parse_url: recognize image/video/audio files + move functions into own class

This commit is contained in:
rabuzarus 2016-11-24 01:11:22 +01:00
parent 0628068977
commit e9226eaf45
5 changed files with 625 additions and 486 deletions

482
include/ParseUrl.php Normal file
View file

@ -0,0 +1,482 @@
<?php
/**
* @file include/ParseUrl.php
* @brief Get informations about a given URL
*/
namespace Friendica;
use \Friendica\Core\Config;
require_once("include/network.php");
require_once("include/Photo.php");
require_once("include/oembed.php");
/**
* @brief Class with methods for extracting certain content from an url
*/
class ParseUrl {
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, $do_oembed);
q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')
ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'",
dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed),
dbesc(serialize($data)), dbesc(datetime_convert()),
dbesc(serialize($data)), dbesc(datetime_convert()));
return $data;
}
public static function getSiteinfo($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, '"');
$url = original_url($url);
$siteinfo["url"] = $url;
$siteinfo["type"] = "link";
$check_cert = Config::get("system", "verifyssl");
$stamp1 = microtime(true);
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, 1);
curl_setopt($ch, CURLOPT_NOBODY, 1);
curl_setopt($ch, CURLOPT_TIMEOUT, 3);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
$header = curl_exec($ch);
$curl_info = @curl_getinfo($ch);
$http_code = $curl_info["http_code"];
curl_close($ch);
$a->save_timestamp($stamp1, "network");
if ((($curl_info["http_code"] == "301") || ($curl_info["http_code"] == "302") || ($curl_info["http_code"] == "303") || ($curl_info["http_code"] == "307"))
&& (($curl_info["redirect_url"] != "") || ($curl_info["location"] != ""))) {
if ($curl_info["redirect_url"] != "") {
$siteinfo = self::getSiteinfo($curl_info["redirect_url"], $no_guessing, $do_oembed, ++$count);
} else {
$siteinfo = self::getSiteinfo($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"] != "") && !strstr(strtolower($curl_info["content_type"]), "html")) {
return($siteinfo);
}
if ($do_oembed) {
$oembed_data = oembed_fetch_url($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"] = $oembed_data->title;
}
if (isset($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, $a->get_useragent());
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
$header = curl_exec($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)) {
$charset = trim(trim(trim(array_pop($matches)), ';,'));
}
if ($charset == "") {
$charset = "utf-8";
}
$pos = strpos($header, "\r\n\r\n");
if ($pos) {
$body = trim(substr($header, $pos));
} else {
$body = $header;
}
if (($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($body);
self::deleteNode($doc, "style");
self::deleteNode($doc, "script");
self::deleteNode($doc, "option");
self::deleteNode($doc, "h1");
self::deleteNode($doc, "h2");
self::deleteNode($doc, "h3");
self::deleteNode($doc, "h4");
self::deleteNode($doc, "h5");
self::deleteNode($doc, "h6");
self::deleteNode($doc, "ol");
self::deleteNode($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"] = $list->item(0)->nodeValue;
}
//$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"] = $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"] == "") || ($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]");
$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"] = $attr["content"];
break;
case "og:description":
$siteinfo["text"] = $attr["content"];
break;
}
}
}
if ((@$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);
$photodata = get_photo_info($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"]);
$photodata = get_photo_info($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 $k Counter for internal use
*/
private static function arrAddHashes(&$tag, $k) {
$tag = "#" . $tag;
}
private static function deleteNode(&$doc, $node) {
$xpath = new \DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child) {
$child->parentNode->removeChild($child);
}
}
private static 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);
}
}

View file

@ -1,5 +1,11 @@
<?php
/**
* @file include/items.php
*/
use \Friendica\ParseUrl;
require_once('include/bbcode.php');
require_once('include/oembed.php');
require_once('include/salmon.php');
@ -216,9 +222,8 @@ function add_page_info_data($data) {
}
function query_page_info($url, $no_photos = false, $photo = "", $keywords = false, $keyword_blacklist = "") {
require_once("mod/parse_url.php");
$data = parseurl_getsiteinfo_cached($url, true);
$data = ParseUrl::getSiteinfoCached($url, true);
if ($photo != "")
$data["images"][0]["src"] = $photo;

