parser.php

来自「PHP 知识管理系统（基于树结构的知识管理系统）, 英文原版的PHP源码。」· PHP 代码 · 共 764 行 · 第 1/2 页

<?php
/**
 *  base include file for SimpleTest
 *  @package    SimpleTest
 *  @subpackage MockObjects
 *  @version    $Id: parser.php 1723 2008-04-08 00:34:10Z lastcraft $
 */

/**#@+
 * Lexer mode stack constants
 */
foreach (array('LEXER_ENTER', 'LEXER_MATCHED',
                'LEXER_UNMATCHED', 'LEXER_EXIT',
                'LEXER_SPECIAL') as $i => $constant) {
    if (! defined($constant)) {
        define($constant, $i + 1);
    }
}
/**#@-*/

/**
 *    Compounded regular expression. Any of
 *    the contained patterns could match and
 *    when one does, it's label is returned.
 *    @package SimpleTest
 *    @subpackage WebTester
 */
class ParallelRegex {
    var $_patterns;
    var $_labels;
    var $_regex;
    var $_case;
    
    /**
     *    Constructor. Starts with no patterns.
     *    @param boolean $case    True for case sensitive, false
     *                            for insensitive.
     *    @access public
     */
    function ParallelRegex($case) {
        $this->_case = $case;
        $this->_patterns = array();
        $this->_labels = array();
        $this->_regex = null;
    }
    
    /**
     *    Adds a pattern with an optional label.
     *    @param string $pattern      Perl style regex, but ( and )
     *                                lose the usual meaning.
     *    @param string $label        Label of regex to be returned
     *                                on a match.
     *    @access public
     */
    function addPattern($pattern, $label = true) {
        $count = count($this->_patterns);
        $this->_patterns[$count] = $pattern;
        $this->_labels[$count] = $label;
        $this->_regex = null;
    }
    
    /**
     *    Attempts to match all patterns at once against
     *    a string.
     *    @param string $subject      String to match against.
     *    @param string $match        First matched portion of
     *                                subject.
     *    @return boolean             True on success.
     *    @access public
     */
    function match($subject, &$match) {
        if (count($this->_patterns) == 0) {
            return false;
        }
        if (! preg_match($this->_getCompoundedRegex(), $subject, $matches)) {
            $match = '';
            return false;
        }
        $match = $matches[0];
        for ($i = 1; $i < count($matches); $i++) {
            if ($matches[$i]) {
                return $this->_labels[$i - 1];
            }
        }
        return true;
    }
    
    /**
     *    Compounds the patterns into a single
     *    regular expression separated with the
     *    "or" operator. Caches the regex.
     *    Will automatically escape (, ) and / tokens.
     *    @param array $patterns    List of patterns in order.
     *    @access private
     */
    function _getCompoundedRegex() {
        if ($this->_regex == null) {
            for ($i = 0, $count = count($this->_patterns); $i < $count; $i++) {
                $this->_patterns[$i] = '(' . str_replace(
                        array('/', '(', ')'),
                        array('\/', '\(', '\)'),
                        $this->_patterns[$i]) . ')';
            }
            $this->_regex = "/" . implode("|", $this->_patterns) . "/" . $this->_getPerlMatchingFlags();
        }
        return $this->_regex;
    }
    
    /**
     *    Accessor for perl regex mode flags to use.
     *    @return string       Perl regex flags.
     *    @access private
     */
    function _getPerlMatchingFlags() {
        return ($this->_case ? "msS" : "msSi");
    }
}

/**
 *    States for a stack machine.
 *    @package SimpleTest
 *    @subpackage WebTester
 */
class SimpleStateStack {
    var $_stack;
    
    /**
     *    Constructor. Starts in named state.
     *    @param string $start        Starting state name.
     *    @access public
     */
    function SimpleStateStack($start) {
        $this->_stack = array($start);
    }
    
    /**
     *    Accessor for current state.
     *    @return string       State.
     *    @access public
     */
    function getCurrent() {
        return $this->_stack[count($this->_stack) - 1];
    }
    
    /**
     *    Adds a state to the stack and sets it
     *    to be the current state.
     *    @param string $state        New state.
     *    @access public
     */
    function enter($state) {
        array_push($this->_stack, $state);
    }
    
    /**
     *    Leaves the current state and reverts
     *    to the previous one.
     *    @return boolean    False if we drop off
     *                       the bottom of the list.
     *    @access public
     */
    function leave() {
        if (count($this->_stack) == 1) {
            return false;
        }
        array_pop($this->_stack);
        return true;
    }
}

/**
 *    Accepts text and breaks it into tokens.
 *    Some optimisation to make the sure the
 *    content is only scanned by the PHP regex
 *    parser once. Lexer modes must not start
 *    with leading underscores.
 *    @package SimpleTest
 *    @subpackage WebTester
 */
class SimpleLexer {
    var $_regexes;
    var $_parser;
    var $_mode;
    var $_mode_handlers;
    var $_case;
    
    /**
     *    Sets up the lexer in case insensitive matching
     *    by default.
     *    @param SimpleSaxParser $parser  Handling strategy by
     *                                    reference.
     *    @param string $start            Starting handler.
     *    @param boolean $case            True for case sensitive.
     *    @access public
     */
    function SimpleLexer(&$parser, $start = "accept", $case = false) {
        $this->_case = $case;
        $this->_regexes = array();
        $this->_parser = &$parser;
        $this->_mode = &new SimpleStateStack($start);
        $this->_mode_handlers = array($start => $start);
    }
    
