nupplaPhil/friendica
1
0
Fork 0
Code Issues Pull requests Releases Wiki Activity

Merge pull request #2973 from rabuzarus/berlin_parse_url

Browse source 
parse_url: recognize image/video/audio files + move functions into own class
This commit is contained in:
Michael Vogel 2016-11-28 21:05:31 +01:00 committed by GitHub
parent 4ff49807ca 743378129b
commit 37e8f6b442
7 changed files with 844 additions and 581 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

552
include/ParseUrl.php Normal file
View file

@ -0,0 +1,552 @@
<?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");
require_once("include/xml.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 type $url The url of the page which should be scraped
* @param type $no_guessing If true the parse doens't search for
* preview pictures
* @param type $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, $do_oembed);
q("INSERT INTO `parsed_url` (`url`, `guessing`, `oembed`, `content`, `created`) VALUES ('%s', %d, %d, '%s', '%s')
ON DUPLICATE KEY UPDATE `content` = '%s', `created` = '%s'",
dbesc(normalise_link($url)), intval(!$no_guessing), intval($do_oembed),
dbesc(serialize($data)), dbesc(datetime_convert()),
dbesc(serialize($data)), dbesc(datetime_convert()));
return $data;
}
/**
* @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 type $url The url of the page which should be scraped
* @param type $no_guessing If true the parse doens't search for
* preview pictures
* @param type $do_oembed The false option is used by the function fetch_oembed()
* to avoid endless loops
* @param type $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, '"');
$url = original_url($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($body);
\xml::deleteNode($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"] = $list->item(0)->nodeValue;
}
//$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;
}
/**
* @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);
}
}

40
include/html2bbcode.php
View file

@ -1,11 +1,14 @@
<?php <?php
/* /**
html2bbcode.php * @file include/html2bbcode.php
Converter for HTML to BBCode * @brief Converter for HTML to BBCode
Made by: ike@piratenpartei.de *
Originally made for the syncom project: http://wiki.piratenpartei.de/Syncom * Made by: ike@piratenpartei.de
https://github.com/annando/Syncom * Originally made for the syncom project: http://wiki.piratenpartei.de/Syncom
*/ * https://github.com/annando/Syncom
*/
require_once("include/xml.php");
function node2bbcode(&$doc, $oldnode, $attributes, $startbb, $endbb) function node2bbcode(&$doc, $oldnode, $attributes, $startbb, $endbb)
{ {
@ -76,15 +79,6 @@ function node2bbcodesub(&$doc, $oldnode, $attributes, $startbb, $endbb)
return($replace); return($replace);
} }
if(!function_exists('deletenode')) {
function deletenode(&$doc, $node)
{
$xpath = new DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child)
$child->parentNode->removeChild($child);
}}
function _replace_code_cb($m){ function _replace_code_cb($m){
return "<code>".str_replace("\n","<br>\n",$m[1]). "</code>"; return "<code>".str_replace("\n","<br>\n",$m[1]). "</code>";
} }
@ -117,12 +111,12 @@ function html2bbcode($message)
@$doc->loadHTML($message); @$doc->loadHTML($message);
deletenode($doc, 'style'); xml::deleteNode($doc, 'style');
deletenode($doc, 'head'); xml::deleteNode($doc, 'head');
deletenode($doc, 'title'); xml::deleteNode($doc, 'title');
deletenode($doc, 'meta'); xml::deleteNode($doc, 'meta');
deletenode($doc, 'xml'); xml::deleteNode($doc, 'xml');
deletenode($doc, 'removeme'); xml::deleteNode($doc, 'removeme');
$xpath = new DomXPath($doc); $xpath = new DomXPath($doc);
$list = $xpath->query("//pre"); $list = $xpath->query("//pre");
@ -239,7 +233,7 @@ function html2bbcode($message)
node2bbcode($doc, 'iframe', array('src'=>'/(.+)/'), '[iframe]$1', '[/iframe]'); node2bbcode($doc, 'iframe', array('src'=>'/(.+)/'), '[iframe]$1', '[/iframe]');
node2bbcode($doc, 'code', array(), '[code]', '[/code]'); node2bbcode($doc, 'code', array(), '[code]', '[/code]');
node2bbcode($doc, 'key', array(), '[code]', '[/code]'); node2bbcode($doc, 'key', array(), '[code]', '[/code]');
$message = $doc->saveHTML(); $message = $doc->saveHTML();

