Strings::toAscii - simplified process when Transliterate is available

Author: dg

src/Utils/Strings.php

@@ -150,34 +150,47 @@ public static function toAscii(string $s): string
 			['"', '"', '"', "'", "'", "'", '^', 'Ya', 'ya', 'Yu', 'yu'],
 			$s
 		);
-		// temporarily hide these characters to distinguish them from the garbage that iconv creates
-		$s = strtr($s, '`\'"^~?', "\x01\x02\x03\x04\x05\x06");
-		if ($transliterator !== null) {
+
+		if ($transliterator) {
 			$s = $transliterator->transliterate($s);
-		}
-		if (ICONV_IMPL === 'glibc') {
-			// glibc implementation is very limited. replace some characters directly
-			$s = str_replace(
-				["\u{BB}", "\u{AB}", "\u{2026}", "\u{2122}", "\u{A9}", "\u{AE}"], // » « … ™ © ®
-				['>>', '<<', '...', 'TM', '(c)', '(R)'],
-				$s
-			);
-			// transliterate the rest into Windows-1250 and then into ASCII, so most Eastern European characters are preserved
-			$s = iconv('UTF-8', 'WINDOWS-1250//TRANSLIT//IGNORE', $s);
-			$s = strtr($s, "\xa5\xa3\xbc\x8c\xa7\x8a\xaa\x8d\x8f\x8e\xaf\xb9\xb3\xbe\x9c\x9a\xba\x9d\x9f\x9e"
-				. "\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3"
-				. "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8"
-				. "\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe"
-				. "\x96\xa0\x8b\x97\x9b\xa6\xad\xb7",
-				'ALLSSSSTZZZallssstzzzRAAAALCCCEEEEIIDDNNOOOOxRUUUUYTsraaaalccceeeeiiddnnooooruuuuyt- <->|-.');
-			$s = self::pcre('preg_replace', ['#[^\x00-\x7F]++#', '', $s]);
-		} else {
+			if (ICONV_IMPL === 'glibc') {
+				// temporarily hide ? to distinguish them from the garbage that iconv creates
+				$s = strtr($s, '?', "\x01");
+			}
+			// use iconv because The transliterator leaves some characters out of ASCII, eg → ʾ
 			$s = iconv('UTF-8', 'ASCII//TRANSLIT//IGNORE', $s);
+			if (ICONV_IMPL === 'glibc') {
+				// remove garbage and restore ? characters
+				$s = str_replace(['?', "\x01"], ['', '?'], $s);
+			}
+		} else {
+			// temporarily hide these characters to distinguish them from the garbage that iconv creates
+			$s = strtr($s, '`\'"^~?', "\x01\x02\x03\x04\x05\x06");
+			if (ICONV_IMPL === 'glibc') {
+				// glibc implementation is very limited. replace some characters directly
+				$s = str_replace(
+					["\u{BB}", "\u{AB}", "\u{2026}", "\u{2122}", "\u{A9}", "\u{AE}"], // » « … ™ © ®
+					['>>', '<<', '...', 'TM', '(c)', '(R)'],
+					$s
+				);
+				// transliterate the rest into Windows-1250 and then into ASCII, so most Eastern European characters are preserved
+				$s = iconv('UTF-8', 'WINDOWS-1250//TRANSLIT//IGNORE', $s);
+				$s = strtr($s, "\xa5\xa3\xbc\x8c\xa7\x8a\xaa\x8d\x8f\x8e\xaf\xb9\xb3\xbe\x9c\x9a\xba\x9d\x9f\x9e"
+					. "\xbf\xc0\xc1\xc2\xc3\xc4\xc5\xc6\xc7\xc8\xc9\xca\xcb\xcc\xcd\xce\xcf\xd0\xd1\xd2\xd3"
+					. "\xd4\xd5\xd6\xd7\xd8\xd9\xda\xdb\xdc\xdd\xde\xdf\xe0\xe1\xe2\xe3\xe4\xe5\xe6\xe7\xe8"
+					. "\xe9\xea\xeb\xec\xed\xee\xef\xf0\xf1\xf2\xf3\xf4\xf5\xf6\xf8\xf9\xfa\xfb\xfc\xfd\xfe"
+					. "\x96\xa0\x8b\x97\x9b\xa6\xad\xb7",
+					'ALLSSSSTZZZallssstzzzRAAAALCCCEEEEIIDDNNOOOOxRUUUUYTsraaaalccceeeeiiddnnooooruuuuyt- <->|-.');
+				$s = self::pcre('preg_replace', ['#[^\x00-\x7F]++#', '', $s]);
+			} else {
+				$s = iconv('UTF-8', 'ASCII//TRANSLIT//IGNORE', $s);
+			}
+			// remove garbage that iconv creates during transliteration (eg Ý -> Y')
+			$s = str_replace(['`', "'", '"', '^', '~', '?'], '', $s);
+			// restore temporarily hidden characters
+			$s = strtr($s, "\x01\x02\x03\x04\x05\x06", '`\'"^~?');
 		}
-		// remove garbage that iconv creates during transliteration (eg Ý -> Y')
-		$s = str_replace(['`', "'", '"', '^', '~', '?'], '', $s);
-		// restore temporarily hidden characters
-		$s = strtr($s, "\x01\x02\x03\x04\x05\x06", '`\'"^~?');
+
 		return $s;
 	}