friendica_2021-01/include/network.php

1081 lines
28 KiB
PHP
Raw Normal View History

<?php
/**
* @file include/network.php
*/
use Friendica\App;
2017-08-26 08:04:21 +02:00
use Friendica\Core\System;
2017-04-30 06:01:26 +02:00
use Friendica\Core\Config;
2017-05-07 20:44:30 +02:00
use Friendica\Network\Probe;
use Friendica\Util\XML;
/**
* @brief Curl wrapper
2017-02-18 04:32:33 +01:00
*
* If binary flag is true, return binary results.
* Set the cookiejar argument to a string (e.g. "/tmp/friendica-cookies.txt")
* to preserve cookies from one request to the next.
2017-02-18 04:32:33 +01:00
*
* @param string $url URL to fetch
* @param boolean $binary default false
* TRUE if asked to return binary results (file download)
* @param integer $redirects The recursion counter for internal use - default 0
* @param integer $timeout Timeout in seconds, default system config value or 60 seconds
* @param string $accept_content supply Accept: header with 'accept_content' as the value
* @param string $cookiejar Path to cookie jar file
2017-02-18 04:32:33 +01:00
*
* @return string The fetched content
*/
function fetch_url($url, $binary = false, &$redirects = 0, $timeout = 0, $accept_content = null, $cookiejar = 0)
{
2015-05-15 20:41:10 +02:00
$ret = z_fetch_url(
$url,
$binary,
$redirects,
array('timeout'=>$timeout,
2015-05-15 20:41:10 +02:00
'accept_content'=>$accept_content,
'cookiejar'=>$cookiejar
)
);
2015-05-15 20:41:10 +02:00
return($ret['body']);
}
2015-05-15 20:41:10 +02:00
/**
* @brief fetches an URL.
*
* @param string $url URL to fetch
* @param boolean $binary default false
* TRUE if asked to return binary results (file download)
* @param int $redirects The recursion counter for internal use - default 0
* @param array $opts (optional parameters) assoziative array with:
* 'accept_content' => supply Accept: header with 'accept_content' as the value
* 'timeout' => int Timeout in seconds, default system config value or 60 seconds
* 'http_auth' => username:password
* 'novalidate' => do not validate SSL certs, default is to validate using our CA list
* 'nobody' => only return the header
* 'cookiejar' => path to cookie jar file
*
2015-05-15 20:41:10 +02:00
* @return array an assoziative array with:
* int 'return_code' => HTTP return code or 0 if timeout or failure
* boolean 'success' => boolean true (if HTTP 2xx result) or false
* string 'redirect_url' => in case of redirect, content was finally retrieved from this URL
* string 'header' => HTTP headers
* string 'body' => fetched content
2015-05-15 20:41:10 +02:00
*/
function z_fetch_url($url, $binary = false, &$redirects = 0, $opts = array())
{
$ret = array('return_code' => 0, 'success' => false, 'header' => '', 'info' => '', 'body' => '');
2015-05-15 20:41:10 +02:00
$stamp1 = microtime(true);
$a = get_app();
if (blocked_url($url)) {
logger('z_fetch_url: domain of ' . $url . ' is blocked', LOGGER_DATA);
return $ret;
}
$ch = @curl_init($url);
if (($redirects > 8) || (!$ch)) {
return $ret;
}
@curl_setopt($ch, CURLOPT_HEADER, true);
2012-07-07 20:40:15 +02:00
if (x($opts, "cookiejar")) {
2015-05-15 20:41:10 +02:00
curl_setopt($ch, CURLOPT_COOKIEJAR, $opts["cookiejar"]);
curl_setopt($ch, CURLOPT_COOKIEFILE, $opts["cookiejar"]);
}
2013-02-22 01:42:34 +01:00
// These settings aren't needed. We're following the location already.
// @curl_setopt($ch, CURLOPT_FOLLOWLOCATION, true);
// @curl_setopt($ch, CURLOPT_MAXREDIRS, 5);
2012-12-20 04:19:02 +01:00
if (x($opts, 'accept_content')) {
curl_setopt(
$ch,
CURLOPT_HTTPHEADER,
array('Accept: ' . $opts['accept_content'])
);
}
2012-07-07 20:40:15 +02:00
@curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
@curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
$range = intval(Config::get('system', 'curl_range_bytes', 0));
if ($range > 0) {
@curl_setopt($ch, CURLOPT_RANGE, '0-' . $range);
}
2015-05-15 20:41:10 +02:00
// Without this setting it seems as if some webservers send compressed content
// This seems to confuse curl so that it shows this uncompressed.
