X-Git-Url: https://git.mxchange.org/?a=blobdiff_plain;f=mod%2Fparse_url.php;h=71d767675f8c51a1b3cba44b3edae4e2bd0c7bf8;hb=5f89cb09234bc79aab2be93778609b344239c3da;hp=bf6639b300a72cd6a24cd2c6a14aa186fe6a9862;hpb=35edf79abe08bcc1833dc695cb9aebfa5d4a78c0;p=friendica.git
diff --git a/mod/parse_url.php b/mod/parse_url.php
index bf6639b300..71d767675f 100644
--- a/mod/parse_url.php
+++ b/mod/parse_url.php
@@ -1,114 +1,531 @@
+
+
-function parse_url_content(&$a) {
+
+ Shiny Trinket
+ 
+ Shiny trinkets are shiny.
+
+*/
- logger('parse_url: ' . $_GET['url']);
+if(!function_exists('deletenode')) {
+ function deletenode(&$doc, $node)
+ {
+ $xpath = new DomXPath($doc);
+ $list = $xpath->query("//".$node);
+ foreach ($list as $child)
+ $child->parentNode->removeChild($child);
+ }
+}
- $url = trim(hex2bin($_GET['url']));
+function completeurl($url, $scheme) {
+ $urlarr = parse_url($url);
- logger('parse_url: ' . $url);
+ if (isset($urlarr["scheme"]))
+ return($url);
- $text = null;
+ $schemearr = parse_url($scheme);
- $template = "%s\n%s";
+ $complete = $schemearr["scheme"]."://".$schemearr["host"];
+ if (@$schemearr["port"] != "")
+ $complete .= ":".$schemearr["port"];
- $arr = array('url' => $url, 'text' => '');
+ if(strpos($urlarr['path'],'/') !== 0)
+ $complete .= '/';
- call_hooks('parse_link', $arr);
+ $complete .= $urlarr["path"];
- if(strlen($arr['text'])) {
- echo $arr['text'];
- killme();
- }
+ if (@$urlarr["query"] != "")
+ $complete .= "?".$urlarr["query"];
- if($url) {
- $s = fetch_url($url);
- } else {
- echo '';
- killme();
+ if (@$urlarr["fragment"] != "")
+ $complete .= "#".$urlarr["fragment"];
+
+ return($complete);
+}
+
+function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) {
+ require_once("include/network.php");
+
+ $a = get_app();
+
+ $siteinfo = array();
+
+ if ($count > 10) {
+ logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
+ return($siteinfo);
}
- logger('parse_url: data: ' . $s, LOGGER_DATA);
+ $url = trim($url, "'");
+ $url = trim($url, '"');
- if(! $s) {
- echo sprintf($template,$url,$url,'');
- killme();
+ $url = original_url($url);
+
+ $siteinfo["url"] = $url;
+ $siteinfo["type"] = "link";
+
+ $stamp1 = microtime(true);
+
+ $ch = curl_init();
+ curl_setopt($ch, CURLOPT_URL, $url);
+ curl_setopt($ch, CURLOPT_HEADER, 1);
+ curl_setopt($ch, CURLOPT_NOBODY, 1);
+ curl_setopt($ch, CURLOPT_TIMEOUT, 3);
+ curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+ //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
+ curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
+
+ $header = curl_exec($ch);
+ $curl_info = @curl_getinfo($ch);
+ $http_code = $curl_info['http_code'];
+ curl_close($ch);
+
+ $a->save_timestamp($stamp1, "network");
+
+ if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302") OR ($curl_info['http_code'] == "303") OR ($curl_info['http_code'] == "307"))
+ AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
+ if ($curl_info['redirect_url'] != "")
+ $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url'], $no_guessing, $do_oembed, ++$count);
+ else
+ $siteinfo = parseurl_getsiteinfo($curl_info['location'], $no_guessing, $do_oembed, ++$count);
+ return($siteinfo);
}
- if(strpos($s,'')) {
- $title = substr($s,strpos($s,'<title>')+7,64);
- if(strpos($title,'<') !== false)
- $title = substr($title,0,strpos($title,'<'));
+ // if the file is too large then exit
+ if ($curl_info["download_content_length"] > 1000000)
+ return($siteinfo);
+
+ // if it isn't a HTML file then exit
+ if (($curl_info["content_type"] != "") AND !strstr(strtolower($curl_info["content_type"]),"html"))
