Shiny trinkets are shiny.
- +/** + * @file mod/parse_url.php + * + * @todo https://developers.google.com/+/plugins/snippet/ + * + * @verbatim + * + * + * + * + * + *Shiny trinkets are shiny.
+ * + * @endverbatim */ if(!function_exists('deletenode')) { @@ -24,50 +28,157 @@ 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"]; if(strpos($urlarr['path'],'/') !== 0) $complete .= '/'; - $complete .= $urlarr["path"]; + $complete .= $urlarr["path"]; - if ($urlarr["query"] != "") - $complete .= "?".$urlarr["query"]; + if (@$urlarr["query"] != "") + $complete .= "?".$urlarr["query"]; - if ($urlarr["fragment"] != "") - $complete .= "#".$urlarr["fragment"]; + if (@$urlarr["fragment"] != "") + $complete .= "#".$urlarr["fragment"]; - return($complete); + return($complete); } -function parseurl_getsiteinfo($url) { +function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) { + + if ($url == "") + return false; + + $r = q("SELECT * FROM `parsed_url` WHERE `url` = '%s' AND `guessing` = %d AND `oembed` = %d", + dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed)); + + if ($r) + $data = $r[0]["content"]; + + if (!is_null($data)) { + $data = unserialize($data); + return $data; + } + + $data = parseurl_getsiteinfo($url, $no_guessing, $do_oembed); + + q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s') + ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'", + dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed), + dbesc(serialize($data)), dbesc(datetime_convert()), + dbesc(serialize($data)), dbesc(datetime_convert())); + + return $data; +} + +function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1) { + require_once("include/network.php"); + require_once("include/Photo.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"); @@ -77,7 +188,12 @@ 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"); $doc = new DOMDocument(); @@ -97,6 +213,28 @@ function parseurl_getsiteinfo($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 $node) @@ -107,25 +245,60 @@ function parseurl_getsiteinfo($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"); - - 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"] = $attr["content"]; - break; - } + foreach ($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 $keyword) + if (!in_array(trim($keyword), $siteinfo["keywords"])) + $siteinfo["keywords"][] = trim($keyword); } //$list = $xpath->query("head/meta[@property]"); @@ -133,34 +306,35 @@ function parseurl_getsiteinfo($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); + $photodata = get_photo_info($src); if (($photodata) && ($photodata[0] > 150) and ($photodata[1] > 150)) { if ($photodata[0] > 300) { @@ -176,13 +350,13 @@ function parseurl_getsiteinfo($url) { "height"=>$photodata[1]); } - } - } else { + } + } elseif ($siteinfo["image"] != "") { $src = completeurl($siteinfo["image"], $url); unset($siteinfo["image"]); - $photodata = @getimagesize($src); + $photodata = get_photo_info($src); if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10)) $siteinfo["images"][] = array("src"=>$src, @@ -190,7 +364,7 @@ function parseurl_getsiteinfo($url) { "height"=>$photodata[1]); } - if ($siteinfo["text"] == "") { + if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$no_guessing) { $text = ""; $list = $xpath->query("//div[@class='article']"); @@ -219,10 +393,14 @@ function parseurl_getsiteinfo($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); } @@ -232,11 +410,14 @@ function arr_add_hashes(&$item,$k) { function parse_url_content(&$a) { + require_once("include/items.php"); + $text = null; $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) @@ -261,12 +442,21 @@ 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 - $template = "' . trim($text) . '
' . htmlspecialchars(trim($text)) . '
'.trim($text).'