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 12namespace Symfony\Component\String; 13 14use Symfony\Component\String\Exception\ExceptionInterface; 15use Symfony\Component\String\Exception\InvalidArgumentException; 16use 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 */ 29abstract 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', 'ᾅ', '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', 'ſt', 'st', 'ﬓ', 'ﬔ', 'ﬕ', 'ﬖ', 'ﬗ']; 41 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', 'ἅι', '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', 'st', 'st', 'մն', 'մե', 'մի', 'վն', 'մխ']; 42 43 // the subset of upper case mappings that map one code point to many code points 44 private const UPPER_FROM = ['ß', 'ff', 'fi', 'fl', 'ffi', 'ffl', 'ſt', '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']; 45 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']; 46 47 // the subset of https://github.com/unicode-org/cldr/blob/master/common/transforms/Latin-ASCII.xml that is not in NFKD 48 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', 'Dž', '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', 'ᴅ', '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', 'ᶅ', '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', '⅓', '⅔', '⅕', '⅖', '⅗', '⅘', '⅙', '⅚', '⅛', '⅜', '⅝', '⅞', '⅟', '〇', '‘', '’', '‚', '‛', '“', '”', '„', '‟', '′', '″', '〝', '〞', '«', '»', '‹', '›', '‐', '‑', '‒', '–', '—', '―', '︱', '︲', '﹘', '‖', '⁄', '⁅', '⁆', '⁎', '、', '。', '〈', '〉', '《', '》', '〔', '〕', '〘', '〙', '〚', '〛', '︑', '︒', '︹', '︺', '︽', '︾', '︿', '﹀', '﹑', '﹝', '﹞', '｟', '｠', '｡', '､', '×', '÷', '−', '∕', '∖', '∣', '∥', '≪', '≫', '⦅', '⦆']; 49 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', '\'', '\'', ',', '\'', '"', '"', ',,', '"', '\'', '"', '"', '"', '<<', '>>', '<', '>', '-', '-', '-', '-', '-', '-', '-', '-', '-', '||', '/', '[', ']', '*', ',', '.', '<', '>', '<<', '>>', '[', ']', '[', ']', '[', ']', ',', '.', '[', ']', '<<', '>>', '<', '>', ',', '[', ']', '((', '))', '.', ',', '*', '/', '-', '/', '\\', '|', '||', '<<', '>>', '((', '))']; 50 51 private static $transliterators = []; 52 53 /** 54 * @return static 55 */ 56 public static function fromCodePoints(int ...$codes): self 57 { 58 $string = ''; 59 60 foreach ($codes as $code) { 61 if (0x80 > $code %= 0x200000) { 62 $string .= \chr($code); 63 } elseif (0x800 > $code) { 64 $string .= \chr(0xC0 | $code >> 6).\chr(0x80 | $code & 0x3F); 65 } elseif (0x10000 > $code) { 66 $string .= \chr(0xE0 | $code >> 12).\chr(0x80 | $code >> 6 & 0x3F).\chr(0x80 | $code & 0x3F); 67 } else { 68 $string .= \chr(0xF0 | $code >> 18).\chr(0x80 | $code >> 12 & 0x3F).\chr(0x80 | $code >> 6 & 0x3F).\chr(0x80 | $code & 0x3F); 69 } 70 } 71 72 return new static($string); 73 } 74 75 /** 76 * Generic UTF-8 to ASCII transliteration. 77 * 78 * Install the intl extension for best results. 79 * 80 * @param string[]|\Transliterator[]|\Closure[] $rules See "*-Latin" rules from Transliterator::listIDs() 81 */ 82 public function ascii(array $rules = []): self 83 { 84 $str = clone $this; 85 $s = $str->string; 86 $str->string = ''; 87 88 array_unshift($rules, 'nfd'); 89 $rules[] = 'latin-ascii'; 90 91 if (\function_exists('transliterator_transliterate')) { 92 $rules[] = 'any-latin/bgn'; 93 } 94 95 $rules[] = 'nfkd'; 96 $rules[] = '[:nonspacing mark:] remove'; 97 98 while (\strlen($s) - 1 > $i = strspn($s, self::ASCII)) { 99 if (0 < --$i) { 100 $str->string .= substr($s, 0, $i); 101 $s = substr($s, $i); 102 } 103 104 if (!