9
include/items.php
View file

@ -1,5 +1,11 @@
<?php <?php
/**
* @file include/items.php
*/
use \Friendica\ParseUrl;
require_once('include/bbcode.php'); require_once('include/bbcode.php');
require_once('include/oembed.php'); require_once('include/oembed.php');
require_once('include/salmon.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, $keyword_blacklist = "") { function query_page_info($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 ($photo != "") if ($photo != "")
$data["images"][0]["src"] = $photo; $data["images"][0]["src"] = $photo;

75
include/oembed.php
View file

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

22
include/plaintext.php
View file

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

155
include/xml.php
View file

@ -1,11 +1,12 @@
<?php <?php
/** /**
* @file include/xml.php * @file include/xml.php
*/ */
/** /**
* @brief This class contain functions to work with XML data * @brief This class contain methods to work with XML data
* *
*/ */
class xml { class xml {
@ -23,15 +24,17 @@ class xml {
public static function from_array($array, &$xml, $remove_header = false, $namespaces = array(), $root = true) { public static function from_array($array, &$xml, $remove_header = false, $namespaces = array(), $root = true) {
if ($root) { if ($root) {
foreach($array as $key => $value) { foreach ($array as $key => $value) {
foreach ($namespaces AS $nskey => $nsvalue) foreach ($namespaces AS $nskey => $nsvalue) {
$key .= " xmlns".($nskey == "" ? "":":").$nskey.'="'.$nsvalue.'"'; $key .= " xmlns".($nskey == "" ? "":":").$nskey.'="'.$nsvalue.'"';
}
if (is_array($value)) { if (is_array($value)) {
$root = new SimpleXMLElement("<".$key."/>"); $root = new SimpleXMLElement("<".$key."/>");
self::from_array($value, $root, $remove_header, $namespaces, false); self::from_array($value, $root, $remove_header, $namespaces, false);
} else } else {
$root = new SimpleXMLElement("<".$key.">".xmlify($value)."</".$key.">"); $root = new SimpleXMLElement("<".$key.">".xmlify($value)."</".$key.">");
}
$dom = dom_import_simplexml($root)->ownerDocument; $dom = dom_import_simplexml($root)->ownerDocument;
$dom->formatOutput = true; $dom->formatOutput = true;
@ -39,16 +42,18 @@ class xml {
$xml_text = $dom->saveXML(); $xml_text = $dom->saveXML();
if ($remove_header) if ($remove_header) {
$xml_text = trim(substr($xml_text, 21)); $xml_text = trim(substr($xml_text, 21));
}
return $xml_text; return $xml_text;
} }
} }
foreach($array as $key => $value) { foreach($array as $key => $value) {
if (!isset($element) AND isset($xml)) if (!isset($element) AND isset($xml)) {
$element = $xml; $element = $xml;
}
if (is_integer($key)) { if (is_integer($key)) {
if (isset($element)) { if (isset($element)) {
@ -62,27 +67,31 @@ class xml {
} }
$element_parts = explode(":", $key); $element_parts = explode(":", $key);
if ((count($element_parts) > 1) AND isset($namespaces[$element_parts[0]])) if ((count($element_parts) > 1) AND isset($namespaces[$element_parts[0]])) {
$namespace = $namespaces[$element_parts[0]]; $namespace = $namespaces[$element_parts[0]];
elseif (isset($namespaces[""])) { } elseif (isset($namespaces[""])) {
$namespace = $namespaces[""]; $namespace = $namespaces[""];
} else } else {
$namespace = NULL; $namespace = NULL;
}
// Remove undefined namespaces from the key // Remove undefined namespaces from the key
if ((count($element_parts) > 1) AND is_null($namespace)) if ((count($element_parts) > 1) AND is_null($namespace)) {
$key = $element_parts[1]; $key = $element_parts[1];
}
if (substr($key, 0, 11) == "@attributes") { if (substr($key, 0, 11) == "@attributes") {
if (!isset($element) OR !is_array($value)) if (!isset($element) OR !is_array($value)) {
continue; continue;
}
foreach ($value as $attr_key => $attr_value) { foreach ($value as $attr_key => $attr_value) {
$element_parts = explode(":", $attr_key); $element_parts = explode(":", $attr_key);
