X-Git-Url: https://git.donarmstrong.com/?a=blobdiff_plain;f=program%2Flib%2Fhtml2text.php;h=9fc96eac7bb519aeb951639babf4e1e5242d721c;hb=76507f7c63a660742e76889ad6e3919f3dde3bb0;hp=e2a5b241eefc8c2f2057a9f82a7733e95c03bacd;hpb=3adad46e27086084a8b28a32fc4fbc953dbfef6c;p=roundcube.git

diff --git a/program/lib/html2text.php b/program/lib/html2text.php
index e2a5b24..9fc96ea 100644
--- a/program/lib/html2text.php
+++ b/program/lib/html2text.php
@@ -145,10 +145,8 @@ class html2text
     var $search = array(
         "/\r/",                                  // Non-legal carriage return
         "/[\n\t]+/",                             // Newlines and tabs
-        '/[ ]{2,}/',                             // Runs of spaces, pre-handling
         '/<script[^>]*>.*?<\/script>/i',         // <script>s -- which strip_tags supposedly has problems with
         '/<style[^>]*>.*?<\/style>/i',           // <style>s -- which strip_tags supposedly has problems with
-        //'/<!-- .* -->/',                         // Comments -- which strip_tags might have problem a with
         '/<p[^>]*>/i',                           // <P>
         '/<br[^>]*>/i',                          // <br>
         '/<i[^>]*>(.*?)<\/i>/i',                 // <i>
@@ -158,26 +156,10 @@ class html2text
         '/<li[^>]*>(.*?)<\/li>/i',               // <li> and </li>
         '/<li[^>]*>/i',                          // <li>
         '/<hr[^>]*>/i',                          // <hr>
+        '/<div[^>]*>/i',                         // <div>
         '/(<table[^>]*>|<\/table>)/i',           // <table> and </table>
         '/(<tr[^>]*>|<\/tr>)/i',                 // <tr> and </tr>
         '/<td[^>]*>(.*?)<\/td>/i',               // <td> and </td>
-        '/&(nbsp|#160);/i',                      // Non-breaking space
-        '/&(quot|rdquo|ldquo|#8220|#8221|#147|#148);/i',
-		                                         // Double quotes
-        '/&(apos|rsquo|lsquo|#8216|#8217);/i',   // Single quotes
-        '/&gt;/i',                               // Greater-than
-        '/&lt;/i',                               // Less-than
-        '/&(amp|#38);/i',                        // Ampersand
-        '/&(copy|#169);/i',                      // Copyright
-        '/&(trade|#8482|#153);/i',               // Trademark
-        '/&(reg|#174);/i',                       // Registered
-        '/&(mdash|#151|#8212);/i',               // mdash
-        '/&(ndash|minus|#8211|#8722);/i',        // ndash
-        '/&(bull|#149|#8226);/i',                // Bullet
-        '/&(pound|#163);/i',                     // Pound sign
-        '/&(euro|#8364);/i',                     // Euro sign
-        '/&[^&;]+;/i',                           // Unknown/unhandled entities
-        '/[ ]{2,}/'                              // Runs of spaces, post-handling
     );
 
