diff --git a/mod/parse_url.php b/mod/parse_url.php
index a38f7e2702..09722341cb 100644
--- a/mod/parse_url.php
+++ b/mod/parse_url.php
@@ -1,7 +1,163 @@
query("//".$node);
+ foreach ($list as $child)
+ $child->parentNode->removeChild($child);
+ }
+}
-require_once('library/HTML5/Parser.php');
-require_once('library/HTMLPurifier.auto.php');
+function parseurl_getsiteinfo($url) {
+ $siteinfo = array();
+
+ $ch = curl_init();
+ curl_setopt($ch, CURLOPT_URL, $url);
+ curl_setopt($ch, CURLOPT_HEADER, 1);
+ curl_setopt($ch, CURLOPT_NOBODY, 0);
+ curl_setopt($ch, CURLOPT_TIMEOUT, 3);
+ curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+ curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
+
+ $header = curl_exec($ch);
+ curl_close($ch);
+
+ if (preg_match('/charset=(.*?)\n/', $header, $matches))
+ $charset = trim(array_pop($matches));
+ else
+ $charset = "utf-8";
+
+ $pos = strpos($header, "\r\n\r\n");
+
+ if ($pos)
+ $body = trim(substr($header, $pos));
+ else
+ $body = $header;
+
+ $body = mb_convert_encoding($body, "UTF-8", $charset);
+ $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
+
+ $doc = new DOMDocument();
+ @$doc->loadHTML($body);
+
+ deletenode($doc, 'style');
+ deletenode($doc, 'script');
+ deletenode($doc, 'option');
+ deletenode($doc, 'h1');
+ deletenode($doc, 'h2');
+ deletenode($doc, 'h3');
+ deletenode($doc, 'h4');
+ deletenode($doc, 'h5');
+ deletenode($doc, 'h6');
+ deletenode($doc, 'ol');
+ deletenode($doc, 'ul');
+
+ $xpath = new DomXPath($doc);
+
+ $list = $xpath->query("head/title");
+ foreach ($list as $node)
+ $siteinfo["title"] = html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
+
+ $list = $xpath->query("head/meta[@name]");
+ foreach ($list as $node) {
+ $attr = array();
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
+
+ switch (strtolower($attr["name"])) {
+ case "fulltitle":
+ $siteinfo["title"] = $attr["content"];
+ break;
+ case "description":
+ $siteinfo["text"] = $attr["content"];
+ break;
+ case "dc.title":
+ $siteinfo["title"] = $attr["content"];
+ break;
+ case "dc.description":
+ $siteinfo["text"] = $attr["content"];
+ break;
+ }
+ }
+
+ $list = $xpath->query("head/meta[@property]");
+ foreach ($list as $node) {
+ $attr = array();
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ $attr["content"] = html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8");
+
+ switch (strtolower($attr["property"])) {
+ case "og:image":
+ $siteinfo["image"] = $attr["content"];
+ break;
+ case "og:title":
+ $siteinfo["title"] = $attr["content"];
+ break;
+ case "og:description":
+ $siteinfo["text"] = $attr["content"];
+ break;
+ }
+ }
+
+ if ($siteinfo["image"] == "") {
+ require_once('include/Photo.php');
+ $list = $xpath->query("//img[@src]");
+ foreach ($list as $node) {
+ $attr = array();
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ // guess mimetype from headers or filename
+ $type = guess_image_type($attr["src"],true);
+
+ $i = fetch_url($attr["src"]);
+ $ph = new Photo($i, $type);
+
+ if(($ph->getWidth() > 200) and ($ph->getHeight() > 200))
+ $siteinfo["image"] = $attr["src"];
+ }
+ }
+
+ if ($siteinfo["text"] == "") {
+ $text = "";
+
+ $list = $xpath->query("//div[@class='article']");
+ foreach ($list as $node)
+ $text .= " ".trim($node->nodeValue);
+
+ if ($text == "") {
+ $list = $xpath->query("//div[@class='content']");
+ foreach ($list as $node)
+ $text .= " ".trim($node->nodeValue);
+ }
+
+ // If none text was found then take the paragraph content
+ if ($text == "") {
+ $list = $xpath->query("//p");
+ foreach ($list as $node)
+ $text .= " ".trim($node->nodeValue);
+ }
+
+ if ($text != "") {
+ $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
+
+ while (strpos($text, " "))
+ $text = trim(str_replace(" ", " ", $text));
+
+ $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
+ }
+ }
+
+ return($siteinfo);
+}
function arr_add_hashes(&$item,$k) {
$item = '#' . $item;
@@ -40,13 +196,11 @@ function parse_url_content(&$a) {
logger('parse_url: ' . $url);
-
if($textmode)
$template = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
else
$template = "
%s%s
";
-
$arr = array('url' => $url, 'text' => '');
call_hooks('parse_link', $arr);
@@ -60,187 +214,53 @@ function parse_url_content(&$a) {
if($url && $title && $text) {
if($textmode)
- $text = $br . $br . '[quote]' . $text . '[/quote]' . $br;
+ $text = $br . $br . '[quote]' . trim($text) . '[/quote]' . $br;
else
- $text = '
' . $text . '
' . trim($text) . '