directlex.php

很棒的在线教学系统

<?php

require_once 'HTMLPurifier/Lexer.php';

HTMLPurifier_ConfigSchema::define(
    'Core', 'DirectLexLineNumberSyncInterval', 0, 'int', '
<p>
  Specifies the number of tokens the DirectLex line number tracking
  implementations should process before attempting to resyncronize the
  current line count by manually counting all previous new-lines. When
  at 0, this functionality is disabled. Lower values will decrease
  performance, and this is only strictly necessary if the counting
  algorithm is buggy (in which case you should report it as a bug).
  This has no effect when %Core.MaintainLineNumbers is disabled or DirectLex is
  not being used. This directive has been available since 2.0.0.
</p>
');

/**
 * Our in-house implementation of a parser.
 * 
 * A pure PHP parser, DirectLex has absolutely no dependencies, making
 * it a reasonably good default for PHP4.  Written with efficiency in mind,
 * it can be four times faster than HTMLPurifier_Lexer_PEARSax3, although it
 * pales in comparison to HTMLPurifier_Lexer_DOMLex.
 * 
 * @todo Reread XML spec and document differences.
 */
class HTMLPurifier_Lexer_DirectLex extends HTMLPurifier_Lexer
{
    
    /**
     * Whitespace characters for str(c)spn.
     * @protected
     */
    var $_whitespace = "\x20\x09\x0D\x0A";
    
    /**
     * Callback function for script CDATA fudge
     * @param $matches, in form of array(opening tag, contents, closing tag)
     * @static
     */
    function scriptCallback($matches) {
        return $matches[1] . htmlspecialchars($matches[2], ENT_COMPAT, 'UTF-8') . $matches[3];
    }
    
    function tokenizeHTML($html, $config, &$context) {
        
        // special normalization for script tags without any armor
        // our "armor" heurstic is a < sign any number of whitespaces after
        // the first script tag
        if ($config->get('HTML', 'Trusted')) {
            $html = preg_replace_callback('#(<script[^>]*>)(\s*[^<].+?)(</script>)#si',
                array('HTMLPurifier_Lexer_DirectLex', 'scriptCallback'), $html);
        }
        
        $html = $this->normalize($html, $config, $context);
        
        $cursor = 0; // our location in the text
        $inside_tag = false; // whether or not we're parsing the inside of a tag
        $array = array(); // result array
        
        $maintain_line_numbers = $config->get('Core', 'MaintainLineNumbers');
        
        if ($maintain_line_numbers === null) {
            // automatically determine line numbering by checking
            // if error collection is on
            $maintain_line_numbers = $config->get('Core', 'CollectErrors');
        }
        
        if ($maintain_line_numbers) $current_line = 1;
        else $current_line = false;
        $context->register('CurrentLine', $current_line);
        $nl = "\n";
        // how often to manually recalculate. This will ALWAYS be right,
        // but it's pretty wasteful. Set to 0 to turn off
        $synchronize_interval = $config->get('Core', 'DirectLexLineNumberSyncInterval'); 
        
        $e = false;
        if ($config->get('Core', 'CollectErrors')) {
            $e =& $context->get('ErrorCollector');
        }
        
        // infinite loop protection
        // has to be pretty big, since html docs can be big
        // we're allow two hundred thousand tags... more than enough?
        // NOTE: this is also used for synchronization, so watch out
        $loops = 0;
        
        while(true) {
            
            // infinite loop protection
            if (++$loops > 200000) return array();
            
            // recalculate lines
            if (
                $maintain_line_numbers && // line number tracking is on
                $synchronize_interval &&  // synchronization is on
                $cursor > 0 &&            // cursor is further than zero
                $loops % $synchronize_interval === 0 // time to synchronize!
            ) {
                $current_line = 1 + $this->substrCount($html, $nl, 0, $cursor);
            }
            
            $position_next_lt = strpos($html, '<', $cursor);
            $position_next_gt = strpos($html, '>', $cursor);
            
