| 1 |
<?php |
| 2 |
|
| 3 |
/* |
| 4 |
* This file is part of the Symfony package. |
| 5 |
* |
| 6 |
* (c) Fabien Potencier <fabien@symfony.com> |
| 7 |
* |
| 8 |
* For the full copyright and license information, please view the LICENSE |
| 9 |
* file that was distributed with this source code. |
| 10 |
*/ |
| 11 |
|
| 12 |
namespace Symfony\Component\String; |
| 13 |
|
| 14 |
use Symfony\Component\String\Exception\ExceptionInterface; |
| 15 |
use Symfony\Component\String\Exception\InvalidArgumentException; |
| 16 |
use Symfony\Component\String\Exception\RuntimeException; |
| 17 |
|
| 18 |
/** |
| 19 |
* Represents a string of abstract Unicode characters. |
| 20 |
* |
| 21 |
* Unicode defines 3 types of "characters" (bytes, code points and grapheme clusters). |
| 22 |
* This class is the abstract type to use as a type-hint when the logic you want to |
| 23 |
* implement is Unicode-aware but doesn't care about code points vs grapheme clusters. |
| 24 |
* |
| 25 |
* @author Nicolas Grekas <p@tchwork.com> |
| 26 |
* |
| 27 |
* @throws ExceptionInterface |
| 28 |
*/ |
| 29 |
abstract class AbstractUnicodeString extends AbstractString |
| 30 |
{ |
| 31 |
public const NFC = \Normalizer::NFC; |
| 32 |
public const NFD = \Normalizer::NFD; |
| 33 |
public const NFKC = \Normalizer::NFKC; |
| 34 |
public const NFKD = \Normalizer::NFKD; |
| 35 |
|
| 36 |
// all ASCII letters sorted by typical frequency of occurrence |
| 37 |
private const ASCII = "\x20\x65\x69\x61\x73\x6E\x74\x72\x6F\x6C\x75\x64\x5D\x5B\x63\x6D\x70\x27\x0A\x67\x7C\x68\x76\x2E\x66\x62\x2C\x3A\x3D\x2D\x71\x31\x30\x43\x32\x2A\x79\x78\x29\x28\x4C\x39\x41\x53\x2F\x50\x22\x45\x6A\x4D\x49\x6B\x33\x3E\x35\x54\x3C\x44\x34\x7D\x42\x7B\x38\x46\x77\x52\x36\x37\x55\x47\x4E\x3B\x4A\x7A\x56\x23\x48\x4F\x57\x5F\x26\x21\x4B\x3F\x58\x51\x25\x59\x5C\x09\x5A\x2B\x7E\x5E\x24\x40\x60\x7F\x00\x01\x02\x03\x04\x05\x06\x07\x08\x0B\x0C\x0D\x0E\x0F\x10\x11\x12\x13\x14\x15\x16\x17\x18\x19\x1A\x1B\x1C\x1D\x1E\x1F"; |
| 38 |
|
| 39 |
// the subset of folded case mappings that is not in lower case mappings |
| 40 |
private const FOLD_FROM = ['İ', 'µ', 'ſ', "\xCD\x85", 'ς', 'ϐ', 'ϑ', 'ϕ', 'ϖ', 'ϰ', 'ϱ', 'ϵ', 'ẛ', "\xE1\xBE\xBE", 'ß', 'İ', 'ʼn', 'ǰ', 'ΐ', 'ΰ', 'և', 'ẖ', 'ẗ', 'ẘ', 'ẙ', 'ẚ', 'ẞ', 'ὐ', 'ὒ', 'ὔ', 'ὖ', 'ᾀ', 'ᾁ', 'ᾂ', 'ᾃ', 'ᾄ', '� |
| 41 |
', 'ᾆ', 'ᾇ', 'ᾈ', 'ᾉ', 'ᾊ', 'ᾋ', 'ᾌ', 'ᾍ', 'ᾎ', 'ᾏ', 'ᾐ', 'ᾑ', 'ᾒ', 'ᾓ', 'ᾔ', 'ᾕ', 'ᾖ', 'ᾗ', 'ᾘ', 'ᾙ', 'ᾚ', 'ᾛ', 'ᾜ', 'ᾝ', 'ᾞ', 'ᾟ', 'ᾠ', 'ᾡ', 'ᾢ', 'ᾣ', 'ᾤ', 'ᾥ', 'ᾦ', 'ᾧ', 'ᾨ', 'ᾩ', 'ᾪ', 'ᾫ', 'ᾬ', 'ᾭ', 'ᾮ', 'ᾯ', 'ᾲ', 'ᾳ', 'ᾴ', 'ᾶ', 'ᾷ', 'ᾼ', 'ῂ', 'ῃ', 'ῄ', 'ῆ', 'ῇ', 'ῌ', 'ῒ', 'ΐ', 'ῖ', 'ῗ', 'ῢ', 'ΰ', 'ῤ', 'ῦ', 'ῧ', 'ῲ', 'ῳ', 'ῴ', 'ῶ', 'ῷ', 'ῼ', 'ff', 'fi', 'fl', 'ffi', 'ffl', '� |
| 42 |
', 'st', 'ﬓ', 'ﬔ', 'ﬕ', 'ﬖ', 'ﬗ']; |
| 43 |
private const FOLD_TO = ['i̇', 'μ', 's', 'ι', 'σ', 'β', 'θ', 'φ', 'π', 'κ', 'ρ', 'ε', 'ṡ', 'ι', 'ss', 'i̇', 'ʼn', 'ǰ', 'ΐ', 'ΰ', 'եւ', 'ẖ', 'ẗ', 'ẘ', 'ẙ', 'aʾ', 'ss', 'ὐ', 'ὒ', 'ὔ', 'ὖ', 'ἀι', 'ἁι', 'ἂι', 'ἃι', 'ἄι', '� |
| 44 |
ι', 'ἆι', 'ἇι', 'ἀι', 'ἁι', 'ἂι', 'ἃι', 'ἄι', '� |
| 45 |
ι', 'ἆι', 'ἇι', 'ἠι', 'ἡι', 'ἢι', 'ἣι', 'ἤι', 'ἥι', 'ἦι', 'ἧι', 'ἠι', 'ἡι', 'ἢι', 'ἣι', 'ἤι', 'ἥι', 'ἦι', 'ἧι', 'ὠι', 'ὡι', 'ὢι', 'ὣι', 'ὤι', 'ὥι', 'ὦι', 'ὧι', 'ὠι', 'ὡι', 'ὢι', 'ὣι', 'ὤι', 'ὥι', 'ὦι', 'ὧι', 'ὰι', 'αι', 'άι', 'ᾶ', 'ᾶι', 'αι', 'ὴι', 'ηι', 'ήι', 'ῆ', 'ῆι', 'ηι', 'ῒ', 'ΐ', 'ῖ', 'ῗ', 'ῢ', 'ΰ', 'ῤ', 'ῦ', 'ῧ', 'ὼι', 'ωι', 'ώι', 'ῶ', 'ῶι', 'ωι', 'ff', 'fi', 'fl', 'ffi', 'ffl', 'st', 'st', 'մն', 'մե', 'մի', 'վն', 'մխ']; |