/// @todo We could possibly set this value to "gzip" or something similar
curl_setopt($ch, CURLOPT_ENCODING, '');
if (x($opts, 'headers')) {
2015-05-15 20:41:10 +02:00
@curl_setopt($ch, CURLOPT_HTTPHEADER, $opts['headers']);
}
if (x($opts, 'nobody')) {
2015-05-15 20:41:10 +02:00
@curl_setopt($ch, CURLOPT_NOBODY, $opts['nobody']);
}
if (x($opts, 'timeout')) {
@curl_setopt($ch, CURLOPT_TIMEOUT, $opts['timeout']);
} else {
$curl_time = intval(Config::get('system', 'curl_timeout'));
@curl_setopt($ch, CURLOPT_TIMEOUT, (($curl_time !== false) ? $curl_time : 60));
2011-08-17 05:05:02 +02:00
}
// by default we will allow self-signed certs
// but you can override this
$check_cert = Config::get('system', 'verifyssl');
@curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
if ($check_cert) {
@curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 2);
}
$proxy = Config::get('system', 'proxy');
if (strlen($proxy)) {
@curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1);
@curl_setopt($ch, CURLOPT_PROXY, $proxy);
$proxyuser = @Config::get('system', 'proxyuser');
if (strlen($proxyuser)) {
@curl_setopt($ch, CURLOPT_PROXYUSERPWD, $proxyuser);
}
}
if (Config::get('system', 'ipv4_resolve', false)) {
curl_setopt($ch, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4);
}
if ($binary) {
@curl_setopt($ch, CURLOPT_BINARYTRANSFER, 1);
}
$a->set_curl_code(0);
// don't let curl abort the entire application
// if it throws any errors.
$s = @curl_exec($ch);
2017-10-16 05:14:32 +02:00
$curl_info = @curl_getinfo($ch);
// Special treatment for HTTP Code 416
// See https://developer.mozilla.org/en-US/docs/Web/HTTP/Status/416
if (($curl_info['http_code'] == 416) && ($range > 0)) {
@curl_setopt($ch, CURLOPT_RANGE, '');
$s = @curl_exec($ch);
$curl_info = @curl_getinfo($ch);
}
2016-01-03 06:59:54 +01:00
if (curl_errno($ch) !== CURLE_OK) {
logger('fetch_url error fetching ' . $url . ': ' . curl_error($ch), LOGGER_NORMAL);
2016-01-03 06:59:54 +01:00
}
$ret['errno'] = curl_errno($ch);
$base = $s;
$ret['info'] = $curl_info;
2015-05-15 20:41:10 +02:00
$http_code = $curl_info['http_code'];
2017-10-16 05:14:32 +02:00
logger('fetch_url ' . $url . ': ' . $http_code . " " . $s, LOGGER_DATA);
$header = '';
// Pull out multiple headers, e.g. proxy and continuation headers
// allow for HTTP/2.x without fixing code
while (preg_match('/^HTTP\/[1-2].+? [1-5][0-9][0-9]/', $base)) {
$chunk = substr($base, 0, strpos($base, "\r\n\r\n") + 4);
$header .= $chunk;
$base = substr($base, strlen($chunk));
}
$a->set_curl_code($http_code);
$a->set_curl_content_type($curl_info['content_type']);
$a->set_curl_headers($header);
if ($http_code == 301 || $http_code == 302 || $http_code == 303 || $http_code == 307) {
$new_location_info = @parse_url($curl_info['redirect_url']);
$old_location_info = @parse_url($curl_info['url']);
$newurl = $curl_info['redirect_url'];
if (($new_location_info['path'] == '') && ( $new_location_info['host'] != '')) {
$newurl = $new_location_info['scheme'] . '://' . $new_location_info['host'] . $old_location_info['path'];
}
$matches = array();
2015-02-10 04:14:08 +01:00
if (preg_match('/(Location:|URI:)(.*?)\n/i', $header, $matches)) {
$newurl = trim(array_pop($matches));
}
if (strpos($newurl, '/') === 0) {
$newurl = $old_location_info["scheme"]."://".$old_location_info["host"].$newurl;