if ((count($element_parts) > 1) AND isset($namespaces[$element_parts[0]])) if ((count($element_parts) > 1) AND isset($namespaces[$element_parts[0]])) {
$namespace = $namespaces[$element_parts[0]]; $namespace = $namespaces[$element_parts[0]];
else } else {
$namespace = NULL; $namespace = NULL;
}
$element->addAttribute($attr_key, $attr_value, $namespace); $element->addAttribute($attr_key, $attr_value, $namespace);
} }
@ -90,9 +99,9 @@ class xml {
continue; continue;
} }
if (!is_array($value)) if (!is_array($value)) {
$element = $xml->addChild($key, xmlify($value), $namespace); $element = $xml->addChild($key, xmlify($value), $namespace);
elseif (is_array($value)) { } elseif (is_array($value)) {
$element = $xml->addChild($key, NULL, $namespace); $element = $xml->addChild($key, NULL, $namespace);
self::from_array($value, $element, $remove_header, $namespaces, false); self::from_array($value, $element, $remove_header, $namespaces, false);
} }
@ -111,8 +120,9 @@ class xml {
$target->addChild($elementname, xmlify($source)); $target->addChild($elementname, xmlify($source));
else { else {
$child = $target->addChild($elementname); $child = $target->addChild($elementname);
foreach ($source->children() AS $childfield => $childentry) foreach ($source->children() AS $childfield => $childentry) {
self::copy($childentry, $child, $childfield); self::copy($childentry, $child, $childfield);
}
} }
} }
@ -168,11 +178,11 @@ class xml {
return(null); return(null);
} }
if (!is_string($xml_element) && if (!is_string($xml_element)
!is_array($xml_element) && && !is_array($xml_element)
(get_class($xml_element) == 'SimpleXMLElement')) { && (get_class($xml_element) == 'SimpleXMLElement')) {
$xml_element_copy = $xml_element; $xml_element_copy = $xml_element;
$xml_element = get_object_vars($xml_element); $xml_element = get_object_vars($xml_element);
} }
if (is_array($xml_element)) { if (is_array($xml_element)) {
@ -181,7 +191,7 @@ class xml {
return (trim(strval($xml_element_copy))); return (trim(strval($xml_element_copy)));
} }
foreach($xml_element as $key=>$value) { foreach ($xml_element as $key => $value) {
$recursion_depth++; $recursion_depth++;
$result_array[strtolower($key)] = $result_array[strtolower($key)] =
@ -223,10 +233,12 @@ class xml {
* *
* @return array The parsed XML in an array form. Use print_r() to see the resulting array structure. * @return array The parsed XML in an array form. Use print_r() to see the resulting array structure.
*/ */
public static function to_array($contents, $namespaces = true, $get_attributes=1, $priority = 'attribute') { public static function to_array($contents, $namespaces = true, $get_attributes = 1, $priority = 'attribute') {
if(!$contents) return array(); if (!$contents) {
return array();
}
if(!function_exists('xml_parser_create')) { if (!function_exists('xml_parser_create')) {
logger('xml::to_array: parser function missing'); logger('xml::to_array: parser function missing');
return array(); return array();
} }
@ -235,12 +247,13 @@ class xml {
libxml_use_internal_errors(true); libxml_use_internal_errors(true);
libxml_clear_errors(); libxml_clear_errors();
if($namespaces) if ($namespaces) {
$parser = @xml_parser_create_ns("UTF-8",':'); $parser = @xml_parser_create_ns("UTF-8",':');
else } else {
$parser = @xml_parser_create(); $parser = @xml_parser_create();
}
if(! $parser) { if (! $parser) {
logger('xml::to_array: xml_parser_create: no resource'); logger('xml::to_array: xml_parser_create: no resource');
return array(); return array();
} }
@ -252,10 +265,11 @@ class xml {
@xml_parse_into_struct($parser, trim($contents), $xml_values); @xml_parse_into_struct($parser, trim($contents), $xml_values);
@xml_parser_free($parser); @xml_parser_free($parser);
if(! $xml_values) { if (! $xml_values) {
logger('xml::to_array: libxml: parse error: ' . $contents, LOGGER_DATA); logger('xml::to_array: libxml: parse error: ' . $contents, LOGGER_DATA);
foreach(libxml_get_errors() as $err) foreach (libxml_get_errors() as $err) {