| 46 |
|
| 47 |
// the subset of upper case mappings that map one code point to many code points |
| 48 |
private const UPPER_FROM = ['ß', 'ff', 'fi', 'fl', 'ffi', 'ffl', '� |
| 49 |
', 'st', 'և', 'ﬓ', 'ﬔ', 'ﬕ', 'ﬖ', 'ﬗ', 'ʼn', 'ΐ', 'ΰ', 'ǰ', 'ẖ', 'ẗ', 'ẘ', 'ẙ', 'ẚ', 'ὐ', 'ὒ', 'ὔ', 'ὖ', 'ᾶ', 'ῆ', 'ῒ', 'ΐ', 'ῖ', 'ῗ', 'ῢ', 'ΰ', 'ῤ', 'ῦ', 'ῧ', 'ῶ']; |
| 50 |
private const UPPER_TO = ['SS', 'FF', 'FI', 'FL', 'FFI', 'FFL', 'ST', 'ST', 'ԵՒ', 'ՄՆ', 'ՄԵ', 'ՄԻ', 'ՎՆ', 'ՄԽ', 'ʼN', 'Ϊ́', 'Ϋ́', 'J̌', 'H̱', 'T̈', 'W̊', 'Y̊', 'Aʾ', 'Υ̓', 'Υ̓̀', 'Υ̓́', 'Υ̓͂', 'Α͂', 'Η͂', 'Ϊ̀', 'Ϊ́', 'Ι͂', 'Ϊ͂', 'Ϋ̀', 'Ϋ́', 'Ρ̓', 'Υ͂', 'Ϋ͂', 'Ω͂']; |
| 51 |
|
| 52 |
// the subset of https://github.com/unicode-org/cldr/blob/master/common/transforms/Latin-ASCII.xml that is not in NFKD |
| 53 |
private const TRANSLIT_FROM = ['Æ', 'Ð', 'Ø', 'Þ', 'ß', 'æ', 'ð', 'ø', 'þ', 'Đ', 'đ', 'Ħ', 'ħ', 'ı', 'ĸ', 'Ŀ', 'ŀ', 'Ł', 'ł', 'ʼn', 'Ŋ', 'ŋ', 'Œ', 'œ', 'Ŧ', 'ŧ', 'ƀ', 'Ɓ', 'Ƃ', 'ƃ', 'Ƈ', 'ƈ', 'Ɖ', 'Ɗ', 'Ƌ', 'ƌ', 'Ɛ', 'Ƒ', 'ƒ', 'Ɠ', 'ƕ', 'Ɩ', 'Ɨ', 'Ƙ', 'ƙ', 'ƚ', 'Ɲ', 'ƞ', 'Ƣ', 'ƣ', 'Ƥ', 'ƥ', 'ƫ', 'Ƭ', 'ƭ', 'Ʈ', 'Ʋ', 'Ƴ', 'ƴ', 'Ƶ', 'ƶ', 'DŽ', '� |
| 54 |
', 'dž', 'Ǥ', 'ǥ', 'ȡ', 'Ȥ', 'ȥ', 'ȴ', 'ȵ', 'ȶ', 'ȷ', 'ȸ', 'ȹ', 'Ⱥ', 'Ȼ', 'ȼ', 'Ƚ', 'Ⱦ', 'ȿ', 'ɀ', 'Ƀ', 'Ʉ', 'Ɇ', 'ɇ', 'Ɉ', 'ɉ', 'Ɍ', 'ɍ', 'Ɏ', 'ɏ', 'ɓ', 'ɕ', 'ɖ', 'ɗ', 'ɛ', 'ɟ', 'ɠ', 'ɡ', 'ɢ', 'ɦ', 'ɧ', 'ɨ', 'ɪ', 'ɫ', 'ɬ', 'ɭ', 'ɱ', 'ɲ', 'ɳ', 'ɴ', 'ɶ', 'ɼ', 'ɽ', 'ɾ', 'ʀ', 'ʂ', 'ʈ', 'ʉ', 'ʋ', 'ʏ', 'ʐ', 'ʑ', 'ʙ', 'ʛ', 'ʜ', 'ʝ', 'ʟ', 'ʠ', 'ʣ', 'ʥ', 'ʦ', 'ʪ', 'ʫ', 'ᴀ', 'ᴁ', 'ᴃ', 'ᴄ', '� |
| 55 |
', 'ᴆ', 'ᴇ', 'ᴊ', 'ᴋ', 'ᴌ', 'ᴍ', 'ᴏ', 'ᴘ', 'ᴛ', 'ᴜ', 'ᴠ', 'ᴡ', 'ᴢ', 'ᵫ', 'ᵬ', 'ᵭ', 'ᵮ', 'ᵯ', 'ᵰ', 'ᵱ', 'ᵲ', 'ᵳ', 'ᵴ', 'ᵵ', 'ᵶ', 'ᵺ', 'ᵻ', 'ᵽ', 'ᵾ', 'ᶀ', 'ᶁ', 'ᶂ', 'ᶃ', 'ᶄ', '� |
| 56 |
', 'ᶆ', 'ᶇ', 'ᶈ', 'ᶉ', 'ᶊ', 'ᶌ', 'ᶍ', 'ᶎ', 'ᶏ', 'ᶑ', 'ᶒ', 'ᶓ', 'ᶖ', 'ᶙ', 'ẚ', 'ẜ', 'ẝ', 'ẞ', 'Ỻ', 'ỻ', 'Ỽ', 'ỽ', 'Ỿ', 'ỿ', '©', '®', '₠', '₢', '₣', '₤', '₧', '₺', '₹', 'ℌ', '℞', '㎧', '㎮', '㏆', '㏗', '㏞', '㏟', '¼', '½', '¾', '� |
| 57 |
�', '� |
| 58 |
�', '� |
| 59 |
�', '� |
| 60 |
�', '� |
| 61 |
�', '� |
| 62 |
�', '� |
| 63 |
�', '� |
| 64 |
�', '� |
| 65 |
�', '� |
| 66 |
�', '� |
| 67 |
�', '� |
| 68 |
�', '� |
| 69 |
�', '〇', '‘', '’', '‚', '‛', '“', '”', '„', '‟', '′', '″', '〝', '〞', '«', '»', '‹', '›', '‐', '‑', '‒', '–', '—', '―', '︱', '︲', '﹘', '‖', '⁄', '� |
| 70 |
', '⁆', '⁎', '、', '。', '〈', '〉', '《', '》', '〔', '〕', '〘', '〙', '〚', '〛', '︑', '︒', '︹', '︺', '︽', '︾', '︿', '﹀', '﹑', '﹝', '﹞', '｟', '｠', '｡', '､', '×', '÷', '−', '∕', '∖', '∣', '∥', '≪', '≫', '� |
| 71 |
', '⦆']; |
| 72 |
private const TRANSLIT_TO = ['AE', 'D', 'O', 'TH', 'ss', 'ae', 'd', 'o', 'th', 'D', 'd', 'H', 'h', 'i', 'q', 'L', 'l', 'L', 'l', '\'n', 'N', 'n', 'OE', 'oe', 'T', 't', 'b', 'B', 'B', 'b', 'C', 'c', 'D', 'D', 'D', 'd', 'E', 'F', 'f', 'G', 'hv', 'I', 'I', 'K', 'k', 'l', 'N', 'n', 'OI', 'oi', 'P', 'p', 't', 'T', 't', 'T', 'V', 'Y', 'y', 'Z', 'z', 'DZ', 'Dz', 'dz', 'G', 'g', 'd', 'Z', 'z', 'l', 'n', 't', 'j', 'db', 'qp', 'A', 'C', 'c', 'L', 'T', 's', 'z', 'B', 'U', 'E', 'e', 'J', 'j', 'R', 'r', 'Y', 'y', 'b', 'c', 'd', 'd', 'e', 'j', 'g', 'g', 'G', 'h', 'h', 'i', 'I', 'l', 'l', 'l', 'm', 'n', 'n', 'N', 'OE', 'r', 'r', 'r', 'R', 's', 't', 'u', 'v', 'Y', 'z', 'z', 'B', 'G', 'H', 'j', 'L', 