}
if (filter_var($newurl, FILTER_VALIDATE_URL)) {
2012-07-07 20:40:15 +02:00
$redirects++;
2015-05-15 20:41:10 +02:00
@curl_close($ch);
return z_fetch_url($newurl, $binary, $redirects, $opts);
2012-07-07 20:40:15 +02:00
}
}
$a->set_curl_code($http_code);
$a->set_curl_content_type($curl_info['content_type']);
$body = substr($s, strlen($header));
2015-05-15 20:41:10 +02:00
$rc = intval($http_code);
$ret['return_code'] = $rc;
$ret['success'] = (($rc >= 200 && $rc <= 299) ? true : false);
$ret['redirect_url'] = $url;
if (!$ret['success']) {
2015-05-15 20:41:10 +02:00
$ret['error'] = curl_error($ch);
$ret['debug'] = $curl_info;
2017-08-11 10:04:01 +02:00
logger('z_fetch_url: error: '.$url.': '.$ret['return_code'].' - '.$ret['error'], LOGGER_DEBUG);
logger('z_fetch_url: debug: '.print_r($curl_info, true), LOGGER_DATA);
2015-05-15 20:41:10 +02:00
}
$ret['body'] = substr($s, strlen($header));
2015-05-15 20:41:10 +02:00
$ret['header'] = $header;
if (x($opts, 'debug')) {
2015-05-15 20:41:10 +02:00
$ret['debug'] = $curl_info;
}
2015-05-15 20:41:10 +02:00
@curl_close($ch);
$a->save_timestamp($stamp1, 'network');
2015-05-15 20:41:10 +02:00
return($ret);
}
/**
* @brief Send POST request to $url
2017-02-18 04:32:33 +01:00
*
* @param string $url URL to post
* @param mixed $params array of POST variables
* @param string $headers HTTP headers
* @param integer $redirects Recursion counter for internal use - default = 0
* @param integer $timeout The timeout in seconds, default system config value or 60 seconds
2017-02-18 04:32:33 +01:00
*
* @return string The content
*/
function post_url($url, $params, $headers = null, &$redirects = 0, $timeout = 0)
{
$stamp1 = microtime(true);
if (blocked_url($url)) {
logger('post_url: domain of ' . $url . ' is blocked', LOGGER_DATA);
return false;
}
$a = get_app();
$ch = curl_init($url);
if (($redirects > 8) || (!$ch)) {
return false;
}
logger('post_url: start ' . $url, LOGGER_DATA);
2014-05-24 11:01:33 +02:00
curl_setopt($ch, CURLOPT_HEADER, true);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $params);
curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
if (Config::get('system', 'ipv4_resolve', false)) {
curl_setopt($ch, CURLOPT_IPRESOLVE, CURL_IPRESOLVE_V4);
}
if (intval($timeout)) {
2011-08-17 05:05:02 +02:00
curl_setopt($ch, CURLOPT_TIMEOUT, $timeout);
} else {
$curl_time = intval(Config::get('system', 'curl_timeout'));
2011-08-17 05:05:02 +02:00
curl_setopt($ch, CURLOPT_TIMEOUT, (($curl_time !== false) ? $curl_time : 60));
}
if (defined('LIGHTTPD')) {
if (!is_array($headers)) {
$headers = array('Expect:');
} else {
if (!in_array('Expect:', $headers)) {
array_push($headers, 'Expect:');
}
}
}
if ($headers) {
curl_setopt($ch, CURLOPT_HTTPHEADER, $headers);
}
$check_cert = Config::get('system', 'verifyssl');
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
if ($check_cert) {
@curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 2);
}
$proxy = Config::get('system', 'proxy');
if (strlen($proxy)) {
curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1);
curl_setopt($ch, CURLOPT_PROXY, $proxy);
$proxyuser = Config::get('system', 'proxyuser');
if (strlen($proxyuser)) {
curl_setopt($ch, CURLOPT_PROXYUSERPWD, $proxyuser);
}
}
$a->set_curl_code(0);
// don't let curl abort the entire application
// if it throws any errors.
