TokenParser.php
183 lines
| 1 | <?php declare(strict_types=1); |
| 2 | |
| 3 | namespace Cybot\Dependencies\PhpDocReader\PhpParser; |
| 4 | |
| 5 | /** |
| 6 | * Parses a file for namespaces/use/class declarations. |
| 7 | * |
| 8 | * Class taken and adapted from doctrine/annotations to avoid pulling the whole package. |
| 9 | * |
| 10 | * @author Fabien Potencier <fabien@symfony.com> |
| 11 | * @author Christian Kaps <christian.kaps@mohiva.com> |
| 12 | */ |
| 13 | class TokenParser |
| 14 | { |
| 15 | /** |
| 16 | * The token list. |
| 17 | * |
| 18 | * @var list<mixed[]> |
| 19 | */ |
| 20 | private $tokens; |
| 21 | |
| 22 | /** |
| 23 | * The number of tokens. |
| 24 | * |
| 25 | * @var int |
| 26 | */ |
| 27 | private $numTokens; |
| 28 | |
| 29 | /** |
| 30 | * The current array pointer. |
| 31 | * |
| 32 | * @var int |
| 33 | */ |
| 34 | private $pointer = 0; |
| 35 | |
| 36 | /** |
| 37 | * @param string $contents |
| 38 | */ |
| 39 | public function __construct($contents) |
| 40 | { |
| 41 | $this->tokens = token_get_all($contents); |
| 42 | |
| 43 | // The PHP parser sets internal compiler globals for certain things. Annoyingly, the last docblock comment it |
| 44 | // saw gets stored in doc_comment. When it comes to compile the next thing to be include()d this stored |
| 45 | // doc_comment becomes owned by the first thing the compiler sees in the file that it considers might have a |
| 46 | // docblock. If the first thing in the file is a class without a doc block this would cause calls to |
| 47 | // getDocBlock() on said class to return our long lost doc_comment. Argh. |
| 48 | // To workaround, cause the parser to parse an empty docblock. Sure getDocBlock() will return this, but at least |
| 49 | // it's harmless to us. |
| 50 | token_get_all("<?php\n/**\n *\n */"); |
| 51 | |
| 52 | $this->numTokens = count($this->tokens); |
| 53 | } |
| 54 | |
| 55 | /** |
| 56 | * Gets all use statements. |
| 57 | * |
| 58 | * @param string $namespaceName The namespace name of the reflected class. |
| 59 | * |
| 60 | * @return array<string, string> A list with all found use statements. |
| 61 | */ |
| 62 | public function parseUseStatements($namespaceName) |
| 63 | { |
| 64 | $statements = []; |
| 65 | while (($token = $this->next())) { |
| 66 | if ($token[0] === T_USE) { |
| 67 | $statements = array_merge($statements, $this->parseUseStatement()); |
| 68 | continue; |
| 69 | } |
| 70 | |
| 71 | if ($token[0] !== T_NAMESPACE || $this->parseNamespace() !== $namespaceName) { |
| 72 | continue; |
| 73 | } |
| 74 | |
| 75 | // Get fresh array for new namespace. This is to prevent the parser to collect the use statements |
| 76 | // for a previous namespace with the same name. This is the case if a namespace is defined twice |
| 77 | // or if a namespace with the same name is commented out. |
| 78 | $statements = []; |
| 79 | } |
| 80 | |
| 81 | return $statements; |
| 82 | } |
| 83 | |
| 84 | /** |
| 85 | * Gets the next non whitespace and non comment token. |
| 86 | * |
| 87 | * @param bool $docCommentIsComment If TRUE then a doc comment is considered a comment and skipped. |
| 88 | * If FALSE then only whitespace and normal comments are skipped. |
| 89 | * |
| 90 | * @return mixed[]|string|null The token if exists, null otherwise. |
| 91 | */ |
| 92 | private function next($docCommentIsComment = true) |
| 93 | { |
| 94 | for ($i = $this->pointer; $i < $this->numTokens; $i++) { |
| 95 | $this->pointer++; |
| 96 | if ( |
| 97 | $this->tokens[$i][0] === T_WHITESPACE || |
| 98 | $this->tokens[$i][0] === T_COMMENT || |
| 99 | ($docCommentIsComment && $this->tokens[$i][0] === T_DOC_COMMENT) |
| 100 | ) { |
| 101 | continue; |
| 102 | } |
| 103 | |
| 104 | return $this->tokens[$i]; |
| 105 | } |
| 106 | |
| 107 | return null; |
| 108 | } |
| 109 | |
| 110 | /** |
| 111 | * Parses a single use statement. |
| 112 | * |
| 113 | * @return array<string, string> A list with all found class names for a use statement. |
| 114 | */ |
| 115 | private function parseUseStatement() |
| 116 | { |
| 117 | $groupRoot = ''; |
| 118 | $class = ''; |
| 119 | $alias = ''; |
| 120 | $statements = []; |
| 121 | $explicitAlias = false; |
| 122 | while (($token = $this->next())) { |
| 123 | if (! $explicitAlias && $token[0] === T_STRING) { |
| 124 | $class .= $token[1]; |
| 125 | $alias = $token[1]; |
| 126 | } elseif ($explicitAlias && $token[0] === T_STRING) { |
| 127 | $alias = $token[1]; |
| 128 | } elseif ( |
| 129 | PHP_VERSION_ID >= 80000 && |
| 130 | ($token[0] === T_NAME_QUALIFIED || $token[0] === T_NAME_FULLY_QUALIFIED) |
| 131 | ) { |
| 132 | $class .= $token[1]; |
| 133 | |
| 134 | $classSplit = explode('\\', $token[1]); |
| 135 | $alias = $classSplit[count($classSplit) - 1]; |
| 136 | } elseif ($token[0] === T_NS_SEPARATOR) { |
| 137 | $class .= '\\'; |
| 138 | $alias = ''; |
| 139 | } elseif ($token[0] === T_AS) { |
| 140 | $explicitAlias = true; |
| 141 | $alias = ''; |
| 142 | } elseif ($token === ',') { |
| 143 | $statements[strtolower($alias)] = $groupRoot . $class; |
| 144 | $class = ''; |
| 145 | $alias = ''; |
| 146 | $explicitAlias = false; |
| 147 | } elseif ($token === ';') { |
| 148 | $statements[strtolower($alias)] = $groupRoot . $class; |
| 149 | break; |
| 150 | } elseif ($token === '{') { |
| 151 | $groupRoot = $class; |
| 152 | $class = ''; |
| 153 | } elseif ($token === '}') { |
| 154 | continue; |
| 155 | } else { |
| 156 | break; |
| 157 | } |
| 158 | } |
| 159 | |
| 160 | return $statements; |
| 161 | } |
| 162 | |
| 163 | /** |
| 164 | * Gets the namespace. |
| 165 | * |
| 166 | * @return string The found namespace. |
| 167 | */ |
| 168 | private function parseNamespace() |
| 169 | { |
| 170 | $name = ''; |
| 171 | while ( |
| 172 | ($token = $this->next()) && ($token[0] === T_STRING || $token[0] === T_NS_SEPARATOR || ( |
| 173 | PHP_VERSION_ID >= 80000 && |
| 174 | ($token[0] === T_NAME_QUALIFIED || $token[0] === T_NAME_FULLY_QUALIFIED) |
| 175 | )) |
| 176 | ) { |
| 177 | $name .= $token[1]; |
| 178 | } |
| 179 | |
| 180 | return $name; |
| 181 | } |
| 182 | } |
| 183 |