+ return($siteinfo);
+
+ if ($do_oembed) {
+ require_once("include/oembed.php");
+
+ $oembed_data = oembed_fetch_url($url);
+
+ if ($oembed_data->type != "error")
+ $siteinfo["type"] = $oembed_data->type;
+
+ if (($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) {
+ if (isset($oembed_data->title))
+ $siteinfo["title"] = $oembed_data->title;
+ if (isset($oembed_data->description))
+ $siteinfo["text"] = trim($oembed_data->description);
+ if (isset($oembed_data->thumbnail_url))
+ $siteinfo["image"] = $oembed_data->thumbnail_url;
+ }
}
- $config = HTMLPurifier_Config::createDefault();
- $config->set('Cache.DefinitionImpl', null);
+ $stamp1 = microtime(true);
- $purifier = new HTMLPurifier($config);
- $s = $purifier->purify($s);
+ // Now fetch the body as well
+ $ch = curl_init();
+ curl_setopt($ch, CURLOPT_URL, $url);
+ curl_setopt($ch, CURLOPT_HEADER, 1);
+ curl_setopt($ch, CURLOPT_NOBODY, 0);
+ curl_setopt($ch, CURLOPT_TIMEOUT, 10);
+ curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
+ curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
- $dom = @HTML5_Parser::parse($s);
+ $header = curl_exec($ch);
+ $curl_info = @curl_getinfo($ch);
+ $http_code = $curl_info['http_code'];
+ curl_close($ch);
- if(! $dom) {
- echo sprintf($template,$url,$url,'');
- killme();
+ $a->save_timestamp($stamp1, "network");
+
+ // Fetch the first mentioned charset. Can be in body or header
+ $charset = "";
+ if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
+ $charset = trim(trim(trim(array_pop($matches)), ';,'));
+
+ if ($charset == "")
+ $charset = "utf-8";
+
+ $pos = strpos($header, "\r\n\r\n");
+
+ if ($pos)
+ $body = trim(substr($header, $pos));
+ else
+ $body = $header;
+
+ if (($charset != '') AND (strtoupper($charset) != "UTF-8")) {
+ logger("parseurl_getsiteinfo: detected charset ".$charset, LOGGER_DEBUG);
+ //$body = mb_convert_encoding($body, "UTF-8", $charset);
+ $body = iconv($charset, "UTF-8//TRANSLIT", $body);
}
- $items = $dom->getElementsByTagName('title');
+ $body = mb_convert_encoding($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');
- if($items) {
- foreach($items as $item) {
- $title = trim($item->textContent);
- break;
+ $xpath = new DomXPath($doc);
+
+ $list = $xpath->query("//meta[@content]");
+ foreach ($list as $node) {
+ $attr = array();
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ if (@$attr["http-equiv"] == 'refresh') {
+ $path = $attr["content"];
+ $pathinfo = explode(";", $path);
+ $content = "";
+ foreach ($pathinfo AS $value) {
+ if (substr(strtolower($value), 0, 4) == "url=")
+ $content = substr($value, 4);
+ }
+ if ($content != "") {
+ $siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
+ return($siteinfo);
+ }
}
}
- $divs = $dom->getElementsByTagName('div');
- if($divs) {
- foreach($divs as $div) {
- $class = $div->getAttribute('class');
- if($class && stristr($class,'article')) {
- $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)
- continue;
- $text = substr($text,0,250) . '...' ;
- break;
- }
+ //$list = $xpath->query("head/title");
+ $list = $xpath->query("//title");
+ foreach ($list as $node)
+ $siteinfo["title"] = html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8");
+
+ //$list = $xpath->query("head/meta[@name]");
+ $list = $xpath->query("//meta[@name]");
+ foreach ($list as $node) {
+ $attr = array();
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
+
+ if ($attr["content"] != "")
+ switch (strtolower($attr["name"])) {
+ case "fulltitle":
+ $siteinfo["title"] = $attr["content"];
+ break;
+ case "description":
+ $siteinfo["text"] = $attr["content"];
+ break;
