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                         $x = strpos( strtolower( $badStr ), '</style>', $x ) + 7; // Moodle
  49                         $chr = '';
  50                     } else if ( !$is_open_tb && strtolower( substr( $badStr, $x + 1, 6 ) ) == 'script' ) {
  51                         $x = strpos( strtolower( $badStr ), '</script>', $x ) + 8;  // Moodle
  52                         $chr = '';
  53                     } else if (!$is_open_tb) {
  54                         $is_open_tb = true;
  55                     } else {
  56                         $chr = '&lt;';
  57                     }
  58                     break;
  59
  60                 case '>':
  61                     if ( !$is_open_tb || $is_open_dq || $is_open_sq ) {
  62                         $chr = '&gt;';
  63                     } else {
  64                         $is_open_tb = false;
  65                     }
  66                     break;
  67
  68                 case '"':
  69                     if ( $is_open_tb && !$is_open_dq && !$is_open_sq ) {
  70                         $is_open_dq = true;
  71                     } else if ( $is_open_tb && $is_open_dq && !$is_open_sq ) {
  72                         $is_open_dq = false;
  73                     } else {
  74                         $chr = '&quot;';
  75                     }
  76                     break;
  77
  78                 case "'":
  79                     if ( $is_open_tb && !$is_open_dq && !$is_open_sq ) {
  80                         $is_open_sq = true;
  81                     } else if ( $is_open_tb && !$is_open_dq && $is_open_sq ) {
  82                         $is_open_sq = false;
  83                     }
  84                     break;
  85             }
  86             $goodStr .= $chr;
  87         }
  88     } // Moodle
  89
  90     //now that the page is valid (I hope) for strip_tags, strip all unwanted tags
  91
  92     $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>' );
  93
  94     //strip extra whitespace except between <pre> and <textarea> tags
  95
  96     $badStr = preg_split( "/<\/?pre[^>]*>/i", $goodStr );
  97
  98     for ( $x = 0; isset($badStr[$x]) && is_string( $badStr[$x] ); $x++ ) { // Moodle: added isset() test
  99         if ( $x % 2 ) { $badStr[$x] = '<pre>'.$badStr[$x].'</pre>'; } else {
 100             $goodStr = preg_split( "/<\/?textarea[^>]*>/i", $badStr[$x] );
 101             for ( $z = 0; isset($goodStr[$z]) && is_string( $goodStr[$z] ); $z++ ) { // Moodle: added isset() test
 102                 if ( $z % 2 ) { $goodStr[$z] = '<textarea>'.$goodStr[$z].'</textarea>'; } else {
 103                     $goodStr[$z] = str_replace('  ', ' ', $goodStr[$z] );
 104                 }
 105             }
 106             $badStr[$x] = implode('',$goodStr);
 107         }
 108     }
 109
 110     $goodStr = implode('',$badStr);
 111
 112     //remove all options from select inputs
 113
 114     $goodStr = preg_replace( "/<option[^>]*>[^<]*/i", '', $goodStr );
 115
 116     //replace all tags with their text equivalents
 117
 118     $goodStr = preg_replace( "/<(\/title|hr)[^>]*>/i", "\n          --------------------\n", $goodStr );
 119
 120     $goodStr = preg_replace( "/<(h|div|p)[^>]*>/i", "\n\n", $goodStr );
 121
 122     $goodStr = preg_replace( "/<sup[^>]*>/i", '^', $goodStr );
 123
 124     $goodStr = preg_replace( "/<(ul|ol|br|dl|dt|table|caption|\/textarea|tr[^>]*>\s*<(td|th))[^>]*>/i", "\n", $goodStr );
 125
 126     $goodStr = preg_replace( "/<li[^>]*>/i", "\n� ", $goodStr );
 127
 128     $goodStr = preg_replace( "/<dd[^>]*>/i", "\n\t", $goodStr );
 129
 130     $goodStr = preg_replace( "/<(th|td)[^>]*>/i", "\t", $goodStr );
 131
 132  // $goodStr = preg_replace( "/<a[^>]* href=(\"((?!\"|#|javascript:)[^\"#]*)(\"|#)|'((?!'|#|javascript:)[^'#]*)('|#)|((?!'|\"|>|#|javascript:)[^#\"'> ]*))[^>]*>/i", "[LINK: $2$4$6] ", $goodStr );   // Moodle
 133     $goodStr = preg_replace( "/<a[^>]* href=(\"((?!\"|#|javascript:)[^\"#]*)(\"|#)|'((?!'|#|javascript:)[^'#]*)('|#)|((?!'|\"|>|#|javascript:)[^#\"'> ]*))[^>]*>([^<]*)<\/a>/i", "$7 [$2$4$6]", $goodStr );
 134
 135     // $goodStr = preg_replace( "/<img[^>]* alt=(\"([^\"]+)\"|'([^']+)'|([^\"'> ]+))[^>]*>/i", "[IMAGE: $2$3$4] ", $goodStr );   // Moodle
 136     $goodStr = preg_replace( "/<img[^>]* alt=(\"([^\"]+)\"|'([^']+)'|([^\"'> ]+))[^>]*>/i", "[$2$3$4] ", $goodStr );
 137
 138     $goodStr = preg_replace( "/<form[^>]* action=(\"([^\"]+)\"|'([^']+)'|([^\"'> ]+))[^>]*>/i", "\n[FORM: $2$3$4] ", $goodStr );
 139
 140     $goodStr = preg_replace( "/<(input|textarea|button|select)[^>]*>/i", "[INPUT] ", $goodStr );
 141
 142     //strip all remaining tags (mostly closing tags)
 143
 144     $goodStr = strip_tags( $goodStr );
 145
 146     //convert HTML entities
 147
 148     $goodStr = strtr( $goodStr, array_flip( get_html_translation_table( HTML_ENTITIES ) ) );
 149
 150     preg_replace( "/&#(\d+);/me", "chr('$1')", $goodStr );
 151
 152     //wordwrap
 153
 154     // $goodStr = wordwrap( $goodStr );   // Moodle
 155     $goodStr = wordwrap( $goodStr, 78 );
 156
 157     //make sure there are no more than 3 linebreaks in a row and trim whitespace
 158     $goodStr = preg_replace("/\r\n?|\f/", "\n", $goodStr);
 159     $goodStr = preg_replace("/\n(\s*\n){2}/", "\n\n\n", $goodStr);
 160     $goodStr = preg_replace("/[ \t]+(\n|$)/", "$1", $goodStr);
 161     $goodStr = preg_replace("/^\n*|\n*$/", '', $goodStr);
 162
 163     return $goodStr;
 164
 165 }
 166
 167 ?>