$rule = array_shift($rules)) { 105 $rules = []; // An empty rule interrupts the next ones 106 } 107 108 if ($rule instanceof \Transliterator) { 109 $s = $rule->transliterate($s); 110 } elseif ($rule instanceof \Closure) { 111 $s = $rule($s); 112 } elseif ($rule) { 113 if ('nfd' === $rule = strtolower($rule)) { 114 normalizer_is_normalized($s, self::NFD) ?: $s = normalizer_normalize($s, self::NFD); 115 } elseif ('nfkd' === $rule) { 116 normalizer_is_normalized($s, self::NFKD) ?: $s = normalizer_normalize($s, self::NFKD); 117 } elseif ('[:nonspacing mark:] remove' === $rule) { 118 $s = preg_replace('/\p{Mn}++/u', '', $s); 119 } elseif ('latin-ascii' === $rule) { 120 $s = str_replace(self::TRANSLIT_FROM, self::TRANSLIT_TO, $s); 121 } elseif ('de-ascii' === $rule) { 122 $s = preg_replace("/([AUO])\u{0308}(?=\p{Ll})/u", '$1e', $s); 123 $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); 124 } elseif (\function_exists('transliterator_transliterate')) { 125 if (null === $transliterator = self::$transliterators[$rule] ?? self::$transliterators[$rule] = \Transliterator::create($rule)) { 126 if ('any-latin/bgn' === $rule) { 127 $rule = 'any-latin'; 128 $transliterator = self::$transliterators[$rule] ?? self::$transliterators[$rule] = \Transliterator::create($rule); 129 } 130 131 if (null === $transliterator) { 132 throw new InvalidArgumentException(sprintf('Unknown transliteration rule "%s".', $rule)); 133 } 134 135 self::$transliterators['any-latin/bgn'] = $transliterator; 136 } 137 138 $s = $transliterator->transliterate($s); 139 } 140 } elseif (!\function_exists('iconv')) { 141 $s = preg_replace('/[^\x00-\x7F]/u', '?', $s); 142 } else { 143 $s = @preg_replace_callback('/[^\x00-\x7F]/u', static function ($c) { 144 $c = (string) iconv('UTF-8', 'ASCII//TRANSLIT', $c[0]); 145 146 if ('' === $c && '' === iconv('UTF-8', 'ASCII//TRANSLIT', '²')) { 147 throw new \LogicException(sprintf('"%s" requires a translit-able iconv implementation, try installing "gnu-libiconv" if you\'re using Alpine Linux.', static::class)); 148 } 149 150 return 1 < \strlen($c) ? ltrim($c, '\'`"^~') : ('' !== $c ? $c : '?'); 151 }, $s); 152 } 153 } 154 155 $str->string .= $s; 156 157 return $str; 158 } 159 160 public function camel(): parent 161 { 162 $str = clone $this; 163 $str->string = str_replace(' ', '', preg_replace_callback('/\b./u', static function ($m) use (&$i) { 164 return 1 === ++$i ? ('İ' === $m[0] ? 'i̇' : mb_strtolower($m[0], 'UTF-8')) : mb_convert_case($m[0], \MB_CASE_TITLE, 'UTF-8'); 165 }, preg_replace('/[^\pL0-9]++/u', ' ', $this->string))); 166 167 return $str; 168 } 169 170 /** 171 * @return int[] 172 */ 173 public function codePointsAt(int $offset): array 174 { 175 $str = $this->slice($offset, 1); 176 177 if ('' === $str->string) { 178 return []; 179 } 180 181 $codePoints = []; 182 183 foreach (preg_split('//u', $str->string, -1, \PREG_SPLIT_NO_EMPTY) as $c) { 184 $codePoints[] = mb_ord($c, 'UTF-8'); 185 } 186 187 return $codePoints; 188 } 189 190 public function folded(bool $compat = true): parent 191 { 192 $str = clone $this; 193 194 if (!$compat || \PHP_VERSION_ID < 70300 || !\defined('Normalizer::NFKC_CF')) { 195 $str->string = normalizer_normalize($str->string, $compat ? \Normalizer::NFKC : \Normalizer::NFC); 196 $str->string = mb_strtolower(str_replace(self::FOLD_FROM, self::FOLD_TO, $this->string), 'UTF-8'); 197 } else { 198 $str->string = normalizer_normalize($str->string, \Normalizer::NFKC_CF); 199 } 200 201 return $str; 202 } 203 204 public function join(array $strings, string $lastGlue = null): parent 205 { 206 $str = clone $this; 207 208 $tail = null !