* **memcache_port** - Portnumber of the memcache daemon. Default is 11211.
* **no_count** (Boolean) - Don't do count calculations (currently only when showing albums)
* **no_oembed** (Boolean) - Don't use OEmbed to fetch more information about a link.
-* **no_oembed_rich_content** (Boolean) - Don't show the rich content (e.g. embedded PDF).
* **no_smilies** (Boolean) - Don't show smilies.
* **no_view_full_size** (Boolean) - Don't add the link "View full size" under a resized image.
* **optimize_items** (Boolean) - Triggers an SQL command to optimize the item table before expiring items.
* @file include/items.php
*/
use Friendica\App;
-use Friendica\ParseUrl;
use Friendica\Content\Feature;
use Friendica\Core\Config;
use Friendica\Core\PConfig;
use Friendica\Protocol\DFRN;
use Friendica\Protocol\OStatus;
use Friendica\Protocol\Feed;
+use Friendica\Util\ParseUrl;
require_once 'include/bbcode.php';
require_once 'include/tags.php';
return false;
}
- $str_allowed = Config::get('system', 'allowed_email');
- if (! $str_allowed) {
- return true;
- }
+ $str_allowed = Config::get('system', 'allowed_email', '');
+ $allowed = explode(',', $str_allowed);
- $found = false;
+ return allowed_domain($domain, $allowed);
+}
- $fnmatch = function_exists('fnmatch');
- $allowed = explode(',', $str_allowed);
+/**
+ * Checks for the existence of a domain in a domain list
+ *
+ * If strict is not set, an empty domain list counts as found
+ *
+ * @brief Checks for the existence of a domain in a domain list
+ * @param string $domain
+ * @param array $domain_list
+ * @param bool $strict
+ * @return boolean
+ */
+function allowed_domain($domain, array $domain_list, $strict = false)
+{
+ $found = false;
- if (count($allowed)) {
- foreach ($allowed as $a) {
- $pat = strtolower(trim($a));
- if (($fnmatch && fnmatch($pat, $domain)) || ($pat == $domain)) {
+ if (count($domain_list)) {
+ foreach ($domain_list as $item) {
+ $pat = strtolower(trim($item));
+ if (fnmatch($pat, $domain) || ($pat == $domain)) {
$found = true;
break;
}
}
+ } elseif(!$strict) {
+ $found = true;
}
return $found;
}
* @file include/plaintext.php
*/
use Friendica\App;
-use Friendica\ParseUrl;
use Friendica\Core\PConfig;
use Friendica\Object\Image;
+use Friendica\Util\ParseUrl;
require_once "include/bbcode.php";
require_once "include/html2plain.php";
$allowed_sites = ((x($_POST,'allowed_sites')) ? notags(trim($_POST['allowed_sites'])) : '');
$allowed_email = ((x($_POST,'allowed_email')) ? notags(trim($_POST['allowed_email'])) : '');
+ $no_oembed_rich_content = x($_POST,'no_oembed_rich_content');
+ $allowed_oembed = ((x($_POST,'allowed_embed')) ? notags(trim($_POST['allowed_embed'])) : '');
$block_public = ((x($_POST,'block_public')) ? True : False);
$force_publish = ((x($_POST,'publish_all')) ? True : False);
$global_directory = ((x($_POST,'directory')) ? notags(trim($_POST['directory'])) : '');
Config::set('config', 'register_text', $register_text);
Config::set('system', 'allowed_sites', $allowed_sites);
Config::set('system', 'allowed_email', $allowed_email);
+ Config::set('system', 'no_oembed_rich_content', $no_oembed_rich_content);
+ Config::set('system', 'allowed_oembed', $allowed_oembed);
Config::set('system', 'block_public', $block_public);
Config::set('system', 'publish_all', $force_publish);
Config::set('system', 'newuser_private', $newuser_private);
'$abandon_days' => array('abandon_days', t('Accounts abandoned after x days'), Config::get('system','account_abandon_days'), t('Will not waste system resources polling external sites for abandonded accounts. Enter 0 for no time limit.')),