            // triggers on "<b>asdf</b>" but not "asdf <b></b>"
            // special case to set up context
            if ($position_next_lt === $cursor) {
                $inside_tag = true;
                $cursor++;
            }
            
            if (!$inside_tag && $position_next_lt !== false) {
                // We are not inside tag and there still is another tag to parse
                $token = new
                    HTMLPurifier_Token_Text(
                        $this->parseData(
                            substr(
                                $html, $cursor, $position_next_lt - $cursor
                            )
                        )
                    );
                if ($maintain_line_numbers) {
                    $token->line = $current_line;
                    $current_line += $this->substrCount($html, $nl, $cursor, $position_next_lt - $cursor);
                }
                $array[] = $token;
                $cursor  = $position_next_lt + 1;
                $inside_tag = true;
                continue;
            } elseif (!$inside_tag) {
                // We are not inside tag but there are no more tags
                // If we're already at the end, break
                if ($cursor === strlen($html)) break;
                // Create Text of rest of string
                $token = new
                    HTMLPurifier_Token_Text(
                        $this->parseData(
                            substr(
                                $html, $cursor
                            )
                        )
                    );
                if ($maintain_line_numbers) $token->line = $current_line;
                $array[] = $token;
                break;
            } elseif ($inside_tag && $position_next_gt !== false) {
                // We are in tag and it is well formed
                // Grab the internals of the tag
                $strlen_segment = $position_next_gt - $cursor;
                
                if ($strlen_segment < 1) {
                    // there's nothing to process!
                    $token = new HTMLPurifier_Token_Text('<');
                    $cursor++;
                    continue;
                }
                
                $segment = substr($html, $cursor, $strlen_segment);
                
                if ($segment === false) {
                    // somehow, we attempted to access beyond the end of
                    // the string, defense-in-depth, reported by Nate Abele
                    break;
                }
                
                // Check if it's a comment
                if (
                    strncmp('!--', $segment, 3) === 0
                ) {
                    // re-determine segment length, looking for -->
                    $position_comment_end = strpos($html, '-->', $cursor);
                    if ($position_comment_end === false) {
                        // uh oh, we have a comment that extends to
                        // infinity. Can't be helped: set comment
                        // end position to end of string
                        if ($e) $e->send(E_WARNING, 'Lexer: Unclosed comment');
                        $position_comment_end = strlen($html);
                        $end = true;
                    } else {
                        $end = false;
                    }
                    $strlen_segment = $position_comment_end - $cursor;
                    $segment = substr($html, $cursor, $strlen_segment);
                    $token = new HTMLPurifier_Token_Comment(substr($segment, 3));
                    if ($maintain_line_numbers) {
                        $token->line = $current_line;
                        $current_line += $this->substrCount($html, $nl, $cursor, $strlen_segment);
                    }
                    $array[] = $token;
                    $cursor = $end ? $position_comment_end : $position_comment_end + 3;
                    $inside_tag = false;
                    continue;
                }
                
                // Check if it's an end tag
                $is_end_tag = (strpos($segment,'/') === 0);
                if ($is_end_tag) {
                    $type = substr($segment, 1);
                    $token = new HTMLPurifier_Token_End($type);
                    if ($maintain_line_numbers) {
                        $token->line = $current_line;
                        $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
                    }
                    $array[] = $token;
                    $inside_tag = false;
                    $cursor = $position_next_gt + 1;
                    continue;
                }
                
                // Check leading character is alnum, if not, we may
                // have accidently grabbed an emoticon. Translate into
                // text and go our merry way
                if (!ctype_alpha($segment[0])) {
                    // XML:  $segment[0] !== '_' && $segment[0] !== ':'
                    if ($e) $e->send(E_NOTICE, 'Lexer: Unescaped lt');
                    $token = new
                        HTMLPurifier_Token_Text(
                            '<' .
                            $this->parseData(
                                $segment
                            ) . 
                            '>'
                        );
                    if ($maintain_line_numbers) {
                        $token->line = $current_line;
                        $current_line += $this->substrCount($html, $nl, $cursor, $position_next_gt - $cursor);
                    }
                    $array[] = $token;
                    $cursor = $position_next_gt + 1;
                    $inside_tag = false;
                    continue;
                }
                
                // Check if it is explicitly self closing, if so, remove
                // trailing slash. Remember, we could have a tag like <br>, so
                // any later token processing scripts must convert improperly
                // classified EmptyTags from StartTags.
                $is_self_closing = (strrpos($segment,'/') === $strlen_segment-1);
                if ($is_self_closing) {
                    $strlen_segment--;
                    $segment = substr($segment, 0, $strlen_segment);
                }
                
                // Check if there are any attributes
                $position_first_space = strcspn($segment, $this->_whitespace);