fallback');
        $this->assertTrue(in_array('en', $languagesDeployed), 'English fallback still present');

        I18nMock::resetAvailableLanguages();
        I18nMock::resetPath();
        Helper::rmDir($path);
    }

    public function testGetCopyHotkey()
    {
        $_SERVER['HTTP_USER_AGENT'] = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7)';
        $this->assertEquals('Cmd', I18n::getCopyHotkey(), 'returns Cmd on macOS');

        $_SERVER['HTTP_USER_AGENT'] = 'Mozilla/5.0 (Windows NT 10.0; Win64; x64)';
        $this->assertEquals('Ctrl', I18n::getCopyHotkey(), 'returns Ctrl on Windows');

        $_SERVER['HTTP_USER_AGENT'] = 'Mozilla/5.0 (X11; Linux x86_64)';
        $this->assertEquals('Ctrl', I18n::getCopyHotkey(), 'returns Ctrl on Linux');

        unset($_SERVER['HTTP_USER_AGENT']);
        $this->assertEquals('Ctrl', I18n::getCopyHotkey(), 'returns Ctrl when user agent absent');
    }

    public function testRightToLeftLanguages()
    {
        foreach (['ar', 'fa', 'he'] as $language) {
            $_COOKIE['lang'] = $language;
            I18n::loadTranslations();
            $this->assertTrue(I18n::isRtl(), "$language is right-to-left");
        }

        $_COOKIE['lang'] = 'en';
        I18n::loadTranslations();
        $this->assertFalse(I18n::isRtl(), 'English is left-to-right');
    }

    public function testMessageIdsExistInAllLanguages()
    {
        $messageIds = [];
        $languages = [];
        foreach (new DirectoryIterator(PATH . 'i18n') as $file) {
            $fileNameLength = strlen($file->getFilename());
            if ($fileNameLength === 7) {       // xx.json
                $language = substr($file->getFilename(), 0, 2);
            } elseif ($fileNameLength === 8) { // jbo.json
                $language = substr($file->getFilename(), 0, 3);
            } elseif ($fileNameLength === 10) { // xx-xx.json
                $language = substr($file->getFilename(), 0, 5);
            } else {
                continue;
            }
            $languageJson = file_get_contents($file->getPathname());
            $languageMessageIds = array_keys(Json::decode($languageJson));
            $messageIds = array_unique(array_merge($messageIds, $languageMessageIds));
            $languages[$language] = $languageMessageIds;
        }
        foreach ($messageIds as $messageId) {
            foreach (array_keys($languages) as $language) {
                // most languages don't translate the data size units, ignore those
                if ($messageId !== 'B' && strlen($messageId) !== 3 && strpos($messageId, 'B', 2) !== 2) {
                    $this->assertContains(
                        $messageId,
                        $languages[$language],
                        "message ID '$messageId' exists in translation file $language.json"
                    );
                }
            }
        }
    }
}