View file

@ -1,8 +1,17 @@
<?php
/**
* @file include/oembed.php
*/
use \Friendica\ParseUrl;
use \Friendica\Core\Config;
function oembed_replacecb($matches){
$embedurl=$matches[1];
$j = oembed_fetch_url($embedurl);
$s = oembed_format_object($j);
return $s;
}
@ -66,7 +75,7 @@ function oembed_fetch_url($embedurl, $no_rich_type = false){
}
if ($txt==false || $txt=="") {
$embedly = get_config("system", "embedly");
$embedly = Config::get("system", "embedly");
if ($embedly != "") {
// try embedly service
$ourl = "https://api.embed.ly/1/oembed?key=".$embedly."&url=".urlencode($embedurl);
@ -110,8 +119,7 @@ function oembed_fetch_url($embedurl, $no_rich_type = false){
// If fetching information doesn't work, then improve via internal functions
if (($j->type == "error") OR ($no_rich_type AND ($j->type == "rich"))) {
require_once("mod/parse_url.php");
$data = parseurl_getsiteinfo_cached($embedurl, true, false);
$data = ParseUrl::getSiteinfoCached($embedurl, true, false);
$j->type = $data["type"];
if ($j->type == "photo") {
@ -143,12 +151,11 @@ function oembed_fetch_url($embedurl, $no_rich_type = false){
function oembed_format_object($j){
require_once("mod/proxy.php");
$a = get_app();
$embedurl = $j->embedurl;
$jhtml = oembed_iframe($j->embedurl,(isset($j->width) ? $j->width : null), (isset($j->height) ? $j->height : null) );
$ret="<span class='oembed ".$j->type."'>";
switch ($j->type) {
case "video": {
case "video":
if (isset($j->thumbnail_url)) {
$tw = (isset($j->thumbnail_width) && intval($j->thumbnail_width)) ? $j->thumbnail_width:200;
$th = (isset($j->thumbnail_height) && intval($j->thumbnail_height)) ? $j->thumbnail_height:180;
@ -158,7 +165,7 @@ function oembed_format_object($j){
$th=120; $tw = $th*$tr;
$tpl=get_markup_template('oembed_video.tpl');
$ret.=replace_macros($tpl, array(
'$baseurl' => $a->get_baseurl(),
'$baseurl' => App::get_baseurl(),
'$embedurl'=>$embedurl,
'$escapedhtml'=>base64_encode($jhtml),
'$tw'=>$tw,
@ -170,43 +177,49 @@ function oembed_format_object($j){
$ret=$jhtml;
}
//$ret.="<br>";
}; break;
case "photo": {
break;
case "photo":
$ret.= "<img width='".$j->width."' src='".proxy_url($j->url)."'>";
}; break;
case "link": {
}; break;
case "rich": {
break;
case "link":
break;
case "rich":
// not so safe..
if (!get_config("system","no_oembed_rich_content"))
if (!Config::get("system","no_oembed_rich_content")) {
$ret.= proxy_parse_html($jhtml);
}; break;
}
break;
}
// add link to source if not present in "rich" type
if ($j->type!='rich' || !strpos($j->html,$embedurl) ){
$ret .= "<h4>";
if (isset($j->title)) {
if (isset($j->provider_name))
if (isset($j->provider_name)) {
$ret .= $j->provider_name.": ";
}
$embedlink = (isset($j->title))?$j->title:$embedurl;
$ret .= "<a href='$embedurl' rel='oembed'>$embedlink</a>";
if (isset($j->author_name))
if (isset($j->author_name)) {
$ret.=" (".$j->author_name.")";
}
} elseif (isset($j->provider_name) OR isset($j->author_name)) {
$embedlink = "";
if (isset($j->provider_name))
if (isset($j->provider_name)) {
$embedlink .= $j->provider_name;
}
if (isset($j->author_name)) {
if ($embedlink != "")
if ($embedlink != "") {
$embedlink .= ": ";
}
$embedlink .= $j->author_name;
}
if (trim($embedlink) == "")
if (trim($embedlink) == "") {
$embedlink = $embedurl;
}
$ret .= "<a href='$embedurl' rel='oembed'>$embedlink</a>";
}
@ -247,15 +260,14 @@ function oembed_iframe($src, $width, $height) {
}
$width = '100%';
$a = get_app();
$s = $a->get_baseurl() . '/oembed/'.base64url_encode($src);
$s = App::get_baseurl() . '/oembed/'.base64url_encode($src);
return '<iframe onload="resizeIframe(this);" class="embed_rich" height="' . $height . '" width="' . $width . '" src="' . $s . '" scrolling="no" frameborder="no">' . t('Embedded content') . '</iframe>';
}
function oembed_bbcode2html($text){
$stopoembed = get_config("system","no_oembed");
$stopoembed = Config::get("system","no_oembed");
if ($stopoembed == true){
return preg_replace("/\[embed\](.+?)\[\/embed\]/is", "<!-- oembed $1 --><i>". t('Embedding disabled') ." : $1</i><!-- /oembed $1 -->" ,$text);
}
@ -268,13 +280,13 @@ function oe_build_xpath($attr, $value){
return "contains( normalize-space( @$attr ), ' $value ' ) or substring( normalize-space( @$attr ), 1, string-length( '$value' ) + 1 ) = '$value ' or substring( normalize-space( @$attr ), string-length( @$attr ) - string-length( '$value' ) ) = ' $value' or @$attr = '$value'";
}
function oe_get_inner_html( $node ) {
$innerHTML= '';
$children = $node->childNodes;
foreach ($children as $child) {
$innerHTML .= $child->ownerDocument->saveXML( $child );
}
return $innerHTML;
function oe_get_inner_html($node) {
$innerHTML= '';
$children = $node->childNodes;
foreach ($children as $child) {
$innerHTML .= $child->ownerDocument->saveXML($child);
}
return $innerHTML;
}
/**
@ -283,15 +295,16 @@ function oe_get_inner_html( $node ) {
*/
function oembed_html2bbcode($text) {
// start parser only if 'oembed' is in text
if (strpos($text, "oembed")){
if (strpos($text, "oembed")) {
// convert non ascii chars to html entities
$html_text = mb_convert_encoding($text, 'HTML-ENTITIES', mb_detect_encoding($text));
// If it doesn't parse at all, just return the text.
$dom = @DOMDocument::loadHTML($html_text);
if(! $dom)
if (! $dom) {
return $text;
}
$xpath = new DOMXPath($dom);
$attr = "oembed";