'q', 'dz', 'dz', 'ts', 'ls', 'lz', 'A', 'AE', 'B', 'C', 'D', 'D', 'E', 'J', 'K', 'L', 'M', 'O', 'P', 'T', 'U', 'V', 'W', 'Z', 'ue', 'b', 'd', 'f', 'm', 'n', 'p', 'r', 'r', 's', 't', 'z', 'th', 'I', 'p', 'U', 'b', 'd', 'f', 'g', 'k', 'l', 'm', 'n', 'p', 'r', 's', 'v', 'x', 'z', 'a', 'd', 'e', 'e', 'i', 'u', 'a', 's', 's', 'SS', 'LL', 'll', 'V', 'v', 'Y', 'y', '(C)', '(R)', 'CE', 'Cr', 'Fr.', 'L.', 'Pts', 'TL', 'Rs', 'x', 'Rx', 'm/s', 'rad/s', 'C/kg', 'pH', 'V/m', 'A/m', ' 1/4', ' 1/2', ' 3/4', ' 1/3', ' 2/3', ' 1/5', ' 2/5', ' 3/5', ' 4/5', ' 1/6', ' 5/6', ' 1/8', ' 3/8', ' 5/8', ' 7/8', ' 1/', '0', '\'', '\'', ',', '\'', '"', '"', ',,', '"', '\'', '"', '"', '"', '<<', '>>', '<', '>', '-', '-', '-', '-', '-', '-', '-', '-', '-', '||', '/', '[', ']', '*', ',', '.', '<', '>', '<<', '>>', '[', ']', '[', ']', '[', ']', ',', '.', '[', ']', '<<', '>>', '<', '>', ',', '[', ']', '((', '))', '.', ',', '*', '/', '-', '/', '\\', '|', '||', '<<', '>>', '((', '))']; |
| 73 |
|
| 74 |
private static $transliterators = []; |
| 75 |
|
| 76 |
/** |
| 77 |
* @return static |
| 78 |
*/ |
| 79 |
public static function fromCodePoints(int ...$codes): self |
| 80 |
{ |
| 81 |
$string = ''; |
| 82 |
|
| 83 |
foreach ($codes as $code) { |
| 84 |
if (0x80 > $code %= 0x200000) { |
| 85 |
$string .= \chr($code); |
| 86 |
} elseif (0x800 > $code) { |
| 87 |
$string .= \chr(0xC0 | $code >> 6).\chr(0x80 | $code & 0x3F); |
| 88 |
} elseif (0x10000 > $code) { |
| 89 |
$string .= \chr(0xE0 | $code >> 12).\chr(0x80 | $code >> 6 & 0x3F).\chr(0x80 | $code & 0x3F); |
| 90 |
} else { |
| 91 |
$string .= \chr(0xF0 | $code >> 18).\chr(0x80 | $code >> 12 & 0x3F).\chr(0x80 | $code >> 6 & 0x3F).\chr(0x80 | $code & 0x3F); |
| 92 |
} |
| 93 |
} |
| 94 |
|
| 95 |
return new static($string); |
| 96 |
} |
| 97 |
|
| 98 |
/** |
| 99 |
* Generic UTF-8 to ASCII transliteration. |
| 100 |
* |
| 101 |
* Install the intl extension for best results. |
| 102 |
* |
| 103 |
* @param string[]|\Transliterator[] $rules See "*-Latin" rules from Transliterator::listIDs() |
| 104 |
*/ |
| 105 |
public function ascii(array $rules = []): self |
| 106 |
{ |
| 107 |
$str = clone $this; |
| 108 |
$s = $str->string; |
| 109 |
$str->string = ''; |
| 110 |
|
| 111 |
array_unshift($rules, 'nfd'); |
| 112 |
$rules[] = 'latin-ascii'; |
| 113 |
|
| 114 |
if (\function_exists('transliterator_transliterate')) { |
| 115 |
$rules[] = 'any-latin/bgn'; |
| 116 |
} |
| 117 |
|
| 118 |
$rules[] = 'nfkd'; |
| 119 |
$rules[] = '[:nonspacing mark:] remove'; |
| 120 |
|
| 121 |
while (\strlen($s) - 1 > $i = strspn($s, self::ASCII)) { |
| 122 |
if (0 < --$i) { |
| 123 |
$str->string .= substr($s, 0, $i); |
| 124 |
$s = substr($s, $i); |
| 125 |
} |
| 126 |
|
| 127 |
if (!$rule = array_shift($rules)) { |
| 128 |
$rules = []; // An empty rule interrupts the next ones |
| 129 |
} |
| 130 |
|
| 131 |
if ($rule instanceof \Transliterator) { |
| 132 |
$s = $rule->transliterate($s); |
| 133 |
} elseif ($rule) { |
| 134 |
if ('nfd' === $rule = strtolower($rule)) { |
| 135 |
normalizer_is_normalized($s, self::NFD) ?: $s = normalizer_normalize($s, self::NFD); |
| 136 |
} elseif ('nfkd' === $rule) { |
| 137 |
normalizer_is_normalized($s, self::NFKD) ?: $s = normalizer_normalize($s, self::NFKD); |
| 138 |
} elseif ('[:nonspacing mark:] remove' === $rule) { |
| 139 |
