diff options
Diffstat (limited to 'vendor/doctrine/lexer/lib')
| -rw-r--r-- | vendor/doctrine/lexer/lib/Doctrine/Common/Lexer/AbstractLexer.php | 328 |
1 files changed, 328 insertions, 0 deletions
diff --git a/vendor/doctrine/lexer/lib/Doctrine/Common/Lexer/AbstractLexer.php b/vendor/doctrine/lexer/lib/Doctrine/Common/Lexer/AbstractLexer.php new file mode 100644 index 0000000..385643a --- /dev/null +++ b/vendor/doctrine/lexer/lib/Doctrine/Common/Lexer/AbstractLexer.php @@ -0,0 +1,328 @@ +<?php + +declare(strict_types=1); + +namespace Doctrine\Common\Lexer; + +use ReflectionClass; +use const PREG_SPLIT_DELIM_CAPTURE; +use const PREG_SPLIT_NO_EMPTY; +use const PREG_SPLIT_OFFSET_CAPTURE; +use function implode; +use function in_array; +use function preg_split; +use function sprintf; +use function substr; + +/** + * Base class for writing simple lexers, i.e. for creating small DSLs. + */ +abstract class AbstractLexer +{ + /** + * Lexer original input string. + * + * @var string + */ + private $input; + + /** + * Array of scanned tokens. + * + * Each token is an associative array containing three items: + * - 'value' : the string value of the token in the input string + * - 'type' : the type of the token (identifier, numeric, string, input + * parameter, none) + * - 'position' : the position of the token in the input string + * + * @var array + */ + private $tokens = []; + + /** + * Current lexer position in input string. + * + * @var int + */ + private $position = 0; + + /** + * Current peek of current lexer position. + * + * @var int + */ + private $peek = 0; + + /** + * The next token in the input. + * + * @var array|null + */ + public $lookahead; + + /** + * The last matched/seen token. + * + * @var array|null + */ + public $token; + + /** + * Composed regex for input parsing. + * + * @var string + */ + private $regex; + + /** + * Sets the input data to be tokenized. + * + * The Lexer is immediately reset and the new input tokenized. + * Any unprocessed tokens from any previous input are lost. + * + * @param string $input The input to be tokenized. + * + * @return void + */ + public function setInput($input) + { + $this->input = $input; + $this->tokens = []; + + $this->reset(); + $this->scan($input); + } + + /** + * Resets the lexer. + * + * @return void + */ + public function reset() + { + $this->lookahead = null; + $this->token = null; + $this->peek = 0; + $this->position = 0; + } + + /** + * Resets the peek pointer to 0. + * + * @return void + */ + public function resetPeek() + { + $this->peek = 0; + } + + /** + * Resets the lexer position on the input to the given position. + * + * @param int $position Position to place the lexical scanner. + * + * @return void + */ + public function resetPosition($position = 0) + { + $this->position = $position; + } + + /** + * Retrieve the original lexer's input until a given position. + * + * @param int $position + * + * @return string + */ + public function getInputUntilPosition($position) + { + return substr($this->input, 0, $position); + } + + /** + * Checks whether a given token matches the current lookahead. + * + * @param int|string $token + * + * @return bool + */ + public function isNextToken($token) + { + return $this->lookahead !== null && $this->lookahead['type'] === $token; + } + + /** + * Checks whether any of the given tokens matches the current lookahead. + * + * @param array $tokens + * + * @return bool + */ + public function isNextTokenAny(array $tokens) + { + return $this->lookahead !== null && in_array($this->lookahead['type'], $tokens, true); + } + + /** + * Moves to the next token in the input string. + * + * @return bool + */ + public function moveNext() + { + $this->peek = 0; + $this->token = $this->lookahead; + $this->lookahead = isset($this->tokens[$this->position]) + ? $this->tokens[$this->position++] : null; + + return $this->lookahead !== null; + } + + /** + * Tells the lexer to skip input tokens until it sees a token with the given value. + * + * @param string $type The token type to skip until. + * + * @return void + */ + public function skipUntil($type) + { + while ($this->lookahead !== null && $this->lookahead['type'] !== $type) { + $this->moveNext(); + } + } + + /** + * Checks if given value is identical to the given token. + * + * @param mixed $value + * @param int|string $token + * + * @return bool + */ + public function isA($value, $token) + { + return $this->getType($value) === $token; + } + + /** + * Moves the lookahead token forward. + * + * @return array|null The next token or NULL if there are no more tokens ahead. + */ + public function peek() + { + if (isset($this->tokens[$this->position + $this->peek])) { + return $this->tokens[$this->position + $this->peek++]; + } + + return null; + } + + /** + * Peeks at the next token, returns it and immediately resets the peek. + * + * @return array|null The next token or NULL if there are no more tokens ahead. + */ + public function glimpse() + { + $peek = $this->peek(); + $this->peek = 0; + + return $peek; + } + + /** + * Scans the input string for tokens. + * + * @param string $input A query string. + * + * @return void + */ + protected function scan($input) + { + if (! isset($this->regex)) { + $this->regex = sprintf( + '/(%s)|%s/%s', + implode(')|(', $this->getCatchablePatterns()), + implode('|', $this->getNonCatchablePatterns()), + $this->getModifiers() + ); + } + + $flags = PREG_SPLIT_NO_EMPTY | PREG_SPLIT_DELIM_CAPTURE | PREG_SPLIT_OFFSET_CAPTURE; + $matches = preg_split($this->regex, $input, -1, $flags); + + if ($matches === false) { + // Work around https://bugs.php.net/78122 + $matches = [[$input, 0]]; + } + + foreach ($matches as $match) { + // Must remain before 'value' assignment since it can change content + $type = $this->getType($match[0]); + + $this->tokens[] = [ + 'value' => $match[0], + 'type' => $type, + 'position' => $match[1], + ]; + } + } + + /** + * Gets the literal for a given token. + * + * @param int|string $token + * + * @return int|string + */ + public function getLiteral($token) + { + $className = static::class; + $reflClass = new ReflectionClass($className); + $constants = $reflClass->getConstants(); + + foreach ($constants as $name => $value) { + if ($value === $token) { + return $className . '::' . $name; + } + } + + return $token; + } + + /** + * Regex modifiers + * + * @return string + */ + protected function getModifiers() + { + return 'iu'; + } + + /** + * Lexical catchable patterns. + * + * @return array + */ + abstract protected function getCatchablePatterns(); + + /** + * Lexical non-catchable patterns. + * + * @return array + */ + abstract protected function getNonCatchablePatterns(); + + /** + * Retrieve token type. Also processes the token value if necessary. + * + * @param string $value + * + * @return int|string|null + */ + abstract protected function getType(&$value); +} |