View file

@ -1,6 +1,15 @@
<?php
/**
* @file include/plaintext.php
*/
use \Friendica\ParseUrl;
require_once("include/Photo.php");
require_once("include/bbcode.php");
require_once("include/html2plain.php");
require_once("include/network.php");
/**
* @brief Fetches attachment data that were generated the old way
@ -181,20 +190,17 @@ function get_attached_data($body) {
// if nothing is found, it maybe having an image.
if (!isset($post["type"])) {
require_once("mod/parse_url.php");
require_once("include/Photo.php");
$URLSearchString = "^\[\]";
if (preg_match_all("(\[url=([$URLSearchString]*)\]\s*\[img\]([$URLSearchString]*)\[\/img\]\s*\[\/url\])ism", $body, $pictures, PREG_SET_ORDER)) {
if (count($pictures) == 1) {
// Checking, if the link goes to a picture
$data = parseurl_getsiteinfo_cached($pictures[0][1], true);
$data = ParseUrl::getSiteinfoCached($pictures[0][1], true);
// Workaround:
// Sometimes photo posts to the own album are not detected at the start.
// So we seem to cannot use the cache for these cases. That's strange.
if (($data["type"] != "photo") AND strstr($pictures[0][1], "/photos/"))
$data = parseurl_getsiteinfo($pictures[0][1], true);
$data = ParseUrl::getSiteinfo($pictures[0][1], true);
if ($data["type"] == "photo") {
$post["type"] = "photo";
@ -246,8 +252,7 @@ function get_attached_data($body) {
$post["text"] = trim($body);
}
} elseif (isset($post["url"]) AND ($post["type"] == "video")) {
require_once("mod/parse_url.php");
$data = parseurl_getsiteinfo_cached($post["url"], true);
$data = ParseUrl::getSiteinfoCached($post["url"], true);
if (isset($data["images"][0]))
$post["image"] = $data["images"][0]["src"];
@ -288,9 +293,6 @@ function shortenmsg($msg, $limit, $twitter = false) {
* @return string The converted message
*/
function plaintext($a, $b, $limit = 0, $includedlinks = false, $htmlmode = 2, $target_network = "") {
require_once("include/bbcode.php");
require_once("include/html2plain.php");
require_once("include/network.php");
// Remove the hash tags
$URLSearchString = "^\[\]";