logger('libxml: parse: ' . $err->code . " at " . $err->line . ":" . $err->column . " : " . $err->message, LOGGER_DATA); logger('libxml: parse: ' . $err->code . " at " . $err->line . ":" . $err->column . " : " . $err->message, LOGGER_DATA);
}
libxml_clear_errors(); libxml_clear_errors();
return; return;
} }
@ -270,8 +284,8 @@ class xml {
// Go through the tags. // Go through the tags.
$repeated_tag_index = array(); // Multiple tags with same name will be turned into an array $repeated_tag_index = array(); // Multiple tags with same name will be turned into an array
foreach($xml_values as $data) { foreach ($xml_values as $data) {
unset($attributes,$value); // Remove existing values, or there will be trouble unset($attributes, $value); // Remove existing values, or there will be trouble
// This command will extract these variables into the foreach scope // This command will extract these variables into the foreach scope
// tag(string), type(string), level(int), attributes(array). // tag(string), type(string), level(int), attributes(array).
@ -280,46 +294,54 @@ class xml {
$result = array(); $result = array();
$attributes_data = array(); $attributes_data = array();
if(isset($value)) { if (isset($value)) {
if($priority == 'tag') $result = $value; if ($priority == 'tag') {
else $result['value'] = $value; // Put the value in a assoc array if we are in the 'Attribute' mode $result = $value;
} else {
$result['value'] = $value; // Put the value in a assoc array if we are in the 'Attribute' mode
}
} }
//Set the attributes too. //Set the attributes too.
if(isset($attributes) and $get_attributes) { if (isset($attributes) and $get_attributes) {
foreach($attributes as $attr => $val) { foreach ($attributes as $attr => $val) {
if($priority == 'tag') $attributes_data[$attr] = $val; if($priority == 'tag') {
else $result['@attributes'][$attr] = $val; // Set all the attributes in a array called 'attr' $attributes_data[$attr] = $val;
} else {
$result['@attributes'][$attr] = $val; // Set all the attributes in a array called 'attr'
}
} }
} }
// See tag status and do the needed. // See tag status and do the needed.
if($namespaces && strpos($tag,':')) { if ($namespaces && strpos($tag, ':')) {
$namespc = substr($tag,0,strrpos($tag,':')); $namespc = substr($tag, 0, strrpos($tag, ':'));
$tag = strtolower(substr($tag,strlen($namespc)+1)); $tag = strtolower(substr($tag, strlen($namespc)+1));
$result['@namespace'] = $namespc; $result['@namespace'] = $namespc;
} }
$tag = strtolower($tag); $tag = strtolower($tag);
if($type == "open") { // The starting of the tag '<tag>' if ($type == "open") { // The starting of the tag '<tag>'
$parent[$level-1] = &$current; $parent[$level-1] = &$current;
if(!is_array($current) or (!in_array($tag, array_keys($current)))) { // Insert New tag if (!is_array($current) or (!in_array($tag, array_keys($current)))) { // Insert New tag
$current[$tag] = $result; $current[$tag] = $result;
if($attributes_data) $current[$tag. '_attr'] = $attributes_data; if ($attributes_data) {
$current[$tag. '_attr'] = $attributes_data;
}
$repeated_tag_index[$tag.'_'.$level] = 1; $repeated_tag_index[$tag.'_'.$level] = 1;
$current = &$current[$tag]; $current = &$current[$tag];
} else { // There was another element with the same tag name } else { // There was another element with the same tag name
if(isset($current[$tag][0])) { // If there is a 0th element it is already an array if (isset($current[$tag][0])) { // If there is a 0th element it is already an array
$current[$tag][$repeated_tag_index[$tag.'_'.$level]] = $result; $current[$tag][$repeated_tag_index[$tag.'_'.$level]] = $result;
$repeated_tag_index[$tag.'_'.$level]++; $repeated_tag_index[$tag.'_'.$level]++;
} else { // This section will make the value an array if multiple tags with the same name appear together } else { // This section will make the value an array if multiple tags with the same name appear together
