lib/html2text.php

   1 <?php
   2
   3 /***************************************************************
   4  * Library to convert HTML into an approximate text equivalent *
   5  ***************************************************************
   6
   7   Version: 1.0.3  (with modifications)
   8   Copyright 2003 Mark Wilton-Jones
   9   License: HowToCreate script license with written permission
  10   URL: http://www.howtocreate.co.uk/php/
  11
  12   For full details about the script and to get the latest version,
  13   please see the HowToCreate web site above.
  14
  15   This version contains modifications for Moodle.  In each case the
  16   lines are marked with "Moodle", so you can  see what has changed.
  17
  18   ********************************************************************/
  19
  20 function html2text( $badStr ) {
  21
  22     $is_open_tb = false;
  23     $is_open_dq = false;
  24     $is_open_sq = false;
  25
  26     //remove comments
  27
  28     while (substr_count($badStr, '<!--') &&
  29            substr_count($badStr, '-->') &&
  30            strpos($badStr, '-->', strpos($badStr, '<!--' ) ) > strpos( $badStr, '<!--' ) ) {
  31            $badStr = substr( $badStr, 0, strpos( $badStr, '<!--' ) ) .
  32                      substr( $badStr, strpos( $badStr, '-->',
  33                      strpos( $badStr, '<!--' ) ) + 3 );
  34     }
  35
  36     //now make sure all HTML tags are correctly written (> not in between quotes)
  37
  38     $len = strlen($badStr); // Moodle
  39     $chr = $badStr{0}; // Moodle
  40     $goodStr = ''; // Moodle
  41
  42     if ($len > 0) { // Moodle
  43         for ($x=0; $x < $len; $x++ ) { // Moodle
  44             $chr = $badStr{$x}; //take each letter in turn and check if that character is permitted there
  45             switch ( $chr ) {
  46                 case '<':
  47                     if ( !$is_open_tb && strtolower( substr( $badStr, $x + 1, 5 ) ) == 'style' ) {
  48                         $badStr = substr( $badStr, 0, $x ) .
  49                                   substr( $badStr, strpos( strtolower( $badStr ), '</style>', $x ) + 7 );
  50                         $chr = '';
  51                     } else if ( !$is_open_tb && strtolower( substr( $badStr, $x + 1, 6 ) ) == 'script' ) {
  52                         $badStr = substr( $badStr, 0, $x ) .
  53                                   substr( $badStr, strpos( strtolower( $badStr ), '</script>', $x ) + 8 );
  54                         $chr = '';
  55                     } else if (!$is_open_tb) {
  56                         $is_open_tb = true;
  57                     } else {
  58                         $chr = '&lt;';
  59                     }
  60                     break;
  61
  62                 case '>':
  63                     if ( !$is_open_tb || $is_open_dq || $is_open_sq ) {
  64                         $chr = '&gt;';
  65                     } else {
  66                         $is_open_tb = false;
  67                     }
  68                     break;
  69
  70                 case '"':
  71                     if ( $is_open_tb && !$is_open_dq && !$is_open_sq ) {
  72                         $is_open_dq = true;
  73                     } else if ( $is_open_tb && $is_open_dq && !$is_open_sq ) {
  74                         $is_open_dq = false;
  75                     } else {
  76                         $chr = '&quot;';
  77                     }
  78                     break;
  79
  80                 case "'":
  81                     if ( $is_open_tb && !$is_open_dq && !$is_open_sq ) {
  82                         $is_open_sq = true;
  83                     } else if ( $is_open_tb && !$is_open_dq && $is_open_sq ) {
  84                         $is_open_sq = false;
  85                     }
  86                     break;
  87             }
  88             $goodStr .= $chr;
