/usr/share/php/PhpParser/Lexer/Emulative.php is in php-parser 1.0.1-1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 | <?php
namespace PhpParser\Lexer;
use PhpParser\Parser;
/**
* ATTENTION: This code is WRITE-ONLY. Do not try to read it.
*/
class Emulative extends \PhpParser\Lexer
{
protected $newKeywords;
protected $inObjectAccess;
const T_ELLIPSIS = 1001;
const T_POW = 1002;
const T_POW_EQUAL = 1003;
const PHP_5_6 = '5.6.0rc1';
const PHP_5_5 = '5.5.0beta1';
const PHP_5_4 = '5.4.0beta1';
public function __construct() {
parent::__construct();
$newKeywordsPerVersion = array(
self::PHP_5_5 => array(
'finally' => Parser::T_FINALLY,
'yield' => Parser::T_YIELD,
),
self::PHP_5_4 => array(
'callable' => Parser::T_CALLABLE,
'insteadof' => Parser::T_INSTEADOF,
'trait' => Parser::T_TRAIT,
'__trait__' => Parser::T_TRAIT_C,
),
);
$this->newKeywords = array();
foreach ($newKeywordsPerVersion as $version => $newKeywords) {
if (version_compare(PHP_VERSION, $version, '>=')) {
break;
}
$this->newKeywords += $newKeywords;
}
if (version_compare(PHP_VERSION, self::PHP_5_6, '<')) {
$this->tokenMap[self::T_ELLIPSIS] = Parser::T_ELLIPSIS;
$this->tokenMap[self::T_POW] = Parser::T_POW;
$this->tokenMap[self::T_POW_EQUAL] = Parser::T_POW_EQUAL;
}
}
public function startLexing($code) {
$this->inObjectAccess = false;
$preprocessedCode = $this->preprocessCode($code);
parent::startLexing($preprocessedCode);
if ($preprocessedCode !== $code) {
$this->postprocessTokens();
}
}
/*
* Replaces new features in the code by ~__EMU__{NAME}__{DATA}__~ sequences.
* ~LABEL~ is never valid PHP code, that's why we can (to some degree) safely
* use it here.
* Later when preprocessing the tokens these sequences will either be replaced
* by real tokens or replaced with their original content (e.g. if they occurred
* inside a string, i.e. a place where they don't have a special meaning).
*/
protected function preprocessCode($code) {
if (version_compare(PHP_VERSION, self::PHP_5_6, '>=')) {
return $code;
}
$code = str_replace('...', '~__EMU__ELLIPSIS__~', $code);
$code = preg_replace('((?<!/)\*\*=)', '~__EMU__POWEQUAL__~', $code);
$code = preg_replace('((?<!/)\*\*(?!/))', '~__EMU__POW__~', $code);
if (version_compare(PHP_VERSION, self::PHP_5_4, '>=')) {
return $code;
}
// binary notation (0b010101101001...)
return preg_replace('(\b0b[01]+\b)', '~__EMU__BINARY__$0__~', $code);
}
/*
* Replaces the ~__EMU__...~ sequences with real tokens or their original
* value.
*/
protected function postprocessTokens() {
// we need to manually iterate and manage a count because we'll change
// the tokens array on the way
for ($i = 0, $c = count($this->tokens); $i < $c; ++$i) {
// first check that the following tokens are of form ~LABEL~,
// then match the __EMU__... sequence.
if ('~' === $this->tokens[$i]
&& isset($this->tokens[$i + 2])
&& '~' === $this->tokens[$i + 2]
&& T_STRING === $this->tokens[$i + 1][0]
&& preg_match('(^__EMU__([A-Z]++)__(?:([A-Za-z0-9]++)__)?$)', $this->tokens[$i + 1][1], $matches)
) {
if ('BINARY' === $matches[1]) {
// the binary number can either be an integer or a double, so return a LNUMBER
// or DNUMBER respectively
$replace = array(
array(is_int(bindec($matches[2])) ? T_LNUMBER : T_DNUMBER, $matches[2], $this->tokens[$i + 1][2])
);
} else if ('ELLIPSIS' === $matches[1]) {
$replace = array(
array(self::T_ELLIPSIS, '...', $this->tokens[$i + 1][2])
);
} else if ('POW' === $matches[1]) {
$replace = array(
array(self::T_POW, '**', $this->tokens[$i + 1][2])
);
} else if ('POWEQUAL' === $matches[1]) {
$replace = array(
array(self::T_POW_EQUAL, '**=', $this->tokens[$i + 1][2])
);
} else {
// just ignore all other __EMU__ sequences
continue;
}
array_splice($this->tokens, $i, 3, $replace);
$c -= 3 - count($replace);
// for multichar tokens (e.g. strings) replace any ~__EMU__...~ sequences
// in their content with the original character sequence
} elseif (is_array($this->tokens[$i])
&& 0 !== strpos($this->tokens[$i][1], '__EMU__')
) {
$this->tokens[$i][1] = preg_replace_callback(
'(~__EMU__([A-Z]++)__(?:([A-Za-z0-9]++)__)?~)',
array($this, 'restoreContentCallback'),
$this->tokens[$i][1]
);
}
}
}
/*
* This method is a callback for restoring EMU sequences in
* multichar tokens (like strings) to their original value.
*/
public function restoreContentCallback(array $matches) {
if ('BINARY' === $matches[1]) {
return $matches[2];
} else if ('ELLIPSIS' === $matches[1]) {
return '...';
} else if ('POW' === $matches[1]) {
return '**';
} else if ('POWEQUAL' === $matches[1]) {
return '**=';
} else {
return $matches[0];
}
}
public function getNextToken(&$value = null, &$startAttributes = null, &$endAttributes = null) {
$token = parent::getNextToken($value, $startAttributes, $endAttributes);
// replace new keywords by their respective tokens. This is not done
// if we currently are in an object access (e.g. in $obj->namespace
// "namespace" stays a T_STRING tokens and isn't converted to T_NAMESPACE)
if (Parser::T_STRING === $token && !$this->inObjectAccess) {
if (isset($this->newKeywords[strtolower($value)])) {
return $this->newKeywords[strtolower($value)];
}
// keep track of whether we currently are in an object access (after ->)
} elseif (Parser::T_OBJECT_OPERATOR === $token) {
$this->inObjectAccess = true;
} else {
$this->inObjectAccess = false;
}
return $token;
}
}
|