View file

@ -17,500 +17,137 @@
* @endverbatim
*/
if(!function_exists('deletenode')) {
function deletenode(&$doc, $node)
{
$xpath = new DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child)
$child->parentNode->removeChild($child);
}
}
use \Friendica\ParseUrl;
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($complete);
}
function parseurl_getsiteinfo_cached($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();
// 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, '"');
$url = original_url($url);
$siteinfo["url"] = $url;
$siteinfo["type"] = "link";
$check_cert = get_config('system','verifyssl');
$stamp1 = microtime(true);
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, 1);
curl_setopt($ch, CURLOPT_NOBODY, 1);
curl_setopt($ch, CURLOPT_TIMEOUT, 3);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
$header = curl_exec($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 (!in_array($oembed_data->type, array("error", "rich"))) {
$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, $a->get_useragent());
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
$header = curl_exec($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))
$charset = trim(trim(trim(array_pop($matches)), ';,'));
if ($charset == "")
$charset = "utf-8";
$pos = strpos($header, "\r\n\r\n");
if ($pos)
$body = trim(substr($header, $pos));
else
$body = $header;
if (($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");
$doc = new DOMDocument();
@$doc->loadHTML($body);
deletenode($doc, 'style');
deletenode($doc, 'script');
deletenode($doc, 'option');
deletenode($doc, 'h1');
deletenode($doc, 'h2');
deletenode($doc, 'h3');
deletenode($doc, 'h4');
deletenode($doc, 'h5');
deletenode($doc, 'h6');
deletenode($doc, 'ol');
deletenode($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 != "") {
$siteinfo = parseurl_getsiteinfo($content, $no_guessing, $do_oembed, ++$count);
return($siteinfo);
}
}
}
$list = $xpath->query("//title");
if ($list->length > 0)
$siteinfo["title"] = $list->item(0)->nodeValue;
//$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"] = $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]");
$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"] = $attr["content"];
break;
case "og:description":
$siteinfo["text"] = $attr["content"];
break;
}
}
if ((@$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);
$photodata = get_photo_info($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]);
}
}
} elseif ($siteinfo["image"] != "") {
$src = completeurl($siteinfo["image"], $url);
unset($siteinfo["image"]);
$photodata = get_photo_info($src);
if (($photodata) && ($photodata[0] > 10) and ($photodata[1] > 10))
$siteinfo["images"][] = array("src"=>$src,
"width"=>$photodata[0],
"height"=>$photodata[1]);
}
if ((@$siteinfo["text"] == "") AND (@$siteinfo["title"] != "") AND !$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);
}
function arr_add_hashes(&$item,$k) {
$item = '#' . $item;
}
require_once("include/items.php");
function parse_url_content(&$a) {
require_once("include/items.php");
$text = null;
$str_tags = '';
$str_tags = "";
$textmode = false;
if(local_user() && (! feature_enabled(local_user(),'richtext')))
if (local_user() && (!feature_enabled(local_user(), "richtext"))) {
$textmode = true;
}
//if($textmode)
$br = (($textmode) ? "\n" : '<br />');
$br = (($textmode) ? "\n" : "<br />");
if(x($_GET,'binurl'))