+ case "thumbnail":
+ $siteinfo["image"] = $attr["content"];
+ break;
+ case "twitter:image":
+ $siteinfo["image"] = $attr["content"];
+ break;
+ case "twitter:image:src":
+ $siteinfo["image"] = $attr["content"];
+ break;
+ case "twitter:card":
+ if (($siteinfo["type"] == "") OR ($attr["content"] == "photo"))
+ $siteinfo["type"] = $attr["content"];
+ break;
+ case "twitter:description":
+ $siteinfo["text"] = $attr["content"];
+ break;
+ case "twitter:title":
+ $siteinfo["title"] = $attr["content"];
+ break;
+ case "dc.title":
+ $siteinfo["title"] = $attr["content"];
+ break;
+ case "dc.description":
+ $siteinfo["text"] = $attr["content"];
+ break;
+ case "keywords":
+ $keywords = explode(",", $attr["content"]);
+ break;
+ case "news_keywords":
+ $keywords = explode(",", $attr["content"]);
+ break;
+ }
+ if ($siteinfo["type"] == "summary")
+ $siteinfo["type"] = "link";
+ }
+
+ if (isset($keywords)) {
+ $siteinfo["keywords"] = array();
+ foreach ($keywords as $keyword)
+ $siteinfo["keywords"][] = trim($keyword);
+ }
+
+ //$list = $xpath->query("head/meta[@property]");
+ $list = $xpath->query("//meta[@property]");
+ foreach ($list as $node) {
+ $attr = array();
+ if ($node->attributes->length)
+ foreach ($node->attributes as $attribute)
+ $attr[$attribute->name] = $attribute->value;
+
+ $attr["content"] = trim(html_entity_decode($attr["content"], ENT_QUOTES, "UTF-8"));
+
+ if ($attr["content"] != "")
+ switch (strtolower($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"] == "") AND !$no_guessing) {
+ $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;
+
+ $src = completeurl($attr["src"], $url);
+ $photodata = @getimagesize($src);
+
+ if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
+ if ($photodata[0] > 300) {
+ $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
+ $photodata[0] = 300;
}
+ if ($photodata[1] > 300) {
+ $photodata[0] = round($photodata[0] * (300 / $photodata[1]));
+ $photodata[1] = 300;
+ }
+ $siteinfo["images"][] = array("src"=>$src,
+ "width"=>$photodata[0],
+ "height"=>$photodata[1]);
}
+
}
+ } else {
+ $src = completeurl($siteinfo["image"], $url);
+
+ unset($siteinfo["image"]);
+
+ $photodata = @getimagesize($src);
+
+ if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
+ $siteinfo["images"][] = array("src"=>$src,
+ "width"=>$photodata[0],
+ "height"=>$photodata[1]);
}
- if(! $text) {
- $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)
- continue;
- $text = substr($text,0,250) . '...' ;
+ if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
+ $text = "";
+
+ $list = $xpath->query("//div[@class='article']");
+ foreach ($list as $node)
+ if (strlen($node->nodeValue) > 40)
+ $text .= " ".trim($node->nodeValue);
+
+ if ($text == "") {
+ $list = $xpath->query("//div[@class='content']");
+ foreach ($list as $node)
+ if (strlen($node->nodeValue) > 40)
+ $text .= " ".trim($node->nodeValue);
+ }
+
+ // If none text was found then take the paragraph content
+ if ($text == "") {
+ $list = $xpath->query("//p");
+ foreach ($list as $node)
+ if (strlen($node->nodeValue) > 40)
+ $text .= " ".trim($node->nodeValue);
+ }
+
+ if ($text != "") {
+ $text = trim(str_replace(array("\n", "\r"), array(" ", " "), $text));
+
+ while (strpos($text, " "))
+ $text = trim(str_replace(" ", " ", $text));
+
+ $siteinfo["text"] = trim(html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...');
+ }
+ }
+
+ logger("parseurl_getsiteinfo: Siteinfo for ".$url." ".print_r($siteinfo, true), LOGGER_DEBUG);
+
+ call_hooks('getsiteinfo', $siteinfo);
+
+ return($siteinfo);
+}
+
+function arr_add_hashes(&$item,$k) {
+ $item = '#' . $item;