$s = preg_replace('/\p{Mn}++/u', '', $s); |
| 140 |
} elseif ('latin-ascii' === $rule) { |
| 141 |
$s = str_replace(self::TRANSLIT_FROM, self::TRANSLIT_TO, $s); |
| 142 |
} elseif ('de-ascii' === $rule) { |
| 143 |
$s = preg_replace("/([AUO])\u{0308}(?=\p{Ll})/u", '$1e', $s); |
| 144 |
$s = str_replace(["a\u{0308}", "o\u{0308}", "u\u{0308}", "A\u{0308}", "O\u{0308}", "U\u{0308}"], ['ae', 'oe', 'ue', 'AE', 'OE', 'UE'], $s); |
| 145 |
} elseif (\function_exists('transliterator_transliterate')) { |
| 146 |
if (null === $transliterator = self::$transliterators[$rule] ?? self::$transliterators[$rule] = \Transliterator::create($rule)) { |
| 147 |
if ('any-latin/bgn' === $rule) { |
| 148 |
$rule = 'any-latin'; |
| 149 |
$transliterator = self::$transliterators[$rule] ?? self::$transliterators[$rule] = \Transliterator::create($rule); |
| 150 |
} |
| 151 |
|
| 152 |
if (null === $transliterator) { |
| 153 |
throw new InvalidArgumentException(sprintf('Unknown transliteration rule "%s".', $rule)); |
| 154 |
} |
| 155 |
|
| 156 |
self::$transliterators['any-latin/bgn'] = $transliterator; |
| 157 |
} |
| 158 |
|
| 159 |
$s = $transliterator->transliterate($s); |
| 160 |
} |
| 161 |
} elseif (!\function_exists('iconv')) { |
| 162 |
$s = preg_replace('/[^\x00-\x7F]/u', '?', $s); |
| 163 |
} elseif (ICONV_IMPL === 'glibc') { |
| 164 |
$s = iconv('UTF-8', 'ASCII//TRANSLIT', $s); |
| 165 |
} else { |
| 166 |
$s = @preg_replace_callback('/[^\x00-\x7F]/u', static function ($c) { |
| 167 |
if ('' === $c = (string) iconv('UTF-8', 'ASCII//IGNORE//TRANSLIT', $c[0])) { |
| 168 |
throw new \LogicException(sprintf('"%s" requires a translit-able iconv implementation, try installing "gnu-libiconv" if you\'re using Alpine Linux.', static::class)); |
| 169 |
} |
| 170 |
|
| 171 |
return 1 < \strlen($c) ? ltrim($c, '\'`"^~') : (\strlen($c) ? $c : '?'); |
| 172 |
}, $s); |
| 173 |
} |
| 174 |
} |
| 175 |
|
| 176 |
$str->string .= $s; |
| 177 |
|
| 178 |
return $str; |
| 179 |
} |
| 180 |
|
| 181 |
public function camel(): parent |
| 182 |
{ |
| 183 |
$str = clone $this; |
| 184 |
$str->string = str_replace(' ', '', preg_replace_callback('/\b./u', static function ($m) use (&$i) { |
| 185 |
return 1 === ++$i ? ('İ' === $m[0] ? 'i̇' : mb_strtolower($m[0], 'UTF-8')) : mb_convert_case($m[0], MB_CASE_TITLE, 'UTF-8'); |
| 186 |
}, preg_replace('/[^\pL0-9]++/u', ' ', $this->string))); |
| 187 |
|
| 188 |
return $str; |
| 189 |
} |
| 190 |
|
| 191 |
/** |
| 192 |
* @return int[] |
| 193 |
*/ |
| 194 |
public function codePointsAt(int $offset): array |
| 195 |
{ |
| 196 |
$str = $this->slice($offset, 1); |
| 197 |
|
| 198 |
if ('' === $str->string) { |
| 199 |
return []; |
| 200 |
} |
| 201 |
|
| 202 |
$codePoints = []; |
| 203 |
|
| 204 |
foreach (preg_split('//u', $str->string, -1, PREG_SPLIT_NO_EMPTY) as $c) { |
| 205 |
$codePoints[] = mb_ord($c, 'UTF-8'); |
| 206 |
} |
| 207 |
|
| 208 |
return $codePoints; |
| 209 |
} |
| 210 |
|
| 211 |
public function folded(bool $compat = true): parent |
| 212 |
{ |
| 213 |
$str = clone $this; |
| 214 |
|
| 215 |
if (!$compat || \PHP_VERSION_ID < 70300 || !\defined('Normalizer::NFKC_CF')) { |
| 216 |
$str->string = normalizer_normalize($str->string, $compat ? \Normalizer::NFKC : \Normalizer::NFC); |
| 217 |
$str->string = mb_strtolower(str_replace(self::FOLD_FROM, self::FOLD_TO, $this->string), 'UTF-8'); |