$current[$tag] = array($current[$tag],$result); // This will combine the existing item and the new item together to make an array $current[$tag] = array($current[$tag], $result); // This will combine the existing item and the new item together to make an array
$repeated_tag_index[$tag.'_'.$level] = 2; $repeated_tag_index[$tag.'_'.$level] = 2;
if(isset($current[$tag.'_attr'])) { // The attribute of the last(0th) tag must be moved as well if (isset($current[$tag.'_attr'])) { // The attribute of the last(0th) tag must be moved as well
$current[$tag]['0_attr'] = $current[$tag.'_attr']; $current[$tag]['0_attr'] = $current[$tag.'_attr'];
unset($current[$tag.'_attr']); unset($current[$tag.'_attr']);
} }
@ -329,35 +351,37 @@ class xml {
$current = &$current[$tag][$last_item_index]; $current = &$current[$tag][$last_item_index];
} }
} elseif($type == "complete") { // Tags that ends in 1 line '<tag />' } elseif ($type == "complete") { // Tags that ends in 1 line '<tag />'
//See if the key is already taken. //See if the key is already taken.
if(!isset($current[$tag])) { //New Key if (!isset($current[$tag])) { //New Key
$current[$tag] = $result; $current[$tag] = $result;
$repeated_tag_index[$tag.'_'.$level] = 1; $repeated_tag_index[$tag.'_'.$level] = 1;
if($priority == 'tag' and $attributes_data) $current[$tag. '_attr'] = $attributes_data; if ($priority == 'tag' and $attributes_data) {
$current[$tag. '_attr'] = $attributes_data;
}
} else { // If taken, put all things inside a list(array) } else { // If taken, put all things inside a list(array)
if(isset($current[$tag][0]) and is_array($current[$tag])) { // If it is already an array... if (isset($current[$tag][0]) and is_array($current[$tag])) { // If it is already an array...
// ...push the new element into that array. // ...push the new element into that array.
$current[$tag][$repeated_tag_index[$tag.'_'.$level]] = $result; $current[$tag][$repeated_tag_index[$tag.'_'.$level]] = $result;
if($priority == 'tag' and $get_attributes and $attributes_data) { if ($priority == 'tag' and $get_attributes and $attributes_data) {
$current[$tag][$repeated_tag_index[$tag.'_'.$level] . '_attr'] = $attributes_data; $current[$tag][$repeated_tag_index[$tag.'_'.$level] . '_attr'] = $attributes_data;
} }
$repeated_tag_index[$tag.'_'.$level]++; $repeated_tag_index[$tag.'_'.$level]++;
} else { // If it is not an array... } else { // If it is not an array...
$current[$tag] = array($current[$tag],$result); //...Make it an array using using the existing value and the new value $current[$tag] = array($current[$tag], $result); //...Make it an array using using the existing value and the new value
$repeated_tag_index[$tag.'_'.$level] = 1; $repeated_tag_index[$tag.'_'.$level] = 1;
if($priority == 'tag' and $get_attributes) { if ($priority == 'tag' and $get_attributes) {
if(isset($current[$tag.'_attr'])) { // The attribute of the last(0th) tag must be moved as well if (isset($current[$tag.'_attr'])) { // The attribute of the last(0th) tag must be moved as well
$current[$tag]['0_attr'] = $current[$tag.'_attr']; $current[$tag]['0_attr'] = $current[$tag.'_attr'];
unset($current[$tag.'_attr']); unset($current[$tag.'_attr']);
} }
if($attributes_data) { if ($attributes_data) {
$current[$tag][$repeated_tag_index[$tag.'_'.$level] . '_attr'] = $attributes_data; $current[$tag][$repeated_tag_index[$tag.'_'.$level] . '_attr'] = $attributes_data;
} }
} }
@ -365,12 +389,25 @@ class xml {
} }
} }
} elseif($type == 'close') { // End of tag '</tag>' } elseif ($type == 'close') { // End of tag '</tag>'
$current = &$parent[$level-1]; $current = &$parent[$level-1];
} }
} }
return($xml_array); return($xml_array);
} }
/**
* @brief Delete a node in a XML object
*
* @param object $doc XML document
* @param string $node Node name
*/
public static function deleteNode(&$doc, $node) {
$xpath = new DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child) {
$child->parentNode->removeChild($child);
}
}
} }
?>

