]> git.mxchange.org Git - friendica.git/blobdiff - mod/parse_url.php
Merge pull request #2118 from annando/1511-supported-networks
[friendica.git] / mod / parse_url.php
index ea05055f0cefc69b461c29b5e009786320bcc2b8..ef051d9f6b433c0e8152c39168c0debd8cfbee0a 100644 (file)
@@ -24,47 +24,144 @@ if(!function_exists('deletenode')) {
 }
 
 function completeurl($url, $scheme) {
-        $urlarr = parse_url($url);
+       $urlarr = parse_url($url);
 
-        if (isset($urlarr["scheme"]))
-                return($url);
+       if (isset($urlarr["scheme"]))
+               return($url);
 
-        $schemearr = parse_url($scheme);
+       $schemearr = parse_url($scheme);
 
-        $complete = $schemearr["scheme"]."://".$schemearr["host"];
+       $complete = $schemearr["scheme"]."://".$schemearr["host"];
 
-        if ($schemearr["port"] != "")
-                $complete .= ":".$schemearr["port"];
+       if (@$schemearr["port"] != "")
+               $complete .= ":".$schemearr["port"];
 
-        $complete .= $urlarr["path"];
+               if(strpos($urlarr['path'],'/') !== 0)
+                       $complete .= '/';
 
-        if ($urlarr["query"] != "")
-                $complete .= "?".$urlarr["query"];
+       $complete .= $urlarr["path"];
 
-        if ($urlarr["fragment"] != "")
-                $complete .= "#".$urlarr["fragment"];
+       if (@$urlarr["query"] != "")
+               $complete .= "?".$urlarr["query"];
 
-        return($complete);
+       if (@$urlarr["fragment"] != "")
+               $complete .= "#".$urlarr["fragment"];
+
+       return($complete);
 }
 
-function parseurl_getsiteinfo($url) {
+function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
+
+       $data = Cache::get("parse_url:".$no_guessing.":".$do_oembed.":".$url);
+       if (!is_null($data)) {
+               $data = unserialize($data);
+               return $data;
+       }
+
+       $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed);
+
+       Cache::set("parse_url:".$no_guessing.":".$do_oembed.":".$url,serialize($data), CACHE_DAY);
+
+       return $data;
+}
+
+function parseurl_getsiteinfo($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);
+       }
+
+       $url = trim($url, "'");
+       $url = trim($url, '"');
+
+       $url = original_url($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($ch, CURLOPT_NOBODY, 0);
+       curl_setopt($ch, CURLOPT_NOBODY, 1);
        curl_setopt($ch, CURLOPT_TIMEOUT, 3);
        curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
-       curl_setopt($ch,CURLOPT_USERAGENT,'Opera/9.64(Windows NT 5.1; U; de) Presto/2.1.1');
+       //curl_setopt($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($siteinfo);
+       }
+
+       // if the file is too large then exit
+       if ($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;
+               }
+       }
+
+       $stamp1 = microtime(true);
+
+       // 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());
 
        $header = curl_exec($ch);
+       $curl_info = @curl_getinfo($ch);
+       $http_code = $curl_info['http_code'];
        curl_close($ch);
 
+       $a->save_timestamp($stamp1, "network");
+
        // Fetch the first mentioned charset. Can be in body or header
+       $charset = "";
        if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
-               $charset = trim(array_pop($matches));
-       else
+               $charset = trim(trim(trim(array_pop($matches)), ';,'));
+
+       if ($charset == "")
                $charset = "utf-8";
 
        $pos = strpos($header, "\r\n\r\n");
@@ -74,7 +171,12 @@ function parseurl_getsiteinfo($url) {
        else
                $body = $header;
 
-       $body = mb_convert_encoding($body, "UTF-8", $charset);
+       if (($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);
+       }
+
        $body = mb_convert_encoding($body, 'HTML-ENTITIES', "UTF-8");
 
        $doc = new DOMDocument();
@@ -94,6 +196,28 @@ function parseurl_getsiteinfo($url) {
 
        $xpath = new DomXPath($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($siteinfo);
+                       }
+               }
+       }
+
        //$list = $xpath->query("head/title");
        $list = $xpath->query("//title");
        foreach ($list as $node)
@@ -104,25 +228,60 @@ function parseurl_getsiteinfo($url) {
        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");
+                       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";
+       }
 