| 218 |
} else { |
| 219 |
$str->string = normalizer_normalize($str->string, \Normalizer::NFKC_CF); |
| 220 |
} |
| 221 |
|
| 222 |
return $str; |
| 223 |
} |
| 224 |
|
| 225 |
public function join(array $strings, string $lastGlue = null): parent |
| 226 |
{ |
| 227 |
$str = clone $this; |
| 228 |
|
| 229 |
$tail = null !== $lastGlue && 1 < \count($strings) ? $lastGlue.array_pop($strings) : ''; |
| 230 |
$str->string = implode($this->string, $strings).$tail; |
| 231 |
|
| 232 |
if (!preg_match('//u', $str->string)) { |
| 233 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 234 |
} |
| 235 |
|
| 236 |
return $str; |
| 237 |
} |
| 238 |
|
| 239 |
public function lower(): parent |
| 240 |
{ |
| 241 |
$str = clone $this; |
| 242 |
$str->string = mb_strtolower(str_replace('İ', 'i̇', $str->string), 'UTF-8'); |
| 243 |
|
| 244 |
return $str; |
| 245 |
} |
| 246 |
|
| 247 |
public function match(string $regexp, int $flags = 0, int $offset = 0): array |
| 248 |
{ |
| 249 |
$match = ((PREG_PATTERN_ORDER | PREG_SET_ORDER) & $flags) ? 'preg_match_all' : 'preg_match'; |
| 250 |
|
| 251 |
if ($this->ignoreCase) { |
| 252 |
$regexp .= 'i'; |
| 253 |
} |
| 254 |
|
| 255 |
set_error_handler(static function ($t, $m) { throw new InvalidArgumentException($m); }); |
| 256 |
|
| 257 |
try { |
| 258 |
if (false === $match($regexp.'u', $this->string, $matches, $flags | PREG_UNMATCHED_AS_NULL, $offset)) { |
| 259 |
$lastError = preg_last_error(); |
| 260 |
|
| 261 |
foreach (get_defined_constants(true)['pcre'] as $k => $v) { |
| 262 |
if ($lastError === $v && '_ERROR' === substr($k, -6)) { |
| 263 |
throw new RuntimeException('Matching failed with '.$k.'.'); |
| 264 |
} |
| 265 |
} |
| 266 |
|
| 267 |
throw new RuntimeException('Matching failed with unknown error code.'); |
| 268 |
} |
| 269 |
} finally { |
| 270 |
restore_error_handler(); |
| 271 |
} |
| 272 |
|
| 273 |
return $matches; |
| 274 |
} |
| 275 |
|
| 276 |
/** |
| 277 |
* @return static |
| 278 |
*/ |
| 279 |
public function normalize(int $form = self::NFC): self |
| 280 |
{ |
| 281 |
if (!\in_array($form, [self::NFC, self::NFD, self::NFKC, self::NFKD])) { |
| 282 |
throw new InvalidArgumentException('Unsupported normalization form.'); |
| 283 |
} |
| 284 |
|
| 285 |
$str = clone $this; |
| 286 |
normalizer_is_normalized($str->string, $form) ?: $str->string = normalizer_normalize($str->string, $form); |
| 287 |
|
| 288 |
return $str; |
| 289 |
} |
| 290 |
|
| 291 |
public function padBoth(int $length, string $padStr = ' '): parent |
| 292 |
{ |
| 293 |
if ('' === $padStr || !preg_match('//u', $padStr)) { |
| 294 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 295 |
} |
| 296 |
|
| 297 |
$pad = clone $this; |
| 298 |
$pad->string = $padStr; |
| 299 |
|
| 300 |
return $this->pad($length, $pad, STR_PAD_BOTH); |
| 301 |
} |
| 302 |
|
| 303 |
public function padEnd(int $length, string $padStr = ' '): parent |
| 304 |
{ |
| 305 |
if ('' === $padStr || !preg_match('//u', $padStr)) { |
| 306 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 307 |
} |
| 308 |
|
| 309 |
$pad = clone $this; |
| 310 |
$pad->string = $padStr; |
| 311 |
|
| 312 |
return $this->pad($length, $pad, STR_PAD_RIGHT); |
| 313 |
} |
| 314 |
|
| 315 |
public function padStart(int $length, string $padStr = ' '): parent |
| 316 |
{ |
| 317 |