     /**
@@ -190,11 +172,9 @@ class html2text
     var $replace = array(
         '',                                     // Non-legal carriage return
         ' ',                                    // Newlines and tabs
-        ' ',                                    // Runs of spaces, pre-handling
         '',                                     // <script>s -- which strip_tags supposedly has problems with
         '',                                     // <style>s -- which strip_tags supposedly has problems with
-        //'',                                     // Comments -- which strip_tags might have problem a with
-        "\n\n",                               // <P>
+        "\n\n",                                 // <P>
         "\n",                                   // <br>
         '_\\1_',                                // <i>
         '_\\1_',                                // <em>
@@ -202,16 +182,53 @@ class html2text
         "\n\n",                                 // <ol> and </ol>
         "\t* \\1\n",                            // <li> and </li>
         "\n\t* ",                               // <li>
-    	"\n-------------------------\n",        // <hr>
-    	"\n\n",                                 // <table> and </table>
+        "\n-------------------------\n",        // <hr>
+        "<div>\n",                              // <div>
+        "\n\n",                                 // <table> and </table>
         "\n",                                   // <tr> and </tr>
         "\t\t\\1\n",                            // <td> and </td>
+    );
+
+    /**
+     *  List of preg* regular expression patterns to search for,
+     *  used in conjunction with $ent_replace.
+     *
+     *  @var array $ent_search
+     *  @access public
+     *  @see $ent_replace
+     */
+    var $ent_search = array(
+        '/&(nbsp|#160);/i',                      // Non-breaking space
+        '/&(quot|rdquo|ldquo|#8220|#8221|#147|#148);/i',
+		                                         // Double quotes
+        '/&(apos|rsquo|lsquo|#8216|#8217);/i',   // Single quotes
+        '/&gt;/i',                               // Greater-than
+        '/&lt;/i',                               // Less-than
+        '/&(copy|#169);/i',                      // Copyright
+        '/&(trade|#8482|#153);/i',               // Trademark
+        '/&(reg|#174);/i',                       // Registered
+        '/&(mdash|#151|#8212);/i',               // mdash
+        '/&(ndash|minus|#8211|#8722);/i',        // ndash
+        '/&(bull|#149|#8226);/i',                // Bullet
+        '/&(pound|#163);/i',                     // Pound sign
+        '/&(euro|#8364);/i',                     // Euro sign
+        '/&(amp|#38);/i',                        // Ampersand: see _converter()
+        '/[ ]{2,}/',                             // Runs of spaces, post-handling
+    );
+
+    /**
+     *  List of pattern replacements corresponding to patterns searched.
+     *
+     *  @var array $ent_replace
+     *  @access public
+     *  @see $ent_search
+     */
+    var $ent_replace = array(
         ' ',                                    // Non-breaking space
         '"',                                    // Double quotes
         "'",                                    // Single quotes
         '>',
         '<',
-        '&',
         '(c)',
         '(tm)',
         '(R)',
@@ -220,8 +237,8 @@ class html2text
         '*',
         '£',
         'EUR',                                  // Euro sign. € ?
-        '',                                     // Unknown/unhandled entities
-        ' '                                     // Runs of spaces, post-handling
+        '|+|amp|+|',                            // Ampersand: see _converter()
+        ' ',                                    // Runs of spaces, post-handling
     );
 
     /**
@@ -249,11 +266,11 @@ class html2text
     *  @see $pre_replace
     */
     var $pre_search = array(
-	"/\n/",
-	"/\t/",
-	'/ /',
-	'/<pre[^>]*>/',
-	'/<\/pre>/'
+        "/\n/",
+        "/\t/",
+        '/ /',
+        '/<pre[^>]*>/',
+        '/<\/pre>/'
     );
 
     /**
@@ -264,11 +281,11 @@ class html2text
      *  @see $pre_search
      */
     var $pre_replace = array(
-	'<br>',
-	'&nbsp;&nbsp;&nbsp;&nbsp;',
-	'&nbsp;',
-	'',
-	''
+        '<br>',
+        '&nbsp;&nbsp;&nbsp;&nbsp;',
+        '&nbsp;',
+        '',
+        ''
     );
 
     /**
@@ -305,7 +322,7 @@ class html2text
      *  @see _build_link_list()
      */
     var $_link_list = '';
-    
+
     /**
      *  Number of valid links detected in the text, used for plain text
      *  display (rendered similar to footnotes).
@@ -316,15 +333,15 @@ class html2text
      */
     var $_link_count = 0;
 
-    /** 
-     * Boolean flag, true if a table of link URLs should be listed after the text. 
-     *  
-     * @var boolean $_do_links 
-     * @access private 
-     * @see html2text() 
+    /**
+     * Boolean flag, true if a table of link URLs should be listed after the text.
+     *
+     * @var boolean $_do_links
+     * @access private
+     * @see html2text()
      */
     var $_do_links = true;
- 
+
     /**
      *  Constructor.
      *
@@ -344,10 +361,10 @@ class html2text
         if ( !empty($source) ) {
             $this->set_html($source, $from_file);
         }
-	
+
         $this->set_base_url();
-	$this->_do_links = $do_links;
-	$this->width = $width;
+        $this->_do_links = $do_links;
+        $this->width = $width;
     }
 
     /**
@@ -361,10 +378,10 @@ class html2text
     function set_html( $source, $from_file = false )
     {
         if ( $from_file && file_exists($source) ) {
-	    $this->html = file_get_contents($source); 
+            $this->html = file_get_contents($source); 
         }
         else
-	    $this->html = $source;
+            $this->html = $source;
 
         $this->_converted = false;
     }
@@ -447,12 +464,7 @@ class html2text
     }
 
     /**
-     *  Workhorse function that does actual conversion.
-     *
-     *  First performs custom tag replacement specified by $search and
-     *  $replace arrays. Then strips any remaining HTML tags, reduces whitespace
-     *  and newlines to a readable format, and word wraps the text to
-     *  $width characters.
+     *  Workhorse function that does actual conversion (calls _converter() method).
      *
      *  @access private
      *  @return void
@@ -465,27 +477,68 @@ class html2text
 
         $text = trim(stripslashes($this->html));
 