'$allowed_sites' => array('allowed_sites', t("Allowed friend domains"), Config::get('system','allowed_sites'), t("Comma separated list of domains which are allowed to establish friendships with this site. Wildcards are accepted. Empty to allow any domains")),
'$allowed_email' => array('allowed_email', t("Allowed email domains"), Config::get('system','allowed_email'), t("Comma separated list of domains which are allowed in email addresses for registrations to this site. Wildcards are accepted. Empty to allow any domains")),
+ '$no_oembed_rich_content' => array('no_oembed_rich_content', t("No OEmbed rich content"), Config::get('system','no_oembed_rich_content'), t("Don't show the rich content (e.g. embedded PDF), except from the domains listed below.")),
+ '$allowed_oembed' => array('allowed_oembed', t("Allowed OEmbed domains"), Config::get('system','allowed_oembed'), t("Comma separated list of domains which oembed content is allowed to be displayed. Wildcards are accepted.")),
'$block_public' => array('block_public', t("Block public"), Config::get('system','block_public'), t("Check to block public access to all otherwise public personal pages on this site unless you are currently logged in.")),
'$force_publish' => array('publish_all', t("Force publish"), Config::get('system','publish_all'), t("Check to force all profiles on this site to be listed in the site directory.")),
'$global_directory' => array('directory', t("Global directory URL"), Config::get('system','directory'), t("URL to the global directory. If this is not set, the global directory is completely unavailable to the application.")),
*/
use Friendica\App;
-use Friendica\ParseUrl;
+use Friendica\Util\ParseUrl;
require_once("include/items.php");
\r
use Friendica\Core\Cache;\r
use Friendica\Core\System;\r
-use Friendica\ParseUrl;\r
use Friendica\Core\Config;\r
use Friendica\Database\DBM;\r
+use Friendica\Util\ParseUrl;\r
use dba;\r
use DOMDocument;\r
use DOMXPath;\r
break;\r
case "rich":\r
// not so safe..\r
- if (!Config::get("system", "no_oembed_rich_content")) {\r
- $ret.= proxy_parse_html($jhtml);\r
+ if (self::isAllowedURL($embedurl)) {\r
+ $ret .= proxy_parse_html($jhtml);\r
}\r
break;\r
}\r
}\r
$width = '100%';\r
\r
- $s = System::baseUrl() . '/oembed/' . base64url_encode($src);\r
+ // Only proxy OEmbed URLs to avoid mixed-content errors\r
+ if (Config::get('system', 'ssl_policy') == SSL_POLICY_FULL && parse_url($src, PHP_URL_SCHEME) !== 'https') {\r
+ $src = System::baseUrl() . '/oembed/' . base64url_encode($src);\r
+ }\r
return '<iframe onload="resizeIframe(this);" class="embed_rich" height="' . $height . '" width="' . $width . '" src="' . $s . '" allowfullscreen scrolling="no" frameborder="no">' . t('Embedded content') . '</iframe>';\r
}\r
\r
}\r
return $innerHTML;\r
}\r
+\r
+ /**\r
+ * Determines if rich content OEmbed is allowed for the provided URL\r
+ *\r
+ * @brief Determines if rich content OEmbed is allowed for the provided URL\r
+ * @param string $url\r
+ * @return boolean\r
+ */\r
+ private static function isAllowedURL($url)\r
+ {\r
+ if (!Config::get('system', 'no_oembed_rich_content')) {\r
+ return true;\r
+ }\r
+\r
+ $domain = parse_url($url, PHP_URL_HOST);\r
+\r
+ $str_allowed = Config::get('system', 'allowed_oembed', '');\r
+ $allowed = explode(',', $str_allowed);\r
+\r
+ return allowed_domain($domain, $allowed, true);\r
+ }\r
}\r
+++ /dev/null
-<?php
-/**
- * @file include/ParseUrl.php
- * @brief Get informations about a given URL
- */
-namespace Friendica;
-