$s = @curl_exec($ch);
$base = $s;
$curl_info = curl_getinfo($ch);
$http_code = $curl_info['http_code'];
logger('post_url: result ' . $http_code . ' - ' . $url, LOGGER_DATA);
2014-05-24 11:01:33 +02:00
$header = '';
// Pull out multiple headers, e.g. proxy and continuation headers
// allow for HTTP/2.x without fixing code
while (preg_match('/^HTTP\/[1-2].+? [1-5][0-9][0-9]/', $base)) {
$chunk = substr($base, 0, strpos($base, "\r\n\r\n") + 4);
$header .= $chunk;
$base = substr($base, strlen($chunk));
}
if ($http_code == 301 || $http_code == 302 || $http_code == 303 || $http_code == 307) {
2014-05-24 11:01:33 +02:00
$matches = array();
preg_match('/(Location:|URI:)(.*?)\n/', $header, $matches);
$newurl = trim(array_pop($matches));
if (strpos($newurl, '/') === 0) {
$newurl = $old_location_info["scheme"] . "://" . $old_location_info["host"] . $newurl;
}
2014-05-24 11:01:33 +02:00
if (filter_var($newurl, FILTER_VALIDATE_URL)) {
$redirects++;
logger('post_url: redirect ' . $url . ' to ' . $newurl);
return post_url($newurl, $params, $headers, $redirects, $timeout);
2014-05-24 11:01:33 +02:00
}
}
$a->set_curl_code($http_code);
$body = substr($s, strlen($header));
$a->set_curl_headers($header);
curl_close($ch);
$a->save_timestamp($stamp1, 'network');
logger('post_url: end ' . $url, LOGGER_DATA);
2014-05-24 11:01:33 +02:00
return $body;
}
// Generic XML return
// Outputs a basic dfrn XML status structure to STDOUT, with a <status> variable
// of $st and an optional text <message> of $message and terminates the current process.
function xml_status($st, $message = '')
{
$result = array('status' => $st);
if ($message != '') {
$result['message'] = $message;
}
if ($st) {
logger('xml_status returning non_zero: ' . $st . " message=" . $message);
}
header("Content-type: text/xml");
$xmldata = array("result" => $result);
echo XML::from_array($xmldata, $xml);
killme();
}
/**
* @brief Send HTTP status header and exit.
*
* @param integer $val HTTP status result value
* @param array $description optional message
* 'title' => header title
* 'description' => optional message
*/
/**
* @brief Send HTTP status header and exit.
*
* @param integer $val HTTP status result value
* @param array $description optional message
* 'title' => header title
* 'description' => optional message
*/
function http_status_exit($val, $description = array())
{
$err = '';
if ($val >= 400) {
2011-08-10 03:55:46 +02:00
$err = 'Error';
if (!isset($description["title"])) {
2015-10-03 23:16:40 +02:00
$description["title"] = $err." ".$val;
}
2015-10-03 23:16:40 +02:00
}
if ($val >= 200 && $val < 300)
2011-08-10 03:55:46 +02:00
$err = 'OK';
logger('http_status_exit ' . $val);
2011-08-10 03:55:46 +02:00
header($_SERVER["SERVER_PROTOCOL"] . ' ' . $val . ' ' . $err);
2015-10-03 23:16:40 +02:00
if (isset($description["title"])) {
$tpl = get_markup_template('http_status.tpl');
echo replace_macros(
$tpl,
array(
'$title' => $description["title"],
'$description' => $description["description"])
);
2015-10-03 23:16:40 +02:00
}
2011-08-10 03:55:46 +02:00
killme();
}
/**
* @brief Check URL to se if ts's real
2017-02-18 04:32:33 +01:00
*
* Take a URL from the wild, prepend http:// if necessary
* and check DNS to see if it's real (or check if is a valid IP address)
2017-02-18 04:32:33 +01:00
*
* @param string $url The URL to be validated
* @return boolean True if it's a valid URL, fals if something wrong with it
*/
function validate_url(&$url)
{
if (Config::get('system', 'disable_url_validation')) {
return true;
}
// no naked subdomains (allow localhost for tests)
if (strpos($url, '.') === false && strpos($url, '/localhost/') === false)
2012-02-12 23:18:32 +01:00
return false;
if (substr($url, 0, 4) != 'http')
$url = 'http://' . $url;
/// @TODO Really supress function outcomes? Why not find them + debug them?
