friendica_2021-01/mod/parse_url.php

152 lines
3.7 KiB
PHP
Raw Normal View History

2010-07-23 07:41:45 +02:00
<?php
2016-11-27 23:41:55 +01:00
/**
* @file mod/parse_url.php
2016-11-25 16:59:31 +01:00
* @brief The parse_url module
*
* This module does parse an url for embeddable content (audio, video, image files or link)
* information and does format this information to BBCode
*
* @see ParseUrl::getSiteinfo() for more information about scraping embeddable content
*/
use \Friendica\ParseUrl;
require_once("include/items.php");
function parse_url_content(App $a) {
$text = null;
$str_tags = "";
$br = "\n";
if (x($_GET,"binurl")) {
$url = trim(hex2bin($_GET["binurl"]));
} else {
$url = trim($_GET["url"]);
}
if ($_GET["title"]) {
$title = strip_tags(trim($_GET["title"]));
}
if ($_GET["description"]) {
$text = strip_tags(trim($_GET["description"]));
}
if ($_GET["tags"]) {
$arr_tags = ParseUrl::convertTagsToArray($_GET["tags"]);
if (count($arr_tags)) {
$str_tags = $br . implode(" ", $arr_tags) . $br;
}
}
// Add url scheme if it is missing
$arrurl = parse_url($url);
if (!x($arrurl, "scheme")) {
if (x($arrurl, "host")) {
$url = "http:".$url;
} else {
$url = "http://".$url;
2015-03-07 23:14:26 +01:00
}
}
logger("prse_url: " . $url);
2016-11-25 16:59:31 +01:00
// Check if the URL is an image, video or audio file. If so 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();
}
2015-03-07 23:14:26 +01:00
}
}
$template = "[bookmark=%s]%s[/bookmark]%s";
2010-07-23 07:41:45 +02:00
$arr = array("url" => $url, "text" => "");
2010-12-26 00:01:02 +01:00
call_hooks("parse_link", $arr);
2010-12-26 00:01:02 +01:00
if (strlen($arr["text"])) {
echo $arr["text"];
2010-12-26 00:01:02 +01:00
killme();
}
2016-11-25 16:59:31 +01:00
// If there is allready some content information submitted we don't
// need to parse the url for content.
if ($url && $title && $text) {
$title = str_replace(array("\r","\n"),array("",""),$title);
$text = "[quote]" . trim($text) . "[/quote]" . $br;
$result = sprintf($template, $url, ($title) ? $title : $url, $text) . $str_tags;
logger("parse_url (unparsed): returns: " . $result);
echo $result;
killme();
}
2016-11-25 16:59:31 +01:00
// Fetch the information directly from the webpage
$siteinfo = ParseUrl::getSiteinfo($url);
unset($siteinfo["keywords"]);
// Format it as BBCode attachment
2016-04-18 20:57:01 +02:00
$info = add_page_info_data($siteinfo);
echo $info;
2010-07-23 07:41:45 +02:00
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;
}