-use Friendica\Content\OEmbed;
-use Friendica\Object\Image;
-use Friendica\Util\XML;
-
-use dba;
-use DOMXPath;
-use DOMDocument;
-
-require_once 'include/dba.php';
-require_once "include/network.php";
-
-/**
- * @brief Class with methods for extracting certain content from an url
- */
-class ParseUrl
-{
- /**
- * @brief Search for chached embeddable data of an url otherwise fetch it
- *
- * @param string $url The url of the page which should be scraped
- * @param bool $no_guessing If true the parse doens't search for
- * preview pictures
- * @param bool $do_oembed The false option is used by the function fetch_oembed()
- * to avoid endless loops
- *
- * @return array which contains needed data for embedding
- * string 'url' => The url of the parsed page
- * string 'type' => Content type
- * string 'title' => The title of the content
- * string 'text' => The description for the content
- * string 'image' => A preview image of the content (only available
- * if $no_geuessing = false
- * array'images' = Array of preview pictures
- * string 'keywords' => The tags which belong to the content
- *
- * @see ParseUrl::getSiteinfo() for more information about scraping
- * embeddable content
- */
- public static function getSiteinfoCached($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;
- }
-
- $data = self::getSiteinfo($url, $no_guessing, $do_oembed);
-
- dba::insert(
- 'parsed_url',
- array(
- 'url' => normalise_link($url), 'guessing' => !$no_guessing,
- 'oembed' => $do_oembed, 'content' => serialize($data),
- 'created' => datetime_convert()),
- true
- );
-
- return $data;
- }
- /**
- * @brief Parse a page for embeddable content information
- *
- * This method parses to url for meta data which can be used to embed
- * the content. If available it prioritizes Open Graph meta tags.
- * If this is not available it uses the twitter cards meta tags.
- * As fallback it uses standard html elements with meta informations
- * like \<title\>Awesome Title\</title\> or
- * \<meta name="description" content="An awesome description"\>
- *
- * @param string $url The url of the page which should be scraped
- * @param bool $no_guessing If true the parse doens't search for
- * preview pictures
- * @param bool $do_oembed The false option is used by the function fetch_oembed()
- * to avoid endless loops
- * @param int $count Internal counter to avoid endless loops
- *
- * @return array which contains needed data for embedding
- * string 'url' => The url of the parsed page
- * string 'type' => Content type
- * string 'title' => The title of the content
- * string 'text' => The description for the content
- * string 'image' => A preview image of the content (only available
- * if $no_geuessing = false
- * array'images' = Array of preview pictures
- * string 'keywords' => The tags which belong to the content
- *
- * @todo https://developers.google.com/+/plugins/snippet/
- * @verbatim
- * <meta itemprop="name" content="Awesome title">
- * <meta itemprop="description" content="An awesome description">
- * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
- *
- * <body itemscope itemtype="http://schema.org/Product">
- * <h1 itemprop="name">Shiny Trinket</h1>
- * <img itemprop="image" src="{image-url}" />
- * <p itemprop="description">Shiny trinkets are shiny.</p>
- * </body>
- * @endverbatim
- */
- public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
- {
- $a = get_app();
-
- $siteinfo = array();
-
- // Check if the URL does contain a scheme
- $scheme = parse_url($url, PHP_URL_SCHEME);
-
- if ($scheme == "") {
