several fixes to parse_url

This commit is contained in:
Friendika 2011-07-11 17:23:42 -07:00
parent eb604deff5
commit 2a65d01ce2
2 changed files with 23 additions and 13 deletions

View file

@ -4,7 +4,7 @@ set_time_limit(0);
ini_set('pcre.backtrack_limit', 250000); ini_set('pcre.backtrack_limit', 250000);
define ( 'FRIENDIKA_VERSION', '2.2.1037' ); define ( 'FRIENDIKA_VERSION', '2.2.1038' );
define ( 'DFRN_PROTOCOL_VERSION', '2.21' ); define ( 'DFRN_PROTOCOL_VERSION', '2.21' );
define ( 'DB_UPDATE_VERSION', 1075 ); define ( 'DB_UPDATE_VERSION', 1075 );

View file

@ -13,7 +13,7 @@ function parse_url_content(&$a) {
$text = null; $text = null;
$template = "<a href=\"%s\" >%s</a>\n%s"; $template = "<br /><a href=\"%s\" >%s</a>%s<br />";
$arr = array('url' => $url, 'text' => ''); $arr = array('url' => $url, 'text' => '');
@ -42,7 +42,7 @@ function parse_url_content(&$a) {
if(strpos($s,'<title>')) { if(strpos($s,'<title>')) {
$title = substr($s,strpos($s,'<title>')+7,64); $title = substr($s,strpos($s,'<title>')+7,64);
if(strpos($title,'<') !== false) if(strpos($title,'<') !== false)
$title = substr($title,0,strpos($title,'<')); $title = strip_tags(substr($title,0,strpos($title,'<')));
} }
$config = HTMLPurifier_Config::createDefault(); $config = HTMLPurifier_Config::createDefault();
@ -51,6 +51,8 @@ function parse_url_content(&$a) {
$purifier = new HTMLPurifier($config); $purifier = new HTMLPurifier($config);
$s = $purifier->purify($s); $s = $purifier->purify($s);
// logger('parse_url: purified: ' . $s, LOGGER_DATA);
$dom = @HTML5_Parser::parse($s); $dom = @HTML5_Parser::parse($s);
if(! $dom) { if(! $dom) {
@ -71,21 +73,27 @@ function parse_url_content(&$a) {
if($divs) { if($divs) {
foreach($divs as $div) { foreach($divs as $div) {
$class = $div->getAttribute('class'); $class = $div->getAttribute('class');
if($class && stristr($class,'article')) { if($class && (stristr($class,'article') || stristr($class,'content'))) {
$items = $div->getElementsByTagName('p'); $items = $div->getElementsByTagName('p');
if($items) { if($items) {
foreach($items as $item) { foreach($items as $item) {
if($item->getElementsByTagName('script'))
continue;
$text = $item->textContent; $text = $item->textContent;
$text = strip_tags($text); if(stristr($text,'<script')) {
if(strlen($text) < 100) $text = '';
continue; continue;
}
$text = strip_tags($text);
if(strlen($text) < 100) {
$text = '';
continue;
}
$text = substr($text,0,250) . '...' ; $text = substr($text,0,250) . '...' ;
break; break;
} }
} }
} }
if($text)
break;
} }
} }
@ -93,12 +101,14 @@ function parse_url_content(&$a) {
$items = $dom->getElementsByTagName('p'); $items = $dom->getElementsByTagName('p');
if($items) { if($items) {
foreach($items as $item) { foreach($items as $item) {
if($item->getElementsByTagName('script'))
continue;
$text = $item->textContent; $text = $item->textContent;
$text = strip_tags($text); if(stristr($text,'<script'))
if(strlen($text) < 100)
continue; continue;
$text = strip_tags($text);
if(strlen($text) < 100) {
$text = '';
continue;
}
$text = substr($text,0,250) . '...' ; $text = substr($text,0,250) . '...' ;
break; break;
} }
@ -106,7 +116,7 @@ function parse_url_content(&$a) {
} }
if(strlen($text)) { if(strlen($text)) {
$text = '<br />' . $text; $text = '<br /><br />' . $text;
} }
echo sprintf($template,$url,($title) ? $title : $url,$text); echo sprintf($template,$url,($title) ? $title : $url,$text);