$url = trim(hex2bin($_GET['binurl']));
else
$url = trim($_GET['url']);
if (x($_GET,"binurl")) {
$url = trim(hex2bin($_GET["binurl"]));
} else {
$url = trim($_GET["url"]);
}
if($_GET['title'])
$title = strip_tags(trim($_GET['title']));
if ($_GET["title"]) {
$title = strip_tags(trim($_GET["title"]));
}
if($_GET['description'])
$text = strip_tags(trim($_GET['description']));
if ($_GET["description"]) {
$text = strip_tags(trim($_GET["description"]));
}
if($_GET['tags']) {
$arr_tags = str_getcsv($_GET['tags']);
if(count($arr_tags)) {
array_walk($arr_tags,'arr_add_hashes');
$str_tags = $br . implode(' ',$arr_tags) . $br;
if ($_GET["tags"]) {
$arr_tags = ParseUrl::convertTagsToArray($_GET["tags"]);
if (count($arr_tags)) {
$str_tags = $br . implode(" ", $arr_tags) . $br;
}
}
// add url scheme if missing
// Add url scheme if it is missing
$arrurl = parse_url($url);
if (!x($arrurl, 'scheme')) {
if (x($arrurl, 'host'))
if (!x($arrurl, "scheme")) {
if (x($arrurl, "host")) {
$url = "http:".$url;
else
} else {
$url = "http://".$url;
}
}
logger('parse_url: ' . $url);
logger("prse_url: " . $url);
if($textmode)
$template = '[bookmark=%s]%s[/bookmark]%s';
else
// If the URL is a image, video or audio file format the URL with the corresponding
// BBCode media tag
$redirects = 0;
// Fetch the header of the URL
$result = z_fetch_url($url, false, $redirects, array("novalidate" => true, "nobody" => true));
if($result["success"]) {
// Convert the header fields into an array
$hdrs = array();
$h = explode("\n", $result["header"]);
foreach ($h as $l) {
list($k,$v) = array_map("trim", explode(":", trim($l), 2));
$hdrs[$k] = $v;
}
if (array_key_exists("Content-Type", $hdrs)) {
$type = $hdrs["Content-Type"];
}
if ($type) {
if(stripos($type, "image/") !== false) {
echo $br . "[img]" . $url . "[/img]" . $br;
killme();
}
if (stripos($type, "video/") !== false) {
echo $br . "[video]" . $url . "[/video]" . $br;
killme();
}
if (stripos($type, "audio/") !== false) {
echo $br . "[audio]" . $url . "[/audio]" . $br;
killme();
}
}
}
if ($textmode) {
$template = "[bookmark=%s]%s[/bookmark]%s";
} else {
$template = "<a class=\"bookmark\" href=\"%s\" >%s</a>%s";
}
$arr = array('url' => $url, 'text' => '');
$arr = array("url" => $url, "text" => "");
call_hooks('parse_link', $arr);
call_hooks("parse_link", $arr);
if(strlen($arr['text'])) {
echo $arr['text'];
if (strlen($arr["text"])) {
echo $arr["text"];
killme();
}
if($url && $title && $text) {
if ($url && $title && $text) {
$title = str_replace(array("\r","\n"),array('',''),$title);
$title = str_replace(array("\r","\n"),array("",""),$title);
if($textmode)
$text = '[quote]' . trim($text) . '[/quote]' . $br;
else {
$text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
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;
$result = sprintf($template, $url, ($title) ? $title : $url, $text) . $str_tags;
logger('parse_url (unparsed): returns: ' . $result);
logger("parse_url (unparsed): returns: " . $result);
echo $result;
killme();
}
$siteinfo = parseurl_getsiteinfo($url);
// Fetch the information from the webpage
$siteinfo = ParseUrl::getSiteinfo($url);
unset($siteinfo["keywords"]);
// Format it as BBCode attachment
$info = add_page_info_data($siteinfo);
if (!$textmode)
if (!$textmode) {
// Replace ' with - not perfect - but the richtext editor has problems otherwise
$info = str_replace(array("&#039;"), array("&#8217;"), $info);
}
echo $info;
killme();
}
?>