- $url = "http://".trim($url, "/");
- }
-
- if ($count > 10) {
- logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
- return($siteinfo);
- }
-
- $url = trim($url, "'");
- $url = trim($url, '"');
-
- $url = strip_tracking_query_params($url);
-
- $siteinfo["url"] = $url;
- $siteinfo["type"] = "link";
-
- $data = z_fetch_url($url);
- if (!$data['success']) {
- return($siteinfo);
- }
-
- // If the file is too large then exit
- if ($data["info"]["download_content_length"] > 1000000) {
- return($siteinfo);
- }
-
- // If it isn't a HTML file then exit
- if (($data["info"]["content_type"] != "") && !strstr(strtolower($data["info"]["content_type"]), "html")) {
- return($siteinfo);
- }
-
- $header = $data["header"];
- $body = $data["body"];
-
- if ($do_oembed) {
- $oembed_data = OEmbed::fetchURL($url);
-
- if (!in_array($oembed_data->type, array("error", "rich", ""))) {
- $siteinfo["type"] = $oembed_data->type;
- }
-
- if (($oembed_data->type == "link") && ($siteinfo["type"] != "photo")) {
- if (isset($oembed_data->title)) {
- $siteinfo["title"] = trim($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;
- }
- }
- }
-
- // 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 == "") {
- $charset = "utf-8";
- }
-
- if (($charset != "") && (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");
-
- $doc = new DOMDocument();
- @$doc->loadHTML($body);
-
- XML::deleteNode($doc, "style");
- XML::deleteNode($doc, "script");
- XML::deleteNode($doc, "option");
- XML::deleteNode($doc, "h1");
- XML::deleteNode($doc, "h2");
- XML::deleteNode($doc, "h3");
- XML::deleteNode($doc, "h4");
- XML::deleteNode($doc, "h5");
- XML::deleteNode($doc, "h6");
- XML::deleteNode($doc, "ol");
- XML::deleteNode($doc, "ul");
-
- $xpath = new DOMXPath($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 != "") {
- $siteinfo = self::getSiteinfo($content, $no_guessing, $do_oembed, ++$count);
- return($siteinfo);
- }
- }
- }
-
- $list = $xpath->query("//title");
- if ($list->length > 0) {
- $siteinfo["title"] = trim($list->item(0)->nodeValue);
- }
-
- //$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":
- $siteinfo["title"] = trim($attr["content"]);
- break;
- case "description":
- $siteinfo["text"] = trim($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"] == "") || ($attr["content"] == "photo")) {
- $siteinfo["type"] = $attr["content"];
- }
- break;
- case "twitter:description":
- $siteinfo["text"] = trim($attr["content"]);
- break;
- case "twitter:title":
- $siteinfo["title"] = trim($attr["content"]);
- break;
- case "dc.title":
- $siteinfo["title"] = trim($attr["content"]);
- break;
- case "dc.description":
- $siteinfo["text"] = trim($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]");
- $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":
- $siteinfo["title"] = trim($attr["content"]);
- break;
- case "og:description":
- $siteinfo["text"] = trim($attr["content"]);
- break;
- }
- }
- }
-
- if ((@$siteinfo["image"] == "") && !$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;
- }
- }
-
- $src = self::completeUrl($attr["src"], $url);
- $photodata = Image::getInfoFromURL($src);
-
- if (($photodata) && ($photodata[0] > 150) && ($photodata[1] > 150)) {
- if ($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" => $photodata[1]);
- }
- }
- } elseif ($siteinfo["image"] != "") {
- $src = self::completeUrl($siteinfo["image"], $url);
-
- unset($siteinfo["image"]);
-
- $photodata = Image::getInfoFromURL($src);
-
- if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) {
- $siteinfo["images"][] = array("src" => $src,