if ('' === $padStr || !preg_match('//u', $padStr)) { |
| 318 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 319 |
} |
| 320 |
|
| 321 |
$pad = clone $this; |
| 322 |
$pad->string = $padStr; |
| 323 |
|
| 324 |
return $this->pad($length, $pad, STR_PAD_LEFT); |
| 325 |
} |
| 326 |
|
| 327 |
public function replaceMatches(string $fromRegexp, $to): parent |
| 328 |
{ |
| 329 |
if ($this->ignoreCase) { |
| 330 |
$fromRegexp .= 'i'; |
| 331 |
} |
| 332 |
|
| 333 |
if (\is_array($to) || $to instanceof \Closure) { |
| 334 |
if (!\is_callable($to)) { |
| 335 |
throw new \TypeError(sprintf('Argument 2 passed to "%s::replaceMatches()" must be callable, array given.', static::class)); |
| 336 |
} |
| 337 |
|
| 338 |
$replace = 'preg_replace_callback'; |
| 339 |
$to = static function (array $m) use ($to): string { |
| 340 |
$to = $to($m); |
| 341 |
|
| 342 |
if ('' !== $to && (!\is_string($to) || !preg_match('//u', $to))) { |
| 343 |
throw new InvalidArgumentException('Replace callback must return a valid UTF-8 string.'); |
| 344 |
} |
| 345 |
|
| 346 |
return $to; |
| 347 |
}; |
| 348 |
} elseif ('' !== $to && !preg_match('//u', $to)) { |
| 349 |
throw new InvalidArgumentException('Invalid UTF-8 string.'); |
| 350 |
} else { |
| 351 |
$replace = 'preg_replace'; |
| 352 |
} |
| 353 |
|
| 354 |
set_error_handler(static function ($t, $m) { throw new InvalidArgumentException($m); }); |
| 355 |
|
| 356 |
try { |
| 357 |
if (null === $string = $replace($fromRegexp.'u', $to, $this->string)) { |
| 358 |
$lastError = preg_last_error(); |
| 359 |
|
| 360 |
foreach (get_defined_constants(true)['pcre'] as $k => $v) { |
| 361 |
if ($lastError === $v && '_ERROR' === substr($k, -6)) { |
| 362 |
throw new RuntimeException('Matching failed with '.$k.'.'); |
| 363 |
} |
| 364 |
} |
| 365 |
|
| 366 |
throw new RuntimeException('Matching failed with unknown error code.'); |
| 367 |
} |
| 368 |
} finally { |
| 369 |
restore_error_handler(); |
| 370 |
} |
| 371 |
|
| 372 |
$str = clone $this; |
| 373 |
$str->string = $string; |
| 374 |
|
| 375 |
return $str; |
| 376 |
} |
| 377 |
|
| 378 |
public function reverse(): parent |
| 379 |
{ |
| 380 |
$str = clone $this; |
| 381 |
$str->string = implode('', array_reverse(preg_split('/(\X)/u', $str->string, -1, PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_NO_EMPTY))); |
| 382 |
|
| 383 |
return $str; |
| 384 |
} |
| 385 |
|
| 386 |
public function snake(): parent |
| 387 |
{ |
| 388 |
$str = $this->camel()->title(); |
| 389 |
$str->string = mb_strtolower(preg_replace(['/(\p{Lu}+)(\p{Lu}\p{Ll})/u', '/([\p{Ll}0-9])(\p{Lu})/u'], '\1_\2', $str->string), 'UTF-8'); |
| 390 |
|
| 391 |
return $str; |
| 392 |
} |
| 393 |
|
| 394 |
public function title(bool $allWords = false): parent |
| 395 |
{ |
| 396 |
$str = clone $this; |
| 397 |
|
| 398 |
$limit = $allWords ? -1 : 1; |
| 399 |
|
| 400 |
$str->string = preg_replace_callback('/\b./u', static function (array $m): string { |
| 401 |
return mb_convert_case($m[0], MB_CASE_TITLE, 'UTF-8'); |
| 402 |
}, $str->string, $limit); |
| 403 |
|
| 404 |
return $str; |
| 405 |
} |
| 406 |
|
| 407 |
public function trim(string $chars = " \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}"): parent |
| 408 |
{ |
| 409 |
if (" \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}" !== $chars && !preg_match('//u', $chars)) { |
