Lexer.php 4.1 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101
  1. <?php
  2. /*
  3. * This file is part of the Symfony package.
  4. *
  5. * (c) Fabien Potencier <fabien@symfony.com>
  6. *
  7. * For the full copyright and license information, please view the LICENSE
  8. * file that was distributed with this source code.
  9. */
  10. namespace Symfony\Component\ExpressionLanguage;
  11. /**
  12. * Lexes an expression.
  13. *
  14. * @author Fabien Potencier <fabien@symfony.com>
  15. */
  16. class Lexer
  17. {
  18. /**
  19. * Tokenizes an expression.
  20. *
  21. * @return TokenStream A token stream instance
  22. *
  23. * @throws SyntaxError
  24. */
  25. public function tokenize(string $expression)
  26. {
  27. $expression = str_replace(["\r", "\n", "\t", "\v", "\f"], ' ', $expression);
  28. $cursor = 0;
  29. $tokens = [];
  30. $brackets = [];
  31. $end = \strlen($expression);
  32. while ($cursor < $end) {
  33. if (' ' == $expression[$cursor]) {
  34. ++$cursor;
  35. continue;
  36. }
  37. if (preg_match('/[0-9]+(?:\.[0-9]+)?([Ee][\+\-][0-9]+)?/A', $expression, $match, 0, $cursor)) {
  38. // numbers
  39. $number = (float) $match[0]; // floats
  40. if (preg_match('/^[0-9]+$/', $match[0]) && $number <= \PHP_INT_MAX) {
  41. $number = (int) $match[0]; // integers lower than the maximum
  42. }
  43. $tokens[] = new Token(Token::NUMBER_TYPE, $number, $cursor + 1);
  44. $cursor += \strlen($match[0]);
  45. } elseif (false !== strpos('([{', $expression[$cursor])) {
  46. // opening bracket
  47. $brackets[] = [$expression[$cursor], $cursor];
  48. $tokens[] = new Token(Token::PUNCTUATION_TYPE, $expression[$cursor], $cursor + 1);
  49. ++$cursor;
  50. } elseif (false !== strpos(')]}', $expression[$cursor])) {
  51. // closing bracket
  52. if (empty($brackets)) {
  53. throw new SyntaxError(sprintf('Unexpected "%s".', $expression[$cursor]), $cursor, $expression);
  54. }
  55. [$expect, $cur] = array_pop($brackets);
  56. if ($expression[$cursor] != strtr($expect, '([{', ')]}')) {
  57. throw new SyntaxError(sprintf('Unclosed "%s".', $expect), $cur, $expression);
  58. }
  59. $tokens[] = new Token(Token::PUNCTUATION_TYPE, $expression[$cursor], $cursor + 1);
  60. ++$cursor;
  61. } elseif (preg_match('/"([^"\\\\]*(?:\\\\.[^"\\\\]*)*)"|\'([^\'\\\\]*(?:\\\\.[^\'\\\\]*)*)\'/As', $expression, $match, 0, $cursor)) {
  62. // strings
  63. $tokens[] = new Token(Token::STRING_TYPE, stripcslashes(substr($match[0], 1, -1)), $cursor + 1);
  64. $cursor += \strlen($match[0]);
  65. } elseif (preg_match('/(?<=^|[\s(])not in(?=[\s(])|\!\=\=|(?<=^|[\s(])not(?=[\s(])|(?<=^|[\s(])and(?=[\s(])|\=\=\=|\>\=|(?<=^|[\s(])or(?=[\s(])|\<\=|\*\*|\.\.|(?<=^|[\s(])in(?=[\s(])|&&|\|\||(?<=^|[\s(])matches|\=\=|\!\=|\*|~|%|\/|\>|\||\!|\^|&|\+|\<|\-/A', $expression, $match, 0, $cursor)) {
  66. // operators
  67. $tokens[] = new Token(Token::OPERATOR_TYPE, $match[0], $cursor + 1);
  68. $cursor += \strlen($match[0]);
  69. } elseif (false !== strpos('.,?:', $expression[$cursor])) {
  70. // punctuation
  71. $tokens[] = new Token(Token::PUNCTUATION_TYPE, $expression[$cursor], $cursor + 1);
  72. ++$cursor;
  73. } elseif (preg_match('/[a-zA-Z_\x7f-\xff][a-zA-Z0-9_\x7f-\xff]*/A', $expression, $match, 0, $cursor)) {
  74. // names
  75. $tokens[] = new Token(Token::NAME_TYPE, $match[0], $cursor + 1);
  76. $cursor += \strlen($match[0]);
  77. } else {
  78. // unlexable
  79. throw new SyntaxError(sprintf('Unexpected character "%s".', $expression[$cursor]), $cursor, $expression);
  80. }
  81. }
  82. $tokens[] = new Token(Token::EOF_TYPE, null, $cursor + 1);
  83. if (!empty($brackets)) {
  84. [$expect, $cur] = array_pop($brackets);
  85. throw new SyntaxError(sprintf('Unclosed "%s".', $expect), $cur, $expression);
  86. }
  87. return new TokenStream($tokens, $expression);
  88. }
  89. }