$h = @parse_url($url);
if ((is_array($h)) && (dns_get_record($h['host'], DNS_A + DNS_CNAME + DNS_PTR) || filter_var($h['host'], FILTER_VALIDATE_IP) )) {
return true;
}
return false;
}
/**
* @brief Checks that email is an actual resolvable internet address
2017-02-18 04:32:33 +01:00
*
* @param string $addr The email address
* @return boolean True if it's a valid email address, false if it's not
*/
function validate_email($addr)
{
if (Config::get('system', 'disable_email_validation')) {
2012-07-02 03:56:00 +02:00
return true;
}
2012-07-02 03:56:00 +02:00
if (! strpos($addr, '@')) {
return false;
}
$h = substr($addr, strpos($addr, '@') + 1);
if (($h) && (dns_get_record($h, DNS_A + DNS_CNAME + DNS_PTR + DNS_MX) || filter_var($h, FILTER_VALIDATE_IP) )) {
return true;
}
return false;
}
/**
* @brief Check if URL is allowed
2017-02-18 04:32:33 +01:00
*
* Check $url against our list of allowed sites,
* wildcards allowed. If allowed_sites is unset return true;
2017-02-18 04:32:33 +01:00
*
* @param string $url URL which get tested
* @return boolean True if url is allowed otherwise return false
*/
function allowed_url($url)
{
$h = @parse_url($url);
if (! $h) {
return false;
}
2017-04-27 04:08:47 +02:00
$str_allowed = Config::get('system', 'allowed_sites');
2017-04-26 04:45:56 +02:00
if (! $str_allowed) {
return true;
2017-04-26 04:45:56 +02:00
}
$found = false;
$host = strtolower($h['host']);
// always allow our own site
2017-04-26 04:45:56 +02:00
if ($host == strtolower($_SERVER['SERVER_NAME'])) {
return true;
2017-04-26 04:45:56 +02:00
}
$fnmatch = function_exists('fnmatch');
2017-04-26 04:45:56 +02:00
$allowed = explode(',', $str_allowed);
if (count($allowed)) {
foreach ($allowed as $a) {
$pat = strtolower(trim($a));
2017-04-26 04:45:56 +02:00
if (($fnmatch && fnmatch($pat, $host)) || ($pat == $host)) {
$found = true;
break;
}
}
}
return $found;
}
2017-04-26 04:45:56 +02:00
/**
* Checks if the provided url domain is on the domain blocklist.
* Returns true if it is or malformed URL, false if not.
2017-04-26 04:45:56 +02:00
*
* @param string $url The url to check the domain from
*
2017-04-26 04:45:56 +02:00
* @return boolean
*/
function blocked_url($url)
{
2017-04-26 04:45:56 +02:00
$h = @parse_url($url);
if (! $h) {
return true;
2017-04-26 04:45:56 +02:00
}
2017-04-27 04:08:47 +02:00
$domain_blocklist = Config::get('system', 'blocklist', array());
2017-04-26 04:45:56 +02:00
if (! $domain_blocklist) {
return false;
2017-04-26 04:45:56 +02:00
}
$host = strtolower($h['host']);
foreach ($domain_blocklist as $domain_block) {
if (strtolower($domain_block['domain']) == $host) {
return true;
2017-04-26 04:45:56 +02:00
}
}
return false;
2017-04-26 04:45:56 +02:00
}
/**
* @brief Check if email address is allowed to register here.
2017-02-18 04:32:33 +01:00
*
* Compare against our list (wildcards allowed).