- "width" => $photodata[0],
- "height" => $photodata[1]);
- }
- }
-
- if ((@$siteinfo["text"] == "") && (@$siteinfo["title"] != "") && !$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);
- }
-
- /**
- * @brief Convert tags from CSV to an array
- *
- * @param string $string Tags
- * @return array with formatted Hashtags
- */
- public static function convertTagsToArray($string)
- {
- $arr_tags = str_getcsv($string);
- if (count($arr_tags)) {
- // add the # sign to every tag
- array_walk($arr_tags, array("self", "arrAddHashes"));
-
- return $arr_tags;
- }
- }
-
- /**
- * @brief Add a hasht sign to a string
- *
- * This method is used as callback function
- *
- * @param string $tag The pure tag name
- * @param int $k Counter for internal use
- * @return void
- */
- private static function arrAddHashes(&$tag, $k)
- {
- $tag = "#" . $tag;
- }
-
- /**
- * @brief Add a scheme to an url
- *
- * The src attribute of some html elements (e.g. images)
- * can miss the scheme so we need to add the correct
- * scheme
- *
- * @param string $url The url which possibly does have
- * a missing scheme (a link to an image)
- * @param string $scheme The url with a correct scheme
- * (e.g. the url from the webpage which does contain the image)
- *
- * @return string The url with a scheme
- */
- private static function completeUrl($url, $scheme)
- {
- $urlarr = parse_url($url);
-
- // If the url does allready have an scheme
- // we can stop the process here
- 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);
- }
-}
--- /dev/null
+<?php
+/**
+ * @file src/Util/ParseUrl.php
+ * @brief Get informations about a given URL
+ */
+namespace Friendica\Util;
+
+use Friendica\Content\OEmbed;
+use Friendica\Object\Image;
+use Friendica\Util\XML;
+
+use dba;
+use DOMXPath;
+use DOMDocument;
+
+require_once 'include/dba.php';
+require_once "include/network.php";
+
+/**
+ * @brief Class with methods for extracting certain content from an url
+ */
+class ParseUrl
+{
+ /**
+ * @brief Search for chached embeddable data of an url otherwise fetch it
+ *
+ * @param string $url The url of the page which should be scraped
+ * @param bool $no_guessing If true the parse doens't search for
+ * preview pictures
+ * @param bool $do_oembed The false option is used by the function fetch_oembed()
+ * to avoid endless loops
+ *
+ * @return array which contains needed data for embedding
+ * string 'url' => The url of the parsed page
+ * string 'type' => Content type
+ * string 'title' => The title of the content
+ * string 'text' => The description for the content
+ * string 'image' => A preview image of the content (only available
+ * if $no_geuessing = false
+ * array'images' = Array of preview pictures
+ * string 'keywords' => The tags which belong to the content
+ *
+ * @see ParseUrl::getSiteinfo() for more information about scraping
+ * embeddable content
+ */
+ public static function getSiteinfoCached($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;
+ }
+
+ $data = self::getSiteinfo($url, $no_guessing, $do_oembed);
+
+ dba::insert(
+ 'parsed_url',
+ array(
+ 'url' => normalise_link($url), 'guessing' => !$no_guessing,
+ 'oembed' => $do_oembed, 'content' => serialize($data),
+ 'created' => datetime_convert()),
+ true
+ );
+
+ return $data;
+ }
+ /**
+ * @brief Parse a page for embeddable content information
+ *
+ * This method parses to url for meta data which can be used to embed
+ * the content. If available it prioritizes Open Graph meta tags.
+ * If this is not available it uses the twitter cards meta tags.
