]> git.mxchange.org Git - friendica.git/blobdiff - mod/parse_url.php
New style for shared links. Preparation for changes in some addons
[friendica.git] / mod / parse_url.php
index 97e1658c8985fd47b6360f9685a91d0549c3ca20..cb7268c1df83acd0e14bb9db4e0b1617346e2bc4 100644 (file)
@@ -1,5 +1,17 @@
 <?php
-require_once('include/Photo.php');
+/* To-Do
+https://developers.google.com/+/plugins/snippet/
+
+<meta itemprop="name" content="Toller Titel">
+<meta itemprop="description" content="Eine tolle Beschreibung">
+<meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
+
+<body itemscope itemtype="http://schema.org/Product">
+  <h1 itemprop="name">Shiny Trinket</h1>
+  <img itemprop="image" src="{image-url}" />
+  <p itemprop="description">Shiny trinkets are shiny.</p>
+</body>
+*/
 
 if(!function_exists('deletenode')) {
        function deletenode(&$doc, $node)
@@ -11,21 +23,73 @@ if(!function_exists('deletenode')) {
        }
 }
 
-function parseurl_getsiteinfo($url) {
-       $siteinfo = array();
+function completeurl($url, $scheme) {
+        $urlarr = parse_url($url);
+
+        if (isset($urlarr["scheme"]))
+                return($url);
+
+        $schemearr = parse_url($scheme);
+
+        $complete = $schemearr["scheme"]."://".$schemearr["host"];
+
+        if (@$schemearr["port"] != "")
+                $complete .= ":".$schemearr["port"];
+
+               if(strpos($urlarr['path'],'/') !== 0)
+                       $complete .= '/';
+
+        $complete .= $urlarr["path"];
+
+        if (@$urlarr["query"] != "")
+                $complete .= "?".$urlarr["query"];
+
+        if (@$urlarr["fragment"] != "")
+                $complete .= "#".$urlarr["fragment"];
 
+        return($complete);
+}
+
+function parseurl_getsiteinfo($url, $no_guessing = false) {
+       $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');
+       //curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
+       curl_setopt($ch,CURLOPT_USERAGENT,' Mozilla/5.0 (Macintosh; Intel Mac OS X 10.8; rv:24.0) Gecko/20100101 Firefox/24.0');
 
        $header = curl_exec($ch);
+       $curl_info = @curl_getinfo($ch);
+        $http_code = $curl_info['http_code'];
        curl_close($ch);
 
-       if (preg_match('/charset=(.*?)\n/', $header, $matches))
+       if ((($curl_info['http_code'] == "301") OR ($curl_info['http_code'] == "302"))
+               AND (($curl_info['redirect_url'] != "") OR ($curl_info['location'] != ""))) {
+               if ($curl_info['redirect_url'] != "")
+                       $siteinfo = parseurl_getsiteinfo($curl_info['redirect_url']);
+               else
+                       $siteinfo = parseurl_getsiteinfo($curl_info['location']);
+               return($siteinfo);
+       }
+
+       require_once("include/oembed.php");
+
+       $oembed_data = oembed_fetch_url($url);
+
+       if ($oembed_data->type == "link") {
+               if (isset($oembed_data->title))
+                       $siteinfo["title"] = $oembed_data->title;
+               if (isset($oembed_data->description))
+                       $siteinfo["text"] = $oembed_data->description;
+               if (isset($oembed_data->thumbnail_url))
+                       $siteinfo["image"] = $oembed_data->thumbnail_url;
+       }
+
+       // Fetch the first mentioned charset. Can be in body or header
+       if (preg_match('/charset=(.*?)['."'".'"\s\n]/', $header, $matches))
                $charset = trim(array_pop($matches));
        else
                $charset = "utf-8";
@@ -57,11 +121,35 @@ function parseurl_getsiteinfo($url) {
 
        $xpath = new DomXPath($doc);
 
-       $list = $xpath->query("head/title");
+       $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);
+                                return($siteinfo);
+                        }
+                }
+       }
+
+       //$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("head/meta[@name]");
+       $list = $xpath->query("//meta[@name]");
        foreach ($list as $node) {
                $attr = array();
                if ($node->attributes->length)
@@ -86,7 +174,8 @@ function parseurl_getsiteinfo($url) {
                }
        }
 
-       $list = $xpath->query("head/meta[@property]");
+       //$list = $xpath->query("head/meta[@property]");
+       $list = $xpath->query("//meta[@property]");
        foreach ($list as $node) {
                $attr = array();
                if ($node->attributes->length)
@@ -108,66 +197,66 @@ function parseurl_getsiteinfo($url) {
                }
        }
 
