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')) { @@ -50,8 +54,36 @@ function completeurl($url, $scheme) { return($complete); } +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"); $a = get_app(); @@ -97,6 +129,14 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co return($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"); @@ -104,15 +144,16 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co if ($oembed_data->type != "error") $siteinfo["type"] = $oembed_data->type; - } - // if the file is too large then exit - if ($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 (($oembed_data->type == "link") AND ($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; + } + } $stamp1 = microtime(true); @@ -194,10 +235,9 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co } } - //$list = $xpath->query("head/title"); $list = $xpath->query("//title"); - foreach ($list as $node) - $siteinfo["title"] = html_entity_decode($node->nodeValue, ENT_QUOTES, "UTF-8"); + if ($list->length > 0) + $siteinfo["title"] = $list->item(0)->nodeValue; //$list = $xpath->query("head/meta[@name]"); $list = $xpath->query("//meta[@name]"); @@ -256,7 +296,8 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co if (isset($keywords)) { $siteinfo["keywords"] = array(); foreach ($keywords as $keyword) - $siteinfo["keywords"][] = trim($keyword); + if (!in_array(trim($keyword), $siteinfo["keywords"])) + $siteinfo["keywords"][] = trim($keyword); } //$list = $xpath->query("head/meta[@property]"); @@ -283,15 +324,6 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co } } - if (isset($oembed_data) AND ($oembed_data->type == "link") AND ($siteinfo["type"] != "photo")) { - if (isset($oembed_data->title) AND (trim($oembed_data->title) != "")) - $siteinfo["title"] = $oembed_data->title; - if (isset($oembed_data->description) AND (trim($oembed_data->description) != "")) - $siteinfo["text"] = trim($oembed_data->description); - if (isset($oembed_data->thumbnail_url) AND (trim($oembed_data->thumbnail_url) != "")) - $siteinfo["image"] = $oembed_data->thumbnail_url; - } - if ((@$siteinfo["image"] == "") AND !$no_guessing) { $list = $xpath->query("//img[@src]"); foreach ($list as $node) { @@ -301,7 +333,7 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co $attr[$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) { @@ -318,12 +350,12 @@ function parseurl_getsiteinfo($url, $no_guessing = false, $do_oembed = true, $co } } - } else { + } elseif ($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, @@ -377,6 +409,8 @@ function arr_add_hashes(&$item,$k) { function parse_url_content(&$a) { + require_once("include/items.php"); + $text = null; $str_tags = ''; @@ -454,77 +488,15 @@ function parse_url_content(&$a) { $siteinfo = parseurl_getsiteinfo($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($siteinfo["images"]) > 0)){ - /* Execute below code only if image is present in siteinfo */ + unset($siteinfo["keywords"]); - $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 - $image .= ''.htmlspecialchars(trim($text)).''; - } + if (!$textmode) + // Replace ' with â - not perfect - but the richtext editor has problems otherwise + $info = str_replace(array("'"), array("’"), $info); - if($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]"; + echo $info; killme(); }