| 410 |
throw new InvalidArgumentException('Invalid UTF-8 chars.'); |
| 411 |
} |
| 412 |
$chars = preg_quote($chars); |
| 413 |
|
| 414 |
$str = clone $this; |
| 415 |
$str->string = preg_replace("{^[$chars]++|[$chars]++$}uD", '', $str->string); |
| 416 |
|
| 417 |
return $str; |
| 418 |
} |
| 419 |
|
| 420 |
public function trimEnd(string $chars = " \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}"): parent |
| 421 |
{ |
| 422 |
if (" \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}" !== $chars && !preg_match('//u', $chars)) { |
| 423 |
throw new InvalidArgumentException('Invalid UTF-8 chars.'); |
| 424 |
} |
| 425 |
$chars = preg_quote($chars); |
| 426 |
|
| 427 |
$str = clone $this; |
| 428 |
$str->string = preg_replace("{[$chars]++$}uD", '', $str->string); |
| 429 |
|
| 430 |
return $str; |
| 431 |
} |
| 432 |
|
| 433 |
public function trimStart(string $chars = " \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}"): parent |
| 434 |
{ |
| 435 |
if (" \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}" !== $chars && !preg_match('//u', $chars)) { |
| 436 |
throw new InvalidArgumentException('Invalid UTF-8 chars.'); |
| 437 |
} |
| 438 |
$chars = preg_quote($chars); |
| 439 |
|
| 440 |
$str = clone $this; |
| 441 |
$str->string = preg_replace("{^[$chars]++}uD", '', $str->string); |
| 442 |
|
| 443 |
return $str; |
| 444 |
} |
| 445 |
|
| 446 |
public function upper(): parent |
| 447 |
{ |
| 448 |
$str = clone $this; |
| 449 |
$str->string = mb_strtoupper($str->string, 'UTF-8'); |
| 450 |
|
| 451 |
if (\PHP_VERSION_ID < 70300) { |
| 452 |
$str->string = str_replace(self::UPPER_FROM, self::UPPER_TO, $str->string); |
| 453 |
} |
| 454 |
|
| 455 |
return $str; |
| 456 |
} |
| 457 |
|
| 458 |
public function width(bool $ignoreAnsiDecoration = true): int |
| 459 |
{ |
| 460 |
$width = 0; |
| 461 |
$s = str_replace(["\x00", "\x05", "\x07"], '', $this->string); |
| 462 |
|
| 463 |
if (false !== strpos($s, "\r")) { |
| 464 |
$s = str_replace(["\r\n", "\r"], "\n", $s); |
| 465 |
} |
| 466 |
|
| 467 |
if (!$ignoreAnsiDecoration) { |
| 468 |
$s = preg_replace('/[\p{Cc}\x7F]++/u', '', $s); |
| 469 |
} |
| 470 |
|
| 471 |
foreach (explode("\n", $s) as $s) { |
| 472 |
if ($ignoreAnsiDecoration) { |
| 473 |
$s = preg_replace('/(?:\x1B(?: |
| 474 |
\[ [\x30-\x3F]*+ [\x20-\x2F]*+ [0x40-\x7E] |
| 475 |
| [P\]X^_] .*? \x1B\\\\ |
| 476 |
| [\x41-\x7E] |
| 477 |
)|[\p{Cc}\x7F]++)/xu', '', $s); |
| 478 |
} |
| 479 |
|
| 480 |
// Non printable characters have been dropped, so wcswidth cannot logically return -1. |
| 481 |
$width += $this->wcswidth($s); |
| 482 |
} |
| 483 |
|
| 484 |
return $width; |
| 485 |
} |
| 486 |
|
| 487 |
/** |
| 488 |
* @return static |
| 489 |
*/ |
| 490 |
private function pad(int $len, self $pad, int $type): parent |
| 491 |
{ |
| 492 |
$sLen = $this->length(); |
| 493 |
|
| 494 |
if ($len <= $sLen) { |
| 495 |
return clone $this; |
| 496 |
} |
| 497 |
|
| 498 |
$padLen = $pad->length(); |
| 499 |
$freeLen = $len - $sLen; |
| 500 |
$len = $freeLen % $padLen; |
| 501 |
|
| 502 |
switch ($type) { |
| 503 |
case STR_PAD_RIGHT: |
| 504 |
return $this->append(str_repeat($pad->string, $freeLen / $padLen).($len ? $pad->slice(0, $len) : '')); |
| 505 |
|
| 506 |
case STR_PAD_LEFT: |
| 507 |
return $this->prepend(str_repeat($pad->string, $freeLen / $padLen).($len ? $pad->slice(0, $len) : '')); |