== $lastGlue && 1 < \count($strings) ? $lastGlue.array_pop($strings) : ''; 209 $str->string = implode($this->string, $strings).$tail; 210 211 if (!preg_match('//u', $str->string)) { 212 throw new InvalidArgumentException('Invalid UTF-8 string.'); 213 } 214 215 return $str; 216 } 217 218 public function lower(): parent 219 { 220 $str = clone $this; 221 $str->string = mb_strtolower(str_replace('İ', 'i̇', $str->string), 'UTF-8'); 222 223 return $str; 224 } 225 226 public function match(string $regexp, int $flags = 0, int $offset = 0): array 227 { 228 $match = ((\PREG_PATTERN_ORDER | \PREG_SET_ORDER) & $flags) ? 'preg_match_all' : 'preg_match'; 229 230 if ($this->ignoreCase) { 231 $regexp .= 'i'; 232 } 233 234 set_error_handler(static function ($t, $m) { throw new InvalidArgumentException($m); }); 235 236 try { 237 if (false === $match($regexp.'u', $this->string, $matches, $flags | \PREG_UNMATCHED_AS_NULL, $offset)) { 238 $lastError = preg_last_error(); 239 240 foreach (get_defined_constants(true)['pcre'] as $k => $v) { 241 if ($lastError === $v && '_ERROR' === substr($k, -6)) { 242 throw new RuntimeException('Matching failed with '.$k.'.'); 243 } 244 } 245 246 throw new RuntimeException('Matching failed with unknown error code.'); 247 } 248 } finally { 249 restore_error_handler(); 250 } 251 252 return $matches; 253 } 254 255 /** 256 * @return static 257 */ 258 public function normalize(int $form = self::NFC): self 259 { 260 if (!\in_array($form, [self::NFC, self::NFD, self::NFKC, self::NFKD])) { 261 throw new InvalidArgumentException('Unsupported normalization form.'); 262 } 263 264 $str = clone $this; 265 normalizer_is_normalized($str->string, $form) ?: $str->string = normalizer_normalize($str->string, $form); 266 267 return $str; 268 } 269 270 public function padBoth(int $length, string $padStr = ' '): parent 271 { 272 if ('' === $padStr || !preg_match('//u', $padStr)) { 273 throw new InvalidArgumentException('Invalid UTF-8 string.'); 274 } 275 276 $pad = clone $this; 277 $pad->string = $padStr; 278 279 return $this->pad($length, $pad, \STR_PAD_BOTH); 280 } 281 282 public function padEnd(int $length, string $padStr = ' '): parent 283 { 284 if ('' === $padStr || !preg_match('//u', $padStr)) { 285 throw new InvalidArgumentException('Invalid UTF-8 string.'); 286 } 287 288 $pad = clone $this; 289 $pad->string = $padStr; 290 291 return $this->pad($length, $pad, \STR_PAD_RIGHT); 292 } 293 294 public function padStart(int $length, string $padStr = ' '): parent 295 { 296 if ('' === $padStr || !preg_match('//u', $padStr)) { 297 throw new InvalidArgumentException('Invalid UTF-8 string.'); 298 } 299 300 $pad = clone $this; 301 $pad->string = $padStr; 302 303 return $this->pad($length, $pad, \STR_PAD_LEFT); 304 } 305 306 public function replaceMatches(string $fromRegexp, $to): parent 307 { 308 if ($this->ignoreCase) { 309 $fromRegexp .= 'i'; 310 } 311 312 if (\is_array($to) || $to instanceof \Closure) { 313 if (!\is_callable($to)) { 314 throw new \TypeError(sprintf('Argument 2 passed to "%s::replaceMatches()" must be callable, array given.', static::class)); 315 } 316 317 $replace = 'preg_replace_callback'; 318 $to = static function (array $m) use ($to): string { 319 $to = $to($m); 320 321 if ('' !== $to && (!\is_string($to) || !preg_match('//u', $to))) { 322 throw new InvalidArgumentException('Replace callback must return a valid UTF-8 string.'); 323 } 324 325 return $to; 326 }; 327 } elseif ('' !