+ * As fallback it uses standard html elements with meta informations
+ * like \<title\>Awesome Title\</title\> or
+ * \<meta name="description" content="An awesome description"\>
+ *
+ * @param string $url The url of the page which should be scraped
+ * @param bool $no_guessing If true the parse doens't search for
+ * preview pictures
+ * @param bool $do_oembed The false option is used by the function fetch_oembed()
+ * to avoid endless loops
+ * @param int $count Internal counter to avoid endless loops
+ *
+ * @return array which contains needed data for embedding
+ * string 'url' => The url of the parsed page
+ * string 'type' => Content type
+ * string 'title' => The title of the content
+ * string 'text' => The description for the content
+ * string 'image' => A preview image of the content (only available
+ * if $no_geuessing = false
+ * array'images' = Array of preview pictures
+ * string 'keywords' => The tags which belong to the content
+ *
+ * @todo https://developers.google.com/+/plugins/snippet/
+ * @verbatim
+ * <meta itemprop="name" content="Awesome title">
+ * <meta itemprop="description" content="An awesome description">
+ * <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
+ *
+ * <body itemscope itemtype="http://schema.org/Product">
+ * <h1 itemprop="name">Shiny Trinket</h1>
+ * <img itemprop="image" src="{image-url}" />
+ * <p itemprop="description">Shiny trinkets are shiny.</p>
+ * </body>
+ * @endverbatim
+ */
+ public static function getSiteinfo($url, $no_guessing = false, $do_oembed = true, $count = 1)
+ {
+ $a = get_app();
+
+ $siteinfo = array();
+
+ // Check if the URL does contain a scheme
+ $scheme = parse_url($url, PHP_URL_SCHEME);
+
+ if ($scheme == "") {
+ $url = "http://".trim($url, "/");
+ }
+
+ if ($count > 10) {
+ logger("parseurl_getsiteinfo: Endless loop detected for ".$url, LOGGER_DEBUG);
+ return($siteinfo);
+ }
+
+ $url = trim($url, "'");
+ $url = trim($url, '"');
+
+ $url = strip_tracking_query_params($url);
+
+ $siteinfo["url"] = $url;
+ $siteinfo["type"] = "link";
+
+ $data = z_fetch_url($url);
+ if (!$data['success']) {
+ return($siteinfo);
+ }
+
+ // If the file is too large then exit
+ if ($data["info"]["download_content_length"] > 1000000) {
+ return($siteinfo);
+ }
+
+ // If it isn't a HTML file then exit
+ if (($data["info"]["content_type"] != "") && !strstr(strtolower($data["info"]["content_type"]), "html")) {
+ return($siteinfo);
+ }
+
+ $header = $data["header"];
+ $body = $data["body"];
+
+ if ($do_oembed) {
+ $oembed_data = OEmbed::fetchURL($url);
+
+ if (!in_array($oembed_data->type, array("error", "rich", ""))) {
+ $siteinfo["type"] = $oembed_data->type;
+ }
+
+ if (($oembed_data->type == "link") && ($siteinfo["type"] != "photo")) {
+ if (isset($oembed_data->title)) {
+ $siteinfo["title"] = trim($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;
+ }
+ }
+ }
+
+ // 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 == "") {
+ $charset = "utf-8";
+ }
+
+ if (($charset != "") && (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");
+
+ $doc = new DOMDocument();
+ @$doc->loadHTML($body);
+
+ XML::deleteNode($doc, "style");
+ XML::deleteNode($doc, "script");
+ XML::deleteNode($doc, "option");
+ XML::deleteNode($doc, "h1");
+ XML::deleteNode($doc, "h2");
+ XML::deleteNode($doc, "h3");
+ XML::deleteNode($doc, "h4");
+ XML::deleteNode($doc, "h5");
+ XML::deleteNode($doc, "h6");
+ XML::deleteNode($doc, "ol");
+ XML::deleteNode($doc, "ul");
+
+ $xpath = new DOMXPath($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 != "") {