-               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"] = $attr["content"];
-                               break;
-               }
+       if (isset($keywords)) {
+               $siteinfo["keywords"] = array();
+               foreach ($keywords as $keyword)
+                       if (!in_array(trim($keyword), $siteinfo["keywords"]))
+                               $siteinfo["keywords"][] = trim($keyword);
        }
 
        //$list = $xpath->query("head/meta[@property]");
@@ -130,36 +289,37 @@ function parseurl_getsiteinfo($url) {
        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"] = $attr["content"];
-                               break;
-               }
+                       foreach ($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"] == "") {
-                $list = $xpath->query("//img[@src]");
-                foreach ($list as $node) {
-                        $attr = array();
-                        if ($node->attributes->length)
-                                foreach ($node->attributes as $attribute)
-                                        $attr[$attribute->name] = $attribute->value;
+       if ((@$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);
+                       $photodata = @getimagesize($src);
 
-                       if (($photodata[0] > 150) and ($photodata[1] > 150)) {
+                       if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) {
                                if ($photodata[0] > 300) {
                                        $photodata[1] = round($photodata[1] * (300 / $photodata[0]));
                                        $photodata[0] = 300;
@@ -173,21 +333,21 @@ function parseurl_getsiteinfo($url) {
                                                                "height"=>$photodata[1]);
                        }
 
-                }
-        } else {
+               }
+    } else {
                $src = completeurl($siteinfo["image"], $url);
 
                unset($siteinfo["image"]);
 
-               $photodata = getimagesize($src);
+               $photodata = @getimagesize($src);
 
-               if (($photodata[0] > 10) and ($photodata[1] > 10))
+               if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
                        $siteinfo["images"][] = array("src"=>$src,
                                                        "width"=>$photodata[0],
                                                        "height"=>$photodata[1]);
        }
 
-       if ($siteinfo["text"] == "") {
+       if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) {
                $text = "";
 
                $list = $xpath->query("//div[@class='article']");
@@ -216,10 +376,14 @@ function parseurl_getsiteinfo($url) {
                        while (strpos($text, "  "))
                                $text = trim(str_replace("  ", " ", $text));
 
-                       $siteinfo["text"] = html_entity_decode(substr($text,0,350), ENT_QUOTES, "UTF-8").'...';
+                       $siteinfo["text"] = trim(html_entity_decode(substr($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);
 }
 
@@ -233,7 +397,8 @@ function parse_url_content(&$a) {
        $str_tags = '';
 
        $textmode = false;
-       if(local_user() && intval(get_pconfig(local_user(),'system','plaintext')))
+
+       if(local_user() && (! feature_enabled(local_user(),'richtext')))
                $textmode = true;
 
        //if($textmode)
@@ -258,12 +423,21 @@ function parse_url_content(&$a) {
                }
        }
 
+       // 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 = $br . '[bookmark=%s]%s[/bookmark]%s' . $br;
+               $template = '[bookmark=%s]%s[/bookmark]%s';
        else
-               $template = "<br /><a class=\"bookmark\" href=\"%s\" >%s</a>%s<br />";
+               $template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
 
        $arr = array('url' => $url, 'text' => '');
 
@@ -277,13 +451,15 @@ function parse_url_content(&$a) {
 
        if($url && $title && $text) {
 
-               if($textmode)
-                       $text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
-               else
-                       $text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
-
                $title = str_replace(array("\r","\n"),array('',''),$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: ' . $result);
@@ -294,43 +470,78 @@ function parse_url_content(&$a) {
 
        $siteinfo = parseurl_getsiteinfo($url);
 
-       if($siteinfo["title"] == "") {
-               echo sprintf($template,$url,$url,'') . $str_tags;
-               killme();
-       } else {
+//     if ($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(sizeof($siteinfo["images"]) > 0){
-            /*
-              Execute below code only if image is present in siteinfo
-             */
-            foreach ($siteinfo["images"] as $imagedata)
-               if($textmode)
-                    $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]';
+       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
-                    $image .= '<img height="'.$imagedata["height"].'" width="'.$imagedata["width"].'" src="'.$imagedata["src"].'" alt="photo" />';
-        }
+                       $max_images = intval($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($text)) {
                if($textmode)
-                       $text = $br.'[quote]'.trim($text).'[/quote]'.$br ;
+                       $text = '[quote]'.trim($text).'[/quote]';
                else
-                       $text = '<br /><blockquote>'.trim($text).'</blockquote><br />';
+                       $text = '<blockquote>'.htmlspecialchars(trim($text)).'</blockquote>';
        }
 
-       if($image) {
+       if($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);
 
-       echo trim($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();
 }
+?>