| 508 |
|
| 509 |
case STR_PAD_BOTH: |
| 510 |
$freeLen /= 2; |
| 511 |
|
| 512 |
$rightLen = ceil($freeLen); |
| 513 |
$len = $rightLen % $padLen; |
| 514 |
$str = $this->append(str_repeat($pad->string, $rightLen / $padLen).($len ? $pad->slice(0, $len) : '')); |
| 515 |
|
| 516 |
$leftLen = floor($freeLen); |
| 517 |
$len = $leftLen % $padLen; |
| 518 |
|
| 519 |
return $str->prepend(str_repeat($pad->string, $leftLen / $padLen).($len ? $pad->slice(0, $len) : '')); |
| 520 |
|
| 521 |
default: |
| 522 |
throw new InvalidArgumentException('Invalid padding type.'); |
| 523 |
} |
| 524 |
} |
| 525 |
|
| 526 |
/** |
| 527 |
* Based on https://github.com/jquast/wcwidth, a Python implementation of https://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c. |
| 528 |
*/ |
| 529 |
private function wcswidth(string $string): int |
| 530 |
{ |
| 531 |
$width = 0; |
| 532 |
|
| 533 |
foreach (preg_split('//u', $string, -1, PREG_SPLIT_NO_EMPTY) as $c) { |
| 534 |
$codePoint = mb_ord($c, 'UTF-8'); |
| 535 |
|
| 536 |
if (0 === $codePoint // NULL |
| 537 |
|| 0x034F === $codePoint // COMBINING GRAPHEME JOINER |
| 538 |
|| (0x200B <= $codePoint && 0x200F >= $codePoint) // ZERO WIDTH SPACE to RIGHT-TO-LEFT MARK |
| 539 |
|| 0x2028 === $codePoint // LINE SEPARATOR |
| 540 |
|| 0x2029 === $codePoint // PARAGRAPH SEPARATOR |
| 541 |
|| (0x202A <= $codePoint && 0x202E >= $codePoint) // LEFT-TO-RIGHT EMBEDDING to RIGHT-TO-LEFT OVERRIDE |
| 542 |
|| (0x2060 <= $codePoint && 0x2063 >= $codePoint) // WORD JOINER to INVISIBLE SEPARATOR |
| 543 |
) { |
| 544 |
continue; |
| 545 |
} |
| 546 |
|
| 547 |
// Non printable characters |
| 548 |
if (32 > $codePoint // C0 control characters |
| 549 |
|| (0x07F <= $codePoint && 0x0A0 > $codePoint) // C1 control characters and DEL |
| 550 |
) { |
| 551 |
return -1; |
| 552 |
} |
| 553 |
|
| 554 |
static $tableZero; |
| 555 |
if (null === $tableZero) { |
| 556 |
$tableZero = require __DIR__.'/Resources/data/wcswidth_table_zero.php'; |
| 557 |
} |
| 558 |
|
| 559 |
if ($codePoint >= $tableZero[0][0] && $codePoint <= $tableZero[$ubound = \count($tableZero) - 1][1]) { |
| 560 |
$lbound = 0; |
| 561 |
while ($ubound >= $lbound) { |
| 562 |
$mid = floor(($lbound + $ubound) / 2); |
| 563 |
|
| 564 |
if ($codePoint > $tableZero[$mid][1]) { |
| 565 |
$lbound = $mid + 1; |
| 566 |
} elseif ($codePoint < $tableZero[$mid][0]) { |
| 567 |
$ubound = $mid - 1; |
| 568 |
} else { |
| 569 |
continue 2; |
| 570 |
} |
| 571 |
} |
| 572 |
} |
| 573 |
|
| 574 |
static $tableWide; |
| 575 |
if (null === $tableWide) { |
| 576 |
$tableWide = require __DIR__.'/Resources/data/wcswidth_table_wide.php'; |
| 577 |
} |
| 578 |
|
| 579 |
if ($codePoint >= $tableWide[0][0] && $codePoint <= $tableWide[$ubound = \count($tableWide) - 1][1]) { |
| 580 |
$lbound = 0; |
| 581 |
while ($ubound >= $lbound) { |
| 582 |
$mid = floor(($lbound + $ubound) / 2); |
| 583 |
|
| 584 |
if ($codePoint > $tableWide[$mid][1]) { |
| 585 |
$lbound = $mid + 1; |
| 586 |
} elseif ($codePoint < $tableWide[$mid][0]) { |
| 587 |
$ubound = $mid - 1; |
| 588 |
} else { |
| 589 |
$width += 2; |
| 590 |
|
| 591 |
continue 2; |
| 592 |
} |
| 593 |
} |
| 594 |
} |
| 595 |
|
| 596 |
++$width; |
| 597 |
} |
| 598 |
|
| 599 |
return $width; |
| 600 |
} |
| 601 |
} |
| 602 |
|
