several fixes to parse_url
This commit is contained in:
parent
eb604deff5
commit
2a65d01ce2
2 changed files with 23 additions and 13 deletions
2
boot.php
2
boot.php
|
@ -4,7 +4,7 @@ set_time_limit(0);
|
|||
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 ( 'DB_UPDATE_VERSION', 1075 );
|
||||
|
||||
|
|
|
@ -13,7 +13,7 @@ function parse_url_content(&$a) {
|
|||
|
||||
$text = null;
|
||||
|
||||
$template = "<a href=\"%s\" >%s</a>\n%s";
|
||||
$template = "<br /><a href=\"%s\" >%s</a>%s<br />";
|
||||
|
||||
|
||||
$arr = array('url' => $url, 'text' => '');
|
||||
|
@ -42,7 +42,7 @@ function parse_url_content(&$a) {
|
|||
if(strpos($s,'<title>')) {
|
||||
$title = substr($s,strpos($s,'<title>')+7,64);
|
||||
if(strpos($title,'<') !== false)
|
||||
$title = substr($title,0,strpos($title,'<'));
|
||||
$title = strip_tags(substr($title,0,strpos($title,'<')));
|
||||
}
|
||||
|
||||
$config = HTMLPurifier_Config::createDefault();
|
||||
|
@ -51,6 +51,8 @@ function parse_url_content(&$a) {
|
|||
$purifier = new HTMLPurifier($config);
|
||||
$s = $purifier->purify($s);
|
||||
|
||||
// logger('parse_url: purified: ' . $s, LOGGER_DATA);
|
||||
|
||||
$dom = @HTML5_Parser::parse($s);
|
||||
|
||||
if(! $dom) {
|
||||
|
@ -71,21 +73,27 @@ function parse_url_content(&$a) {
|
|||
if($divs) {
|
||||
foreach($divs as $div) {
|
||||
$class = $div->getAttribute('class');
|
||||
if($class && stristr($class,'article')) {
|
||||
if($class && (stristr($class,'article') || stristr($class,'content'))) {
|
||||
$items = $div->getElementsByTagName('p');
|
||||
if($items) {
|
||||
foreach($items as $item) {
|
||||
if($item->getElementsByTagName('script'))
|
||||
continue;
|
||||
$text = $item->textContent;
|
||||
$text = strip_tags($text);
|
||||
if(strlen($text) < 100)
|
||||
if(stristr($text,'<script')) {
|
||||
$text = '';
|
||||
continue;
|
||||
}
|
||||
$text = strip_tags($text);
|
||||
if(strlen($text) < 100) {
|
||||
$text = '';
|
||||
continue;
|
||||
}
|
||||
$text = substr($text,0,250) . '...' ;
|
||||
break;
|
||||
}
|
||||
}
|
||||
}
|
||||
if($text)
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
||||
|
@ -93,12 +101,14 @@ function parse_url_content(&$a) {
|
|||
$items = $dom->getElementsByTagName('p');
|
||||
if($items) {
|
||||
foreach($items as $item) {
|
||||
if($item->getElementsByTagName('script'))
|
||||
continue;
|
||||
$text = $item->textContent;
|
||||
$text = strip_tags($text);
|
||||
if(strlen($text) < 100)
|
||||
if(stristr($text,'<script'))
|
||||
continue;
|
||||
$text = strip_tags($text);
|
||||
if(strlen($text) < 100) {
|
||||
$text = '';
|
||||
continue;
|
||||
}
|
||||
$text = substr($text,0,250) . '...' ;
|
||||
break;
|
||||
}
|
||||
|
@ -106,7 +116,7 @@ function parse_url_content(&$a) {
|
|||
}
|
||||
|
||||
if(strlen($text)) {
|
||||
$text = '<br />' . $text;
|
||||
$text = '<br /><br />' . $text;
|
||||
}
|
||||
|
||||
echo sprintf($template,$url,($title) ? $title : $url,$text);
|
||||
|
|
Loading…
Reference in a new issue