2017-02-18 04:32:33 +01:00
*
* @param string $email email address
* @return boolean False if not allowed, true if allowed
* or if allowed list is not configured
*/
function allowed_email($email)
{
$domain = strtolower(substr($email, strpos($email, '@') + 1));
if (! $domain) {
return false;
}
$str_allowed = Config::get('system', 'allowed_email');
if (! $str_allowed) {
return true;
}
$found = false;
$fnmatch = function_exists('fnmatch');
$allowed = explode(',', $str_allowed);
if (count($allowed)) {
foreach ($allowed as $a) {
$pat = strtolower(trim($a));
if (($fnmatch && fnmatch($pat, $domain)) || ($pat == $domain)) {
$found = true;
break;
}
}
}
return $found;
}
function avatar_img($email)
{
$avatar['size'] = 175;
$avatar['email'] = $email;
$avatar['url'] = '';
2012-04-11 00:33:43 +02:00
$avatar['success'] = false;
call_hooks('avatar_lookup', $avatar);
if (! $avatar['success']) {
$avatar['url'] = System::baseUrl() . '/images/person-175.jpg';
}
2012-04-11 00:33:43 +02:00
logger('Avatar: ' . $avatar['email'] . ' ' . $avatar['url'], LOGGER_DEBUG);
return $avatar['url'];
}
function parse_xml_string($s, $strict = true)
{
2017-09-25 23:34:44 +02:00
// the "strict" parameter is deactivated
/// @todo Move this function to the xml class
libxml_use_internal_errors(true);
2017-09-25 23:34:44 +02:00
$x = @simplexml_load_string($s);
if (!$x) {
logger('libxml: parse: error: ' . $s, LOGGER_DATA);
foreach (libxml_get_errors() as $err) {
logger('libxml: parse: ' . $err->code." at ".$err->line.":".$err->column." : ".$err->message, LOGGER_DATA);
}
libxml_clear_errors();
}
return $x;
}
2011-08-18 13:20:30 +02:00
function scale_external_images($srctext, $include_link = true, $scale_replace = false)
{
// Suppress "view full size"
if (intval(Config::get('system', 'no_view_full_size'))) {
$include_link = false;
}
2012-02-25 23:22:51 +01:00
$a = get_app();
// Picture addresses can contain special characters
2013-05-03 07:19:16 +02:00
$s = htmlspecialchars_decode($srctext);
2012-02-25 23:22:51 +01:00
$matches = null;
$c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism', $s, $matches, PREG_SET_ORDER);
if ($c) {
require_once 'include/Photo.php';
foreach ($matches as $mtch) {
2012-02-25 23:22:51 +01:00
logger('scale_external_image: ' . $mtch[1]);
$hostname = str_replace('www.', '', substr(System::baseUrl(), strpos(System::baseUrl(), '://') + 3));
if (stristr($mtch[1], $hostname)) {
2012-02-25 23:22:51 +01:00
continue;
}
// $scale_replace, if passed, is an array of two elements. The
// first is the name of the full-size image. The second is the
// name of a remote, scaled-down version of the full size image.
// This allows Friendica to display the smaller remote image if
// one exists, while still linking to the full-size image
if ($scale_replace) {
$scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
} else {
$scaled = $mtch[1];
}
$i = fetch_url($scaled);
if (! $i) {
2013-05-03 07:19:16 +02:00
return $srctext;
}
2012-06-07 17:42:13 +02:00
// guess mimetype from headers or filename
$type = guess_image_type($mtch[1], true);
2012-11-17 00:50:39 +01:00
if ($i) {
2012-06-07 17:42:13 +02:00
$ph = new Photo($i, $type);
if ($ph->is_valid()) {
2012-02-25 23:22:51 +01:00
$orig_width = $ph->getWidth();
$orig_height = $ph->getHeight();
if ($orig_width > 640 || $orig_height > 640) {
2012-02-25 23:22:51 +01:00
$ph->scaleImage(640);
$new_width = $ph->getWidth();
$new_height = $ph->getHeight();
logger('scale_external_images: ' . $orig_width . '->' . $new_width . 'w ' . $orig_height . '->' . $new_height . 'h' . ' match: ' . $mtch[0], LOGGER_DEBUG);
$s = str_replace(
$mtch[0],
'[img=' . $new_width . 'x' . $new_height. ']' . $scaled . '[/img]'
. "\n" . (($include_link)
2012-02-25 23:22:51 +01:00
? '[url=' . $mtch[1] . ']' . t('view full size') . '[/url]' . "\n"
: ''),
$s
);
2012-02-25 23:22:51 +01:00
logger('scale_external_images: new string: ' . $s, LOGGER_DEBUG);
}
}
}
}
}
// replace the special char encoding
$s = htmlspecialchars($s, ENT_NOQUOTES, 'UTF-8');
2012-02-25 23:22:51 +01:00
return $s;
}
function fix_contact_ssl_policy(&$contact, $new_policy)
{
$ssl_changed = false;
if ((intval($new_policy) == SSL_POLICY_SELFSIGN || $new_policy === 'self') && strstr($contact['url'], 'https:')) {
$ssl_changed = true;
$contact['url'] = str_replace('https:', 'http:', $contact['url']);
$contact['request'] = str_replace('https:', 'http:', $contact['request']);
$contact['notify'] = str_replace('https:', 'http:', $contact['notify']);
$contact['poll'] = str_replace('https:', 'http:', $contact['poll']);
$contact['confirm'] = str_replace('https:', 'http:', $contact['confirm']