-       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;
-
-                        // 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->is_valid() and ($ph->getWidth() > 200) and ($ph->getHeight() > 200)) {
-                               if ($siteinfo["image"] == "")
-                                       $siteinfo["image"] = $attr["src"];
-
-                               if($ph->getWidth() > 300 || $ph->getHeight() > 300) {
-                                       $ph->scaleImage(300);
-                                       $siteinfo["images"][] = array("src"=>$attr["src"],
-                                                                       "width"=>$ph->getWidth(),
-                                                                       "height"=>$ph->getHeight());
-                               } else
-                                       $siteinfo["images"][] = array("src"=>$attr["src"],
-                                                                       "width"=>$ph->getWidth(),
-                                                                       "height"=>$ph->getHeight());
+       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 {
-               // guess mimetype from headers or filename
-                $type = guess_image_type($siteinfo["image"],true);
 
-                $i = fetch_url($siteinfo["image"]);
-                $ph = new Photo($i, $type);
+               }
+    } else {
+               $src = completeurl($siteinfo["image"], $url);
+
+               unset($siteinfo["image"]);
 
-               if($ph->is_valid())
-                       $siteinfo["images"][] = array("src"=>$siteinfo["image"],
-                                                       "width"=>$ph->getWidth(),
-                                                       "height"=>$ph->getHeight());
+               $photodata = @getimagesize($src);
+
+               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 = $xpath->query("//div[@class='article']");
                foreach ($list as $node)
-                       $text .= " ".trim($node->nodeValue);
+                       if (strlen($node->nodeValue) > 40)
+                               $text .= " ".trim($node->nodeValue);
 
                if ($text == "") {
                        $list = $xpath->query("//div[@class='content']");
                        foreach ($list as $node)
-                               $text .= " ".trim($node->nodeValue);
+                               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)
-                               $text .= " ".trim($node->nodeValue);
+                               if (strlen($node->nodeValue) > 40)
+                                       $text .= " ".trim($node->nodeValue);
                }
 
                if ($text != "") {
@@ -193,7 +282,8 @@ function parse_url_content(&$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($textmode)
@@ -238,9 +328,9 @@ function parse_url_content(&$a) {
        if($url && $title && $text) {
 
                if($textmode)
-                       $text = $br . $br . '[quote]' . trim($text) . '[/quote]' . $br;
+                       $text = $br . '[quote]' . trim($text) . '[/quote]' . $br;
                else
-                       $text = '<br /><br /><blockquote>' . trim($text) . '</blockquote><br />';
+                       $text = '<br /><blockquote>' . trim($text) . '</blockquote><br />';
 
                $title = str_replace(array("\r","\n"),array('',''),$title);
 
@@ -264,38 +354,26 @@ function parse_url_content(&$a) {
 
        $image = "";
 
-       foreach ($siteinfo["images"] as $imagedata)
-               if($textmode)
-                       $image .= '[img='.$imagedata["width"].'x'.$imagedata["height"].']'.$imagedata["src"].'[/img]';
+       if(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" />';
-
-/*     if ($image != "") {
-               $i = fetch_url($image);
-               if($i) {
-                       require_once('include/Photo.php');
-                       // guess mimetype from headers or filename
-                       $type = guess_image_type($image,true);
-
-                       $ph = new Photo($i, $type);
-                       if($ph->is_valid()) {
-                               if($ph->getWidth() > 300 || $ph->getHeight() > 300) {
-                                       $ph->scaleImage(300);
-                                       $new_width = $ph->getWidth();
-                                       $new_height = $ph->getHeight();
-                                       if($textmode)
-                                               $image = $br . $br . '[img=' . $new_width . 'x' . $new_height . ']' . $image . '[/img]';
-                                       else
-                                               $image = '<br /><br /><img height="' . $new_height . '" width="' . $new_width . '" src="' .$image . '" alt="photo" />';
-                               } else {
-                                       if($textmode)
-                                               $image = $br.$br.'[img]'.$image.'[/img]';
-                                       else
-                                               $image = '<br /><br /><img src="'.$image.'" 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)
@@ -305,7 +383,7 @@ function parse_url_content(&$a) {
        }
 
        if($image) {
-               $text = $br.$br.$image.$br.$text;
+               $text = $br.$br.$image.$text;
        }
        $title = str_replace(array("\r","\n"),array('',''),$title);
 
@@ -313,6 +391,6 @@ function parse_url_content(&$a) {
 
        logger('parse_url: returns: ' . $result);
 
-       echo $result;
+       echo trim($result);
        killme();
 }