572
mod/parse_url.php
View file

@ -1,516 +1,176 @@
<?php <?php
/** /**
* @file mod/parse_url.php * @file mod/parse_url.php
* @brief The parse_url module
* *
* @todo https://developers.google.com/+/plugins/snippet/ * This module does parse an url for embedable content (audio, video, image files or link)
* information and does format this information to BBCode or html (this depends
* on the user settings - default is BBCode output).
* If the user has enabled the richtext editor setting the output will be in html
* (Note: This is not always possible and in some case not useful because
* the richtext editor doesn't support all kind of html).
* Otherwise the output will be constructed BBCode.
* *
* @verbatim * @see ParseUrl::getSiteinfo() for more information about scraping embeddable content
* <meta itemprop="name" content="Toller Titel">
* <meta itemprop="description" content="Eine tolle Beschreibung">
* <meta itemprop="image" content="http://maple.libertreeproject.org/images/tree-icon.png">
*
* <body itemscope itemtype="http://schema.org/Product">
* <h1 itemprop="name">Shiny Trinket</h1>
* <img itemprop="image" src="{image-url}" />
* <p itemprop="description">Shiny trinkets are shiny.</p>
* </body>
* @endverbatim
*/ */
if(!function_exists('deletenode')) { use \Friendica\ParseUrl;
function deletenode(&$doc, $node)
{
$xpath = new DomXPath($doc);
$list = $xpath->query("//".$node);
foreach ($list as $child)
$child->parentNode->removeChild($child);
}
}
function completeurl($url, $scheme) { require_once("include/items.php");
$urlarr = parse_url($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 = '#' . $item;
}
function parse_url_content(&$a) { function parse_url_content(&$a) {
require_once("include/items.php");
$text = null; $text = null;
$str_tags = ''; $str_tags = "";
$textmode = false; $textmode = false;
if(local_user() && (! feature_enabled(local_user(),'richtext'))) if (local_user() && (!feature_enabled(local_user(), "richtext"))) {
$textmode = true; $textmode = true;
}
//if($textmode) $br = (($textmode) ? "\n" : "<br />");
$br = (($textmode) ? "\n" : '<br />');
if(x($_GET,'binurl')) if (x($_GET,"binurl")) {
$url = trim(hex2bin($_GET['binurl'])); $url = trim(hex2bin($_GET["binurl"]));
else } else {
$url = trim($_GET['url']); $url = trim($_GET["url"]);
}
if($_GET['title']) if ($_GET["title"]) {
$title = strip_tags(trim($_GET['title'])); $title = strip_tags(trim($_GET["title"]));
}
if($_GET['description']) if ($_GET["description"]) {
$text = strip_tags(trim($_GET['description'])); $text = strip_tags(trim($_GET["description"]));
}
if($_GET['tags']) { if ($_GET["tags"]) {
$arr_tags = str_getcsv($_GET['tags']); $arr_tags = ParseUrl::convertTagsToArray($_GET["tags"]);
if(count($arr_tags)) { if (count($arr_tags)) {
array_walk($arr_tags,'arr_add_hashes'); $str_tags = $br . implode(" ", $arr_tags) . $br;
$str_tags = $br . implode(' ',$arr_tags) . $br;
