Code Coverage |
||||||||||
Lines |
Functions and Methods |
Classes and Traits |
||||||||
| Total | |
100.00% |
50 / 50 |
|
100.00% |
6 / 6 |
CRAP | |
100.00% |
1 / 1 |
| JavascriptEscape | |
100.00% |
50 / 50 |
|
100.00% |
6 / 6 |
32 | |
100.00% |
1 / 1 |
| decode | |
100.00% |
13 / 13 |
|
100.00% |
1 / 1 |
10 | |||
| unicode | |
100.00% |
25 / 25 |
|
100.00% |
1 / 1 |
13 | |||
| codepoint | |
100.00% |
3 / 3 |
|
100.00% |
1 / 1 |
2 | |||
| highSurrogate | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
2 | |||
| lowSurrogate | |
100.00% |
2 / 2 |
|
100.00% |
1 / 1 |
2 | |||
| hex | |
100.00% |
5 / 5 |
|
100.00% |
1 / 1 |
3 | |||
| 1 | <?php |
| 2 | |
| 3 | declare(strict_types=1); |
| 4 | |
| 5 | namespace Celema\Verba\Tool; |
| 6 | |
| 7 | /** |
| 8 | * Decodes escape sequences from JavaScript string literals. |
| 9 | * |
| 10 | * @internal |
| 11 | */ |
| 12 | final class JavascriptEscape |
| 13 | { |
| 14 | /** |
| 15 | * Decodes the escape at $offset and advances it to the final consumed byte. |
| 16 | */ |
| 17 | public static function decode(string $raw, int &$offset): string |
| 18 | { |
| 19 | $offset++; |
| 20 | $char = $raw[$offset] ?? ''; |
| 21 | |
| 22 | return match ($char) { |
| 23 | 'b' => "\b", |
| 24 | 'f' => "\f", |
| 25 | 'n' => "\n", |
| 26 | 'r' => "\r", |
| 27 | 't' => "\t", |
| 28 | 'v' => "\v", |
| 29 | 'u' => self::unicode($raw, $offset), |
| 30 | 'x' => self::hex($raw, $offset), |
| 31 | default => $char, |
| 32 | }; |
| 33 | } |
| 34 | |
| 35 | private static function unicode(string $raw, int &$offset): string |
| 36 | { |
| 37 | if (($raw[$offset + 1] ?? '') === '{') { |
| 38 | $end = strpos($raw, '}', $offset + 2); |
| 39 | |
| 40 | if ($end === false) { |
| 41 | return 'u'; |
| 42 | } |
| 43 | |
| 44 | $hex = substr($raw, $offset + 2, $end - $offset - 2); |
| 45 | |
| 46 | if ($hex === '' || !ctype_xdigit($hex)) { |
| 47 | return 'u'; |
| 48 | } |
| 49 | |
| 50 | $offset = $end; |
| 51 | |
| 52 | return self::codepoint($hex); |
| 53 | } |
| 54 | |
| 55 | $hex = substr($raw, $offset + 1, 4); |
| 56 | |
| 57 | if (strlen($hex) !== 4 || !ctype_xdigit($hex)) { |
| 58 | return 'u'; |
| 59 | } |
| 60 | |
| 61 | $escape = '\\u' . $hex; |
| 62 | $offset += 4; |
| 63 | |
| 64 | if (self::highSurrogate($hex)) { |
| 65 | $low = substr($raw, $offset + 3, 4); |
| 66 | |
| 67 | if ( |
| 68 | ($raw[$offset + 1] ?? '') === '\\' |
| 69 | && ($raw[$offset + 2] ?? '') === 'u' |
| 70 | && strlen($low) === 4 |
| 71 | && ctype_xdigit($low) |
| 72 | && self::lowSurrogate($low) |
| 73 | ) { |
| 74 | $escape .= '\\u' . $low; |
| 75 | $offset += 6; |
| 76 | } |
| 77 | } |
| 78 | |
| 79 | /** @var string|null $decoded */ |
| 80 | $decoded = json_decode('"' . $escape . '"'); |
| 81 | |
| 82 | return $decoded ?? ''; |
| 83 | } |
| 84 | |
| 85 | private static function codepoint(string $hex): string |
| 86 | { |
| 87 | // Braced escapes already hold a full codepoint; fixed-width escapes use JSON for surrogates. |
| 88 | $entity = '&#x' . $hex . ';'; |
| 89 | $decoded = html_entity_decode($entity, ENT_QUOTES | ENT_HTML5, 'UTF-8'); |
| 90 | |
| 91 | return $decoded === $entity ? '' : $decoded; |
| 92 | } |
| 93 | |
| 94 | private static function highSurrogate(string $hex): bool |
| 95 | { |
| 96 | $code = hexdec($hex); |
| 97 | |
| 98 | return $code >= 0xD800 && $code <= 0xDBFF; |
| 99 | } |
| 100 | |
| 101 | private static function lowSurrogate(string $hex): bool |
| 102 | { |
| 103 | $code = hexdec($hex); |
| 104 | |
| 105 | return $code >= 0xDC00 && $code <= 0xDFFF; |
| 106 | } |
| 107 | |
| 108 | private static function hex(string $raw, int &$offset): string |
| 109 | { |
| 110 | $hex = substr($raw, $offset + 1, 2); |
| 111 | |
| 112 | if (strlen($hex) !== 2 || !ctype_xdigit($hex)) { |
| 113 | return 'x'; |
| 114 | } |
| 115 | |
| 116 | $offset += 2; |
| 117 | |
| 118 | return chr(hexdec($hex)); |
| 119 | } |
| 120 | } |