Fix handling of links with various URI schemes e.g. "skype:" (#1488106)

Fix handling of links inside PRE elements on html to text conversion
Fix indexing of links on html to text conversion
pull/13/head
Aleksander Machniak 12 years ago
parent f8c96f737c
commit 8c188058cf

@ -1,6 +1,9 @@
CHANGELOG Roundcube Webmail CHANGELOG Roundcube Webmail
=========================== ===========================
- Fix handling of links with various URI schemes e.g. "skype:" (#1488106)
- Fix handling of links inside PRE elements on html to text conversion
- Fix indexing of links on html to text conversion
- Add mail attachments using drag & drop on HTML5 enabled browsers - Add mail attachments using drag & drop on HTML5 enabled browsers
- Add workaround for invalid BODYSTRUCTURE response - parse message with Mail_mimeDecode package (#1485585) - Add workaround for invalid BODYSTRUCTURE response - parse message with Mail_mimeDecode package (#1485585)
- Decode header value in rcube_mime::get() by default (#1488511) - Decode header value in rcube_mime::get() by default (#1488511)

@ -249,12 +249,11 @@ class html2text
* @access public * @access public
*/ */
var $callback_search = array( var $callback_search = array(
'/<(a) [^>]*href=("|\')([^"\']+)\2[^>]*>(.*?)<\/a>/i', '/<(a) [^>]*href=("|\')([^"\']+)\2[^>]*>(.*?)<\/a>/i', // <a href="">
// <a href=""> '/<(h)[123456]( [^>]*)?>(.*?)<\/h[123456]>/i', // h1 - h6
'/<(h)[123456][^>]*>(.*?)<\/h[123456]>/i', // H1 - H3 '/<(b)( [^>]*)?>(.*?)<\/b>/i', // <b>
'/<(b)[^>]*>(.*?)<\/b>/i', // <b> '/<(strong)( [^>]*)?>(.*?)<\/strong>/i', // <strong>
'/<(strong)[^>]*>(.*?)<\/strong>/i', // <strong> '/<(th)( [^>]*)?>(.*?)<\/th>/i', // <th> and </th>
'/<(th)[^>]*>(.*?)<\/th>/i', // <th> and </th>
); );
/** /**
@ -560,11 +559,11 @@ class html2text
} }
// Ignored link types // Ignored link types
if (preg_match('!^(javascript|mailto|#):!i', $link)) { if (preg_match('!^(javascript:|mailto:|#)!i', $link)) {
return $display; return $display;
} }
if (preg_match('!^(https?://)!i', $link)) { if (preg_match('!^([a-z][a-z0-9.+-]+:)!i', $link)) {
$url = $link; $url = $link;
} }
else { else {
@ -576,8 +575,8 @@ class html2text
} }
if (($index = array_search($url, $this->_link_list)) === false) { if (($index = array_search($url, $this->_link_list)) === false) {
$this->_link_list[] = $url;
$index = count($this->_link_list); $index = count($this->_link_list);
$this->_link_list[] = $url;
} }
return $display . ' [' . ($index+1) . ']'; return $display . ' [' . ($index+1) . ']';
@ -593,12 +592,20 @@ class html2text
{ {
// get the content of PRE element // get the content of PRE element
while (preg_match('/<pre[^>]*>(.*)<\/pre>/ismU', $text, $matches)) { while (preg_match('/<pre[^>]*>(.*)<\/pre>/ismU', $text, $matches)) {
$this->pre_content = $matches[1];
// Run our defined tags search-and-replace with callback
$this->pre_content = preg_replace_callback($this->callback_search,
array('html2text', '_preg_callback'), $this->pre_content);
// convert the content // convert the content
$this->pre_content = sprintf('<div><br>%s<br></div>', $this->pre_content = sprintf('<div><br>%s<br></div>',
preg_replace($this->pre_search, $this->pre_replace, $matches[1])); preg_replace($this->pre_search, $this->pre_replace, $this->pre_content));
// replace the content (use callback because content can contain $0 variable) // replace the content (use callback because content can contain $0 variable)
$text = preg_replace_callback('/<pre[^>]*>.*<\/pre>/ismU', $text = preg_replace_callback('/<pre[^>]*>.*<\/pre>/ismU',
array('html2text', '_preg_pre_callback'), $text, 1); array('html2text', '_preg_pre_callback'), $text, 1);
// free memory // free memory
$this->pre_content = ''; $this->pre_content = '';
} }
@ -671,11 +678,11 @@ class html2text
switch (strtolower($matches[1])) { switch (strtolower($matches[1])) {
case 'b': case 'b':
case 'strong': case 'strong':
return $this->_toupper($matches[2]); return $this->_toupper($matches[3]);
case 'th': case 'th':
return $this->_toupper("\t\t". $matches[2] ."\n"); return $this->_toupper("\t\t". $matches[3] ."\n");
case 'h': case 'h':
return $this->_toupper("\n\n". $matches[2] ."\n\n"); return $this->_toupper("\n\n". $matches[3] ."\n\n");
case 'a': case 'a':
// Remove spaces in URL (#1487805) // Remove spaces in URL (#1487805)
$url = str_replace(' ', '', $matches[3]); $url = str_replace(' ', '', $matches[3]);

@ -202,7 +202,7 @@ class washtml
$key = strtolower($key); $key = strtolower($key);
$value = $node->getAttribute($key); $value = $node->getAttribute($key);
if (isset($this->_html_attribs[$key]) || if (isset($this->_html_attribs[$key]) ||
($key == 'href' && preg_match('!^(http:|https:|ftp:|mailto:|//|#).+!i', $value))) ($key == 'href' && preg_match('!^([a-z][a-z0-9.+-]+:|//|#).+!i', $value)))
$t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"'; $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
else if ($key == 'style' && ($style = $this->wash_style($value))) { else if ($key == 'style' && ($style = $this->wash_style($value))) {
$quot = strpos($style, '"') !== false ? "'" : '"'; $quot = strpos($style, '"') !== false ? "'" : '"';

Loading…
Cancel
Save