+ $siteinfo = self::getSiteinfo($content, $no_guessing, $do_oembed, ++$count);
+ return($siteinfo);
+ }
+ }
+ }
+
+ $list = $xpath->query("//title");
+ if ($list->length > 0) {
+ $siteinfo["title"] = trim($list->item(0)->nodeValue);
+ }
+
+ //$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":
+ $siteinfo["title"] = trim($attr["content"]);
+ break;
+ case "description":
+ $siteinfo["text"] = trim($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"] == "") || ($attr["content"] == "photo")) {
+ $siteinfo["type"] = $attr["content"];
+ }
+ break;
+ case "twitter:description":
+ $siteinfo["text"] = trim($attr["content"]);
+ break;
+ case "twitter:title":
+ $siteinfo["title"] = trim($attr["content"]);
+ break;
+ case "dc.title":
+ $siteinfo["title"] = trim($attr["content"]);
+ break;
+ case "dc.description":
+ $siteinfo["text"] = trim($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]");
+ $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":
+ $siteinfo["title"] = trim($attr["content"]);
+ break;
+ case "og:description":
+ $siteinfo["text"] = trim($attr["content"]);
+ break;
+ }
+ }
+ }
+
+ if ((@$siteinfo["image"] == "") && !$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;
+ }
+ }
+
+ $src = self::completeUrl($attr["src"], $url);
+ $photodata = Image::getInfoFromURL($src);
+
+ if (($photodata) && ($photodata[0] > 150) && ($photodata[1] > 150)) {
+ if ($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" => $photodata[1]);
+ }
+ }
+ } elseif ($siteinfo["image"] != "") {
+ $src = self::completeUrl($siteinfo["image"], $url);
+
+ unset($siteinfo["image"]);
+
+ $photodata = Image::getInfoFromURL($src);
+
+ if (($photodata) && ($photodata[0] > 10) && ($photodata[1] > 10)) {
+ $siteinfo["images"][] = array("src" => $src,
+ "width" => $photodata[0],
+ "height" => $photodata[1]);
+ }
+ }
+
+ if ((@$siteinfo["text"] == "") && (@$siteinfo["title"] != "") && !$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);
+ }
+
+ /**
+ * @brief Convert tags from CSV to an array
+ *
+ * @param string $string Tags
+ * @return array with formatted Hashtags
+ */
+ public static function convertTagsToArray($string)
+ {
+ $arr_tags = str_getcsv($string);
+ if (count($arr_tags)) {
+ // add the # sign to every tag
+ array_walk($arr_tags, array("self", "arrAddHashes"));
+
+ return $arr_tags;
+ }
+ }
+
+ /**
+ * @brief Add a hasht sign to a string
+ *
+ * This method is used as callback function
+ *
+ * @param string $tag The pure tag name
+ * @param int $k Counter for internal use
+ * @return void
+ */
+ private static function arrAddHashes(&$tag, $k)
+ {
+ $tag = "#" . $tag;
+ }
+
+ /**
+ * @brief Add a scheme to an url
+ *
+ * The src attribute of some html elements (e.g. images)
+ * can miss the scheme so we need to add the correct
+ * scheme
+ *
+ * @param string $url The url which possibly does have
+ * a missing scheme (a link to an image)
+ * @param string $scheme The url with a correct scheme
+ * (e.g. the url from the webpage which does contain the image)
+ *
+ * @return string The url with a scheme
+ */
+ private static function completeUrl($url, $scheme)
+ {
+ $urlarr = parse_url($url);
+
+ // If the url does allready have an scheme
+ // we can stop the process here
+ 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);
+ }
+}
'Friendica\\Model\\Profile' => $baseDir . '/src/Model/Profile.php',
'Friendica\\Model\\User' => $baseDir . '/src/Model/User.php',
'Friendica\\Module\\Feed' => $baseDir . '/src/Module/Feed.php',