} }
} }
// add url scheme if missing // Add url scheme if it is missing
$arrurl = parse_url($url); $arrurl = parse_url($url);
if (!x($arrurl, 'scheme')) { if (!x($arrurl, "scheme")) {
if (x($arrurl, 'host')) if (x($arrurl, "host")) {
$url = "http:".$url; $url = "http:".$url;
else } else {
$url = "http://".$url; $url = "http://".$url;
}
} }
logger('parse_url: ' . $url); logger("prse_url: " . $url);
if($textmode) // Check if the URL is an image, video or audio file. If so format
$template = '[bookmark=%s]%s[/bookmark]%s'; // the URL with the corresponding BBCode media tag
else $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"; $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'])) { if (strlen($arr["text"])) {
echo $arr['text']; echo $arr["text"];
killme(); killme();
} }
// If there is allready some content information submitted we don't
// need to parse the url for content.
if ($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]" . $br;
if($textmode) } else {
$text = '[quote]' . trim($text) . '[/quote]' . $br; $text = "<blockquote>" . htmlspecialchars(trim($text)) . "</blockquote><br />";
else {
$text = '<blockquote>' . htmlspecialchars(trim($text)) . '</blockquote><br />';
$title = htmlspecialchars($title); $title = htmlspecialchars($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; echo $result;
killme(); killme();
} }
$siteinfo = parseurl_getsiteinfo($url); // Fetch the information directly from the webpage
$siteinfo = ParseUrl::getSiteinfo($url);
unset($siteinfo["keywords"]); unset($siteinfo["keywords"]);
// Format it as BBCode attachment
$info = add_page_info_data($siteinfo); $info = add_page_info_data($siteinfo);
if (!$textmode) if (!$textmode) {
// Replace ' with ’ - not perfect - but the richtext editor has problems otherwise // Replace ' with ’ - not perfect - but the richtext editor has problems otherwise
$info = str_replace(array("&#039;"), array("&#8217;"), $info); $info = str_replace(array("&#039;"), array("&#8217;"), $info);
}
echo $info; echo $info;
killme(); killme();
} }
?>
/**
* @brief Legacy function to call ParseUrl::getSiteinfoCached
*
* Note: We have moved the function to ParseUrl.php. This function is only for
* legacy support and will be remove in the future
*
* @param type $url The url of the page which should be scraped
* @param type $no_guessing If true the parse doens't search for
* preview pictures
* @param type $do_oembed The false option is used by the function fetch_oembed()
* to avoid endless loops
*
* @return array which contains needed data for embedding
*
* @see ParseUrl::getSiteinfoCached()
*
* @todo Remove this function after all Addons has been changed to use
* ParseUrl::getSiteinfoCached
*/
function parseurl_getsiteinfo_cached($url, $no_guessing = false, $do_oembed = true) {
$siteinfo = ParseUrl::getSiteinfoCached($url, $no_guessing, $do_oembed);
return $siteinfo;
}