-	// Convert <PRE>
+        // Convert HTML to TXT
+        $this->_converter($text);
+
+        // Add link list
+        if ( !empty($this->_link_list) ) {
+            $text .= "\n\nLinks:\n------\n" . $this->_link_list;
+        }
+
+        $this->text = $text;
+
+        $this->_converted = true;
+    }
+
+    /**
+     *  Workhorse function that does actual conversion.
+     *
+     *  First performs custom tag replacement specified by $search and
+     *  $replace arrays. Then strips any remaining HTML tags, reduces whitespace
+     *  and newlines to a readable format, and word wraps the text to
+     *  $width characters.
+     *
+     *  @param string Reference to HTML content string
+     *
+     *  @access private
+     *  @return void
+     */
+    function _converter(&$text)
+    {
+        // Convert <BLOCKQUOTE> (before PRE!)
+        $this->_convert_blockquotes($text);
+
+        // Convert <PRE>
         $this->_convert_pre($text);
 
-        // Run our defined search-and-replace
+        // Run our defined tags search-and-replace
         $text = preg_replace($this->search, $this->replace, $text);
-        $text = preg_replace_callback($this->callback_search, array('html2text', '_preg_callback'), $text);
 
-	// Replace known html entities
-	$text = html_entity_decode($text, ENT_COMPAT, 'UTF-8');
+        // Run our defined tags search-and-replace with callback
+        $text = preg_replace_callback($this->callback_search, array('html2text', '_preg_callback'), $text);
 
         // Strip any other HTML tags
         $text = strip_tags($text, $this->allowed_tags);
 
+        // Run our defined entities/characters search-and-replace
+        $text = preg_replace($this->ent_search, $this->ent_replace, $text);
+
+        // Replace known html entities
+        $text = html_entity_decode($text, ENT_COMPAT, 'UTF-8');
+
+        // Remove unknown/unhandled entities (this cannot be done in search-and-replace block)
+        $text = preg_replace('/&([a-zA-Z0-9]{2,6}|#[0-9]{2,4});/', '', $text);
+
+        // Convert "|+|amp|+|" into "&", need to be done after handling of unknown entities
+        // This properly handles situation of "&amp;quot;" in input string
+        $text = str_replace('|+|amp|+|', '&', $text);
+
         // Bring down number of empty lines to 2 max
         $text = preg_replace("/\n\s+\n/", "\n\n", $text);
         $text = preg_replace("/[\n]{3,}/", "\n\n", $text);
 
-        // Add link list
-        if ( !empty($this->_link_list) ) {
-            $text .= "\n\nLinks:\n------\n" . $this->_link_list;
-        }
+        // remove leading empty lines (can be produced by eg. P tag on the beginning)
+        $text = ltrim($text, "\n");
 
         // Wrap the text to a readable format
         // for PHP versions >= 4.0.2. Default width is 75
@@ -493,10 +546,6 @@ class html2text
         if ( $this->width > 0 ) {
         	$text = wordwrap($text, $this->width);
         }
-
-        $this->text = $text;
-
-        $this->_converted = true;
     }
 
     /**
@@ -514,20 +563,20 @@ class html2text
      */
     function _build_link_list( $link, $display )
     {
-	if ( !$this->_do_links ) return $display;
-	
-	if ( substr($link, 0, 7) == 'http://' || substr($link, 0, 8) == 'https://' ||
-             substr($link, 0, 7) == 'mailto:' ) {
+	    if ( !$this->_do_links )
+	        return $display;
+
+	    if ( preg_match('!^(https?://|mailto:)!', $link) ) {
             $this->_link_count++;
-            $this->_link_list .= "[" . $this->_link_count . "] $link\n";
+            $this->_link_list .= '[' . $this->_link_count . "] $link\n";
             $additional = ' [' . $this->_link_count . ']';
-		} elseif ( substr($link, 0, 11) == 'javascript:' ) {
-			// Don't count the link; ignore it
-			$additional = '';
+	    } elseif ( substr($link, 0, 11) == 'javascript:' ) {
+		    // Don't count the link; ignore it
+		    $additional = '';
 		// what about href="#anchor" ?
         } else {
             $this->_link_count++;
-            $this->_link_list .= "[" . $this->_link_count . "] " . $this->url;
+            $this->_link_list .= '[' . $this->_link_count . '] ' . $this->url;
             if ( substr($link, 0, 1) != '/' ) {
                 $this->_link_list .= '/';
             }
@@ -537,7 +586,7 @@ class html2text
 