- 'Friendica\\Module\\Login' => $baseDir . '/src/Module/Login.php',
- 'Friendica\\Module\\Logout' => $baseDir . '/src/Module/Logout.php',
'Friendica\\Module\\Oembed' => $baseDir . '/src/Module/Oembed.php',
'Friendica\\Network\\FKOAuth1' => $baseDir . '/src/Network/FKOAuth1.php',
'Friendica\\Network\\FKOAuthDataStore' => $baseDir . '/src/Network/FKOAuthDataStore.php',
'Friendica\\Object\\Image' => $baseDir . '/src/Object/Image.php',
'Friendica\\Object\\Post' => $baseDir . '/src/Object/Post.php',
'Friendica\\Object\\Thread' => $baseDir . '/src/Object/Thread.php',
- 'Friendica\\ParseUrl' => $baseDir . '/src/ParseUrl.php',
'Friendica\\Protocol\\DFRN' => $baseDir . '/src/Protocol/DFRN.php',
'Friendica\\Protocol\\Diaspora' => $baseDir . '/src/Protocol/Diaspora.php',
'Friendica\\Protocol\\Email' => $baseDir . '/src/Protocol/Email.php',
'Friendica\\Util\\ExAuth' => $baseDir . '/src/Util/ExAuth.php',
'Friendica\\Util\\Lock' => $baseDir . '/src/Util/Lock.php',
'Friendica\\Util\\Map' => $baseDir . '/src/Util/Map.php',
+ 'Friendica\\Util\\ParseUrl' => $baseDir . '/src/Util/ParseUrl.php',
'Friendica\\Util\\PidFile' => $baseDir . '/src/Util/Pidfile.php',
'Friendica\\Util\\XML' => $baseDir . '/src/Util/XML.php',
'Friendica\\Worker\\CheckVersion' => $baseDir . '/src/Worker/CheckVersion.php',
'Friendica\\Model\\Profile' => __DIR__ . '/../..' . '/src/Model/Profile.php',
'Friendica\\Model\\User' => __DIR__ . '/../..' . '/src/Model/User.php',
'Friendica\\Module\\Feed' => __DIR__ . '/../..' . '/src/Module/Feed.php',
- 'Friendica\\Module\\Login' => __DIR__ . '/../..' . '/src/Module/Login.php',
- 'Friendica\\Module\\Logout' => __DIR__ . '/../..' . '/src/Module/Logout.php',
'Friendica\\Module\\Oembed' => __DIR__ . '/../..' . '/src/Module/Oembed.php',
'Friendica\\Network\\FKOAuth1' => __DIR__ . '/../..' . '/src/Network/FKOAuth1.php',
'Friendica\\Network\\FKOAuthDataStore' => __DIR__ . '/../..' . '/src/Network/FKOAuthDataStore.php',
'Friendica\\Object\\Image' => __DIR__ . '/../..' . '/src/Object/Image.php',
'Friendica\\Object\\Post' => __DIR__ . '/../..' . '/src/Object/Post.php',
'Friendica\\Object\\Thread' => __DIR__ . '/../..' . '/src/Object/Thread.php',
- 'Friendica\\ParseUrl' => __DIR__ . '/../..' . '/src/ParseUrl.php',
'Friendica\\Protocol\\DFRN' => __DIR__ . '/../..' . '/src/Protocol/DFRN.php',
'Friendica\\Protocol\\Diaspora' => __DIR__ . '/../..' . '/src/Protocol/Diaspora.php',
'Friendica\\Protocol\\Email' => __DIR__ . '/../..' . '/src/Protocol/Email.php',
'Friendica\\Util\\ExAuth' => __DIR__ . '/../..' . '/src/Util/ExAuth.php',
'Friendica\\Util\\Lock' => __DIR__ . '/../..' . '/src/Util/Lock.php',
'Friendica\\Util\\Map' => __DIR__ . '/../..' . '/src/Util/Map.php',
+ 'Friendica\\Util\\ParseUrl' => __DIR__ . '/../..' . '/src/Util/ParseUrl.php',
'Friendica\\Util\\PidFile' => __DIR__ . '/../..' . '/src/Util/Pidfile.php',
'Friendica\\Util\\XML' => __DIR__ . '/../..' . '/src/Util/XML.php',
'Friendica\\Worker\\CheckVersion' => __DIR__ . '/../..' . '/src/Worker/CheckVersion.php',
<h3>{{$corporate}}</h3>
{{include file="field_input.tpl" field=$allowed_sites}}
{{include file="field_input.tpl" field=$allowed_email}}
+ {{include file="field_checkbox.tpl" field=$no_oembed_rich_content}}
+ {{include file="field_input.tpl" field=$allowed_oembed}}
{{include file="field_checkbox.tpl" field=$block_public}}
{{include file="field_checkbox.tpl" field=$force_publish}}
{{include file="field_select.tpl" field=$community_page_style}}
projects / friendica.git / commitdiff