  89         }
  90     } // Moodle
  91
  92     //now that the page is valid (I hope) for strip_tags, strip all unwanted tags
  93
  94     $goodStr = strip_tags( $goodStr, '<title><hr><h1><h2><h3><h4><h5><h6><div><p><pre><sup><ul><ol><br><dl><dt><table><caption><tr><li><dd><th><td><a><area><img><form><input><textarea><button><select><option>' );
  95
  96     //strip extra whitespace except between <pre> and <textarea> tags
  97
  98     $badStr = preg_split( "/<\/?pre[^>]*>/i", $goodStr );
  99
 100     for ( $x = 0; isset($badStr[$x]) && is_string( $badStr[$x] ); $x++ ) { // Moodle: added isset() test
 101         if ( $x % 2 ) { $badStr[$x] = '<pre>'.$badStr[$x].'</pre>'; } else {
 102             $goodStr = preg_split( "/<\/?textarea[^>]*>/i", $badStr[$x] );
 103             for ( $z = 0; isset($goodStr[$z]) && is_string( $goodStr[$z] ); $z++ ) { // Moodle: added isset() test
 104                 if ( $z % 2 ) { $goodStr[$z] = '<textarea>'.$goodStr[$z].'</textarea>'; } else {
 105                     $goodStr[$z] = str_replace('  ', ' ', $goodStr[$z] );
 106                 }
 107             }
 108             $badStr[$x] = implode('',$goodStr);
 109         }
 110     }
 111
 112     $goodStr = implode('',$badStr);
 113
 114     //remove all options from select inputs
 115
 116     $goodStr = preg_replace( "/<option[^>]*>[^<]*/i", '', $goodStr );
 117
 118     //replace all tags with their text equivalents
 119
 120     $goodStr = preg_replace( "/<(\/title|hr)[^>]*>/i", "\n          --------------------\n", $goodStr );
 121
 122     $goodStr = preg_replace( "/<(h|div|p)[^>]*>/i", "\n\n", $goodStr );
 123
 124     $goodStr = preg_replace( "/<sup[^>]*>/i", '^', $goodStr );
 125
 126     $goodStr = preg_replace( "/<(ul|ol|br|dl|dt|table|caption|\/textarea|tr[^>]*>\s*<(td|th))[^>]*>/i", "\n", $goodStr );
 127
 128     $goodStr = preg_replace( "/<li[^>]*>/i", "\n· ", $goodStr );
 129
 130     $goodStr = preg_replace( "/<dd[^>]*>/i", "\n\t", $goodStr );
 131
 132     $goodStr = preg_replace( "/<(th|td)[^>]*>/i", "\t", $goodStr );
 133
 134  // $goodStr = preg_replace( "/<a[^>]* href=(\"((?!\"|#|javascript:)[^\"#]*)(\"|#)|'((?!'|#|javascript:)[^'#]*)('|#)|((?!'|\"|>|#|javascript:)[^#\"'> ]*))[^>]*>/i", "[LINK: $2$4$6] ", $goodStr );   // Moodle
 135     $goodStr = preg_replace( "/<a[^>]* href=(\"((?!\"|#|javascript:)[^\"#]*)(\"|#)|'((?!'|#|javascript:)[^'#]*)('|#)|((?!'|\"|>|#|javascript:)[^#\"'> ]*))[^>]*>([^<]*)<\/a>/i", "$7 [$2$4$6]", $goodStr );
 136
 137     // $goodStr = preg_replace( "/<img[^>]* alt=(\"([^\"]+)\"|'([^']+)'|([^\"'> ]+))[^>]*>/i", "[IMAGE: $2$3$4] ", $goodStr );   // Moodle
 138     $goodStr = preg_replace( "/<img[^>]* alt=(\"([^\"]+)\"|'([^']+)'|([^\"'> ]+))[^>]*>/i", "[$2$3$4] ", $goodStr );
 139
 140     $goodStr = preg_replace( "/<form[^>]* action=(\"([^\"]+)\"|'([^']+)'|([^\"'> ]+))[^>]*>/i", "\n[FORM: $2$3$4] ", $goodStr );
 141
 142     $goodStr = preg_replace( "/<(input|textarea|button|select)[^>]*>/i", "[INPUT] ", $goodStr );
 143
 144     //strip all remaining tags (mostly closing tags)
 145
 146     $goodStr = strip_tags( $goodStr );
 147
 148     //convert HTML entities
 149
 150     $goodStr = strtr( $goodStr, array_flip( get_html_translation_table( HTML_ENTITIES ) ) );
 151
 152     preg_replace( "/&#(\d+);/me", "chr('$1')", $goodStr );
 153
 154     //wordwrap
 155
 156     // $goodStr = wordwrap( $goodStr );   // Moodle
 157     $goodStr = wordwrap( $goodStr, 78 );
 158
 159     //make sure there are no more than 3 linebreaks in a row and trim whitespace
 160     $goodStr = preg_replace("/\r\n?|\f/", "\n", $goodStr);
 161     $goodStr = preg_replace("/\n(\s*\n){2}/", "\n\n\n", $goodStr);
 162     $goodStr = preg_replace("/[ \t]+(\n|$)/", "$1", $goodStr);
 163     $goodStr = preg_replace("/^\n*|\n*$/", '', $goodStr);
 164
 165     return $goodStr;
 166
 167 }
 168
 169 ?>