== $to && !preg_match('//u', $to)) { 328 throw new InvalidArgumentException('Invalid UTF-8 string.'); 329 } else { 330 $replace = 'preg_replace'; 331 } 332 333 set_error_handler(static function ($t, $m) { throw new InvalidArgumentException($m); }); 334 335 try { 336 if (null === $string = $replace($fromRegexp.'u', $to, $this->string)) { 337 $lastError = preg_last_error(); 338 339 foreach (get_defined_constants(true)['pcre'] as $k => $v) { 340 if ($lastError === $v && '_ERROR' === substr($k, -6)) { 341 throw new RuntimeException('Matching failed with '.$k.'.'); 342 } 343 } 344 345 throw new RuntimeException('Matching failed with unknown error code.'); 346 } 347 } finally { 348 restore_error_handler(); 349 } 350 351 $str = clone $this; 352 $str->string = $string; 353 354 return $str; 355 } 356 357 public function reverse(): parent 358 { 359 $str = clone $this; 360 $str->string = implode('', array_reverse(preg_split('/(\X)/u', $str->string, -1, \PREG_SPLIT_DELIM_CAPTURE | \PREG_SPLIT_NO_EMPTY))); 361 362 return $str; 363 } 364 365 public function snake(): parent 366 { 367 $str = $this->camel()->title(); 368 $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'); 369 370 return $str; 371 } 372 373 public function title(bool $allWords = false): parent 374 { 375 $str = clone $this; 376 377 $limit = $allWords ? -1 : 1; 378 379 $str->string = preg_replace_callback('/\b./u', static function (array $m): string { 380 return mb_convert_case($m[0], \MB_CASE_TITLE, 'UTF-8'); 381 }, $str->string, $limit); 382 383 return $str; 384 } 385 386 public function trim(string $chars = " \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}"): parent 387 { 388 if (" \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}" !== $chars && !preg_match('//u', $chars)) { 389 throw new InvalidArgumentException('Invalid UTF-8 chars.'); 390 } 391 $chars = preg_quote($chars); 392 393 $str = clone $this; 394 $str->string = preg_replace("{^[$chars]++|[$chars]++$}uD", '', $str->string); 395 396 return $str; 397 } 398 399 public function trimEnd(string $chars = " \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}"): parent 400 { 401 if (" \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}" !== $chars && !preg_match('//u', $chars)) { 402 throw new InvalidArgumentException('Invalid UTF-8 chars.'); 403 } 404 $chars = preg_quote($chars); 405 406 $str = clone $this; 407 $str->string = preg_replace("{[$chars]++$}uD", '', $str->string); 408 409 return $str; 410 } 411 412 public function trimStart(string $chars = " \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}"): parent 413 { 414 if (" \t\n\r\0\x0B\x0C\u{A0}\u{FEFF}" !== $chars && !preg_match('//u', $chars)) { 415 throw new InvalidArgumentException('Invalid UTF-8 chars.'); 416 } 417 $chars = preg_quote($chars); 418 419 $str = clone $this; 420 $str->string = preg_replace("{^[$chars]++}uD", '', $str->string); 421 422 return $str; 423 } 424 425 public function upper(): parent 426 { 427 $str = clone $this; 428 $str->string = mb_strtoupper($str->string, 'UTF-8'); 429 430 if (\PHP_VERSION_ID < 70300) { 431 $str->string = str_replace(self::UPPER_FROM, self::UPPER_TO, $str->string); 432 } 433 434 return $str; 435 } 436 437 public function width(bool $ignoreAnsiDecoration = true): int 438 { 439 $width = 0; 440 $s = str_replace(["\x00", "\x05", "\x07"], '', $this->string); 441 442 if (false !== strpos($s, "\r")) { 443 $s = str_replace(["\r\n", "\r"], "\n", $s); 444 } 445 446 if (!$ignoreAnsiDecoration) { 447 $s = preg_replace('/[\p{Cc}\x7F]++/u', '', $s); 448 } 449 450 foreach (explode("\n", $s) as $s) { 451 if ($ignoreAnsiDecoration) { 452 $s = preg_replace('/(?:\x1B(?: 453 \[ [\x30-\x3F]*+ [\x20-\x2F]*+ [0x40-\x7E] 454 | [P\]X^_] .