+}
+
+function parse_url_content(&$a) {
+
+ $text = null;
+ $str_tags = '';
+
+ $textmode = false;
+
+ if(local_user() && (! feature_enabled(local_user(),'richtext')))
+ $textmode = true;
+
+ //if($textmode)
+ $br = (($textmode) ? "\n" : '<br />');
+
+ if(x($_GET,'binurl'))
+ $url = trim(hex2bin($_GET['binurl']));
+ else
+ $url = trim($_GET['url']);
+
+ if($_GET['title'])
+ $title = strip_tags(trim($_GET['title']));
+
+ if($_GET['description'])
+ $text = strip_tags(trim($_GET['description']));
+
+ if($_GET['tags']) {
+ $arr_tags = str_getcsv($_GET['tags']);
+ if(count($arr_tags)) {
+ array_walk($arr_tags,'arr_add_hashes');
+ $str_tags = $br . implode(' ',$arr_tags) . $br;
+ }
+ }
+
+ // add url scheme if missing
+ $arrurl = parse_url($url);
+ if (!x($arrurl, 'scheme')) {
+ if (x($arrurl, 'host'))
+ $url = "http:".$url;
+ else
+ $url = "http://".$url;
+ }
+
+ logger('parse_url: ' . $url);
+
+ if($textmode)
+ $template = '[bookmark=%s]%s[/bookmark]%s';
+ else
+ $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
+
+ $arr = array('url' => $url, 'text' => '');
+
+ call_hooks('parse_link', $arr);
+
+ if(strlen($arr['text'])) {
+ echo $arr['text'];
+ killme();
+ }
+
+
+ if($url && $title && $text) {
+
+ $title = str_replace(array("\r","\n"),array('',''),$title);
+
+ if($textmode)
+ $text = '[quote]' . trim($text) . '[/quote]' . $br;
+ else {
+ $text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
+ $title = htmlspecialchars($title);
+ }
+
+ $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
+
+ logger('parse_url (unparsed): returns: ' . $result);
+
+ echo $result;
+ killme();
+ }
+
+ $siteinfo = parseurl_getsiteinfo($url);
+
+// if ($textmode) {
+// require_once("include/items.php");
+//
+// echo add_page_info_data($siteinfo);
+// killme();
+// }
+
+ $url= $siteinfo["url"];
+
+ // If the link contains BBCode stuff, make a short link out of this to avoid parsing problems
+ if (strpos($url, '[') OR strpos($url, ']')) {
+ require_once("include/network.php");
+ $url = short_link($url);
+ }
+
+ $sitedata = "";
+
+ if($siteinfo["title"] != "") {
+ $text = $siteinfo["text"];
+ $title = $siteinfo["title"];
+ }
+
+ $image = "";
+
+ if (($siteinfo["type"] != "video") AND (sizeof($siteinfo["images"]) > 0)){
+ /* Execute below code only if image is present in siteinfo */
+
+ $total_images = 0;
+ $max_images = get_config('system','max_bookmark_images');
+ if($max_images === false)
+ $max_images = 2;
+ else
+ $max_images = intval($max_images);
+
+ foreach ($siteinfo["images"] as $imagedata) {
+ if($textmode)
+ $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]' . "\n";
+ else
+ $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" /><br />';
+ $total_images ++;
+ if($max_images && $max_images >= $total_images)
break;
- }
}
}
if(strlen($text)) {
- $text = '<br />' . $text;
+ if($textmode)
+ $text = '[quote]'.trim($text).'[/quote]';
+ else
+ $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
}
- echo sprintf($template,$url,($title) ? $title : $url,$text);
+ if($image)
+ $text = $br.$br.$image.$text;
+ else
+ $text = $br.$text;
+
+ $title = str_replace(array("\r","\n"),array('',''),$title);
+
+ $result = sprintf($template,$url,($title) ? $title : $url,$text) . $str_tags;
+
+ logger('parse_url: returns: ' . $result);
+
+ $sitedata .= trim($result);
+
+ if (($siteinfo["type"] == "video") AND ($url != ""))
+ echo "[class=type-video]".$sitedata."[/class]";
+ elseif (($siteinfo["type"] != "photo"))
+ echo "[class=type-link]".$sitedata."[/class]";
+ else
+ echo "[class=type-photo]".$title.$br.$image."[/class]";
+
killme();
}
+?>