         return $display . $additional;
     }
-    
+
     /**
      *  Helper function for PRE body conversion.
      *
@@ -546,11 +595,73 @@ class html2text
      */
     function _convert_pre(&$text)
     {
-	while(preg_match('/<pre[^>]*>(.*)<\/pre>/ismU', $text, $matches))
-	{
-	    $result = preg_replace($this->pre_search, $this->pre_replace, $matches[1]);
-	    $text = preg_replace('/<pre[^>]*>.*<\/pre>/ismU', '<div><br>' . $result . '<br></div>', $text, 1);
-	}
+        // get the content of PRE element
+        while (preg_match('/<pre[^>]*>(.*)<\/pre>/ismU', $text, $matches)) {
+            // convert the content
+            $this->pre_content = sprintf('<div><br>%s<br></div>',
+                preg_replace($this->pre_search, $this->pre_replace, $matches[1]));
+            // replace the content (use callback because content can contain $0 variable)
+            $text = preg_replace_callback('/<pre[^>]*>.*<\/pre>/ismU', 
+                array('html2text', '_preg_pre_callback'), $text, 1);
+            // free memory
+            $this->pre_content = '';
+        }
+    }
+
+    /**
+     *  Helper function for BLOCKQUOTE body conversion.
+     *
+     *  @param string HTML content
+     *  @access private
+     */
+    function _convert_blockquotes(&$text)
+    {
+        if (preg_match_all('/<\/*blockquote[^>]*>/i', $text, $matches, PREG_OFFSET_CAPTURE)) {
+            $level = 0;
+            $diff = 0;
+            foreach ($matches[0] as $m) {
+                if ($m[0][0] == '<' && $m[0][1] == '/') {
+                    $level--;
+                    if ($level < 0) {
+                        $level = 0; // malformed HTML: go to next blockquote
+                    }
+                    else if ($level > 0) {
+                        // skip inner blockquote
+                    }
+                    else {
+                        $end  = $m[1];
+                        $len  = $end - $taglen - $start;
+                        // Get blockquote content
+                        $body = substr($text, $start + $taglen - $diff, $len);
+
+                        // Set text width
+                        $p_width = $this->width;
+                        if ($this->width > 0) $this->width -= 2;
+                        // Convert blockquote content
+                        $body = trim($body);
+                        $this->_converter($body);
+                        // Add citation markers and create PRE block
+                        $body = preg_replace('/((^|\n)>*)/', '\\1> ', trim($body));
+                        $body = '<pre>' . htmlspecialchars($body) . '</pre>';
+                        // Re-set text width
+                        $this->width = $p_width;
+                        // Replace content
+                        $text = substr($text, 0, $start - $diff)
+                            . $body . substr($text, $end + strlen($m[0]) - $diff);
+
+                        $diff = $len + $taglen + strlen($m[0]) - strlen($body);
+                        unset($body);
+                    }
+                }
+                else {
+                    if ($level == 0) {
+                        $start = $m[1];
+                        $taglen = strlen($m[0]);
+                    }
+                    $level ++;
+                }
+            }
+        }
     }
 
     /**
@@ -558,38 +669,46 @@ class html2text
      *
      *  @param  array PREG matches
      *  @return string
-     *  @access private
      */
-    function _preg_callback($matches)
+    private function _preg_callback($matches)
+    {
+        switch($matches[1]) {
+        case 'b':
+        case 'strong':
+            return $this->_strtoupper($matches[2]);
+        case 'th':
+            return $this->_strtoupper("\t\t". $matches[2] ."\n");
+        case 'h':
+            return $this->_strtoupper("\n\n". $matches[2] ."\n\n");
+        case 'a':
+            // Remove spaces in URL (#1487805)
+            $url = str_replace(' ', '', $matches[3]);
+            return $this->_build_link_list($url, $matches[4]);
+        }
+    }
+
+    /**
+     *  Callback function for preg_replace_callback use in PRE content handler.
+     *
+     *  @param  array PREG matches
+     *  @return string
+     */
+    private function _preg_pre_callback($matches)
     {
-	switch($matches[1])
-	{
-	    case 'b':
-	    case 'strong':
-		return $this->_strtoupper($matches[2]);
-	    case 'hr':
-		return $this->_strtoupper("\t\t". $matches[2] ."\n");
-	    case 'h':
-		return $this->_strtoupper("\n\n". $matches[2] ."\n\n");
-	    case 'a':
-    	        return $this->_build_link_list($matches[3], $matches[4]);
-	}
+        return $this->pre_content;
     }
-    
+
     /**
      *  Strtoupper multibyte wrapper function
      *
      *  @param  string
      *  @return string
-     *  @access private
      */
-    function _strtoupper($str)
+    private function _strtoupper($str)
     {
-	if (function_exists('mb_strtoupper'))
-    	    return mb_strtoupper($str);
-    	else
-	    return strtoupper($str);
+        if (function_exists('mb_strtoupper'))
+            return mb_strtoupper($str);
+        else
+            return strtoupper($str);
     }
 }
-
-?>