*? \x1B\\\\ 455 | [\x41-\x7E] 456 )|[\p{Cc}\x7F]++)/xu', '', $s); 457 } 458 459 // Non printable characters have been dropped, so wcswidth cannot logically return -1. 460 $width += $this->wcswidth($s); 461 } 462 463 return $width; 464 } 465 466 /** 467 * @return static 468 */ 469 private function pad(int $len, self $pad, int $type): parent 470 { 471 $sLen = $this->length(); 472 473 if ($len <= $sLen) { 474 return clone $this; 475 } 476 477 $padLen = $pad->length(); 478 $freeLen = $len - $sLen; 479 $len = $freeLen % $padLen; 480 481 switch ($type) { 482 case \STR_PAD_RIGHT: 483 return $this->append(str_repeat($pad->string, $freeLen / $padLen).($len ? $pad->slice(0, $len) : '')); 484 485 case \STR_PAD_LEFT: 486 return $this->prepend(str_repeat($pad->string, $freeLen / $padLen).($len ? $pad->slice(0, $len) : '')); 487 488 case \STR_PAD_BOTH: 489 $freeLen /= 2; 490 491 $rightLen = ceil($freeLen); 492 $len = $rightLen % $padLen; 493 $str = $this->append(str_repeat($pad->string, $rightLen / $padLen).($len ? $pad->slice(0, $len) : '')); 494 495 $leftLen = floor($freeLen); 496 $len = $leftLen % $padLen; 497 498 return $str->prepend(str_repeat($pad->string, $leftLen / $padLen).($len ? $pad->slice(0, $len) : '')); 499 500 default: 501 throw new InvalidArgumentException('Invalid padding type.'); 502 } 503 } 504 505 /** 506 * Based on https://github.com/jquast/wcwidth, a Python implementation of https://www.cl.cam.ac.uk/~mgk25/ucs/wcwidth.c. 507 */ 508 private function wcswidth(string $string): int 509 { 510 $width = 0; 511 512 foreach (preg_split('//u', $string, -1, \PREG_SPLIT_NO_EMPTY) as $c) { 513 $codePoint = mb_ord($c, 'UTF-8'); 514 515 if (0 === $codePoint // NULL 516 || 0x034F === $codePoint // COMBINING GRAPHEME JOINER 517 || (0x200B <= $codePoint && 0x200F >= $codePoint) // ZERO WIDTH SPACE to RIGHT-TO-LEFT MARK 518 || 0x2028 === $codePoint // LINE SEPARATOR 519 || 0x2029 === $codePoint // PARAGRAPH SEPARATOR 520 || (0x202A <= $codePoint && 0x202E >= $codePoint) // LEFT-TO-RIGHT EMBEDDING to RIGHT-TO-LEFT OVERRIDE 521 || (0x2060 <= $codePoint && 0x2063 >= $codePoint) // WORD JOINER to INVISIBLE SEPARATOR 522 ) { 523 continue; 524 } 525 526 // Non printable characters 527 if (32 > $codePoint // C0 control characters 528 || (0x07F <= $codePoint && 0x0A0 > $codePoint) // C1 control characters and DEL 529 ) { 530 return -1; 531 } 532 533 static $tableZero; 534 if (null === $tableZero) { 535 $tableZero = require __DIR__.'/Resources/data/wcswidth_table_zero.php'; 536 } 537 538 if ($codePoint >= $tableZero[0][0] && $codePoint <= $tableZero[$ubound = \count($tableZero) - 1][1]) { 539 $lbound = 0; 540 while ($ubound >= $lbound) { 541 $mid = floor(($lbound + $ubound) / 2); 542 543 if ($codePoint > $tableZero[$mid][1]) { 544 $lbound = $mid + 1; 545 } elseif ($codePoint < $tableZero[$mid][0]) { 546 $ubound = $mid - 1; 547 } else { 548 continue 2; 549 } 550 } 551 } 552 553 static $tableWide; 554 if (null === $tableWide) { 555 $tableWide = require __DIR__.'/Resources/data/wcswidth_table_wide.php'; 556 } 557 558 if ($codePoint >= $tableWide[0][0] && $codePoint <= $tableWide[$ubound = \count($tableWide) - 1][1]) { 559 $lbound = 0; 560 while ($ubound >= $lbound) { 561 $mid = floor(($lbound + $ubound) / 2); 562 563 if ($codePoint > $tableWide[$mid][1]) { 564 $lbound = $mid + 1; 565 } elseif ($codePoint < $tableWide[$mid][0]) { 566 $ubound = $mid - 1; 567 } else { 568 $width += 2; 569 570 continue 2; 571 } 572 } 573 } 574 575 ++$width; 576 } 577 578 return $width; 579 } 580} 581