    /**
     *    Adds a token search pattern for a particular
     *    parsing mode. The pattern does not change the
     *    current mode.
     *    @param string $pattern      Perl style regex, but ( and )
     *                                lose the usual meaning.
     *    @param string $mode         Should only apply this
     *                                pattern when dealing with
     *                                this type of input.
     *    @access public
     */
    function addPattern($pattern, $mode = "accept") {
        if (! isset($this->_regexes[$mode])) {
            $this->_regexes[$mode] = new ParallelRegex($this->_case);
        }
        $this->_regexes[$mode]->addPattern($pattern);
        if (! isset($this->_mode_handlers[$mode])) {
            $this->_mode_handlers[$mode] = $mode;
        }
    }
    
    /**
     *    Adds a pattern that will enter a new parsing
     *    mode. Useful for entering parenthesis, strings,
     *    tags, etc.
     *    @param string $pattern      Perl style regex, but ( and )
     *                                lose the usual meaning.
     *    @param string $mode         Should only apply this
     *                                pattern when dealing with
     *                                this type of input.
     *    @param string $new_mode     Change parsing to this new
     *                                nested mode.
     *    @access public
     */
    function addEntryPattern($pattern, $mode, $new_mode) {
        if (! isset($this->_regexes[$mode])) {
            $this->_regexes[$mode] = new ParallelRegex($this->_case);
        }
        $this->_regexes[$mode]->addPattern($pattern, $new_mode);
        if (! isset($this->_mode_handlers[$new_mode])) {
            $this->_mode_handlers[$new_mode] = $new_mode;
        }
    }
    
    /**
     *    Adds a pattern that will exit the current mode
     *    and re-enter the previous one.
     *    @param string $pattern      Perl style regex, but ( and )
     *                                lose the usual meaning.
     *    @param string $mode         Mode to leave.
     *    @access public
     */
    function addExitPattern($pattern, $mode) {
        if (! isset($this->_regexes[$mode])) {
            $this->_regexes[$mode] = new ParallelRegex($this->_case);
        }
        $this->_regexes[$mode]->addPattern($pattern, "__exit");
        if (! isset($this->_mode_handlers[$mode])) {
            $this->_mode_handlers[$mode] = $mode;
        }
    }
    
    /**
     *    Adds a pattern that has a special mode. Acts as an entry
     *    and exit pattern in one go, effectively calling a special
     *    parser handler for this token only.
     *    @param string $pattern      Perl style regex, but ( and )
     *                                lose the usual meaning.
     *    @param string $mode         Should only apply this
     *                                pattern when dealing with
     *                                this type of input.
     *    @param string $special      Use this mode for this one token.
     *    @access public
     */
    function addSpecialPattern($pattern, $mode, $special) {
        if (! isset($this->_regexes[$mode])) {
            $this->_regexes[$mode] = new ParallelRegex($this->_case);
        }
        $this->_regexes[$mode]->addPattern($pattern, "_$special");
        if (! isset($this->_mode_handlers[$special])) {
            $this->_mode_handlers[$special] = $special;
        }
    }
    
    /**
     *    Adds a mapping from a mode to another handler.
     *    @param string $mode        Mode to be remapped.
     *    @param string $handler     New target handler.
     *    @access public
     */
    function mapHandler($mode, $handler) {
        $this->_mode_handlers[$mode] = $handler;
    }
    
    /**
     *    Splits the page text into tokens. Will fail
     *    if the handlers report an error or if no
     *    content is consumed. If successful then each
     *    unparsed and parsed token invokes a call to the
     *    held listener.
     *    @param string $raw        Raw HTML text.
     *    @return boolean           True on success, else false.
     *    @access public
     */
    function parse($raw) {
        if (! isset($this->_parser)) {
            return false;
        }
        $length = strlen($raw);
        while (is_array($parsed = $this->_reduce($raw))) {
            list($raw, $unmatched, $matched, $mode) = $parsed;
            if (! $this->_dispatchTokens($unmatched, $matched, $mode)) {
                return false;
            }
            if ($raw === '') {
                return true;
            }
            if (strlen($raw) == $length) {
                return false;
            }
            $length = strlen($raw);
        }
        if (! $parsed) {
            return false;
        }
        return $this->_invokeParser($raw, LEXER_UNMATCHED);
    }
    
    /**
     *    Sends the matched token and any leading unmatched
     *    text to the parser changing the lexer to a new
     *    mode if one is listed.
     *    @param string $unmatched    Unmatched leading portion.
     *    @param string $matched      Actual token match.
     *    @param string $mode         Mode after match. A boolean
     *                                false mode causes no change.
     *    @return boolean             False if there was any error
     *                                from the parser.
     *    @access private
     */
    function _dispatchTokens($unmatched, $matched, $mode = false) {
        if (! $this->_invokeParser($unmatched, LEXER_UNMATCHED)) {
            return false;
        }
        if (is_bool($mode)) {
            return $this->_invokeParser($matched, LEXER_MATCHED);
        }
        if ($this->_isModeEnd($mode)) {
            if (! $this->_invokeParser($matched, LEXER_EXIT)) {
                return false;
            }
            return $this->_mode->leave();
        }
        if ($this->_isSpecialMode($mode)) {
            $this->_mode->enter($this->_decodeSpecial($mode));
            if (! $this->_invokeParser($matched, LEXER_SPECIAL)) {
                return false;
            }
            return $this->_mode->leave();
        }
        $this->_mode->enter($mode);
        return $this->_invokeParser($matched, LEXER_ENTER);
    }
    
    /**
     *    Tests to see if the new mode is actually to leave
     *    the current mode and pop an item from the matching
     *    mode stack.
     *    @param string $mode    Mode to test.
     *    @return boolean        True if this is the exit mode.
     *    @access private
     */
    function _isModeEnd($mode) {
        return ($mode === "__exit");
    }
    
    /**
     *    Test to see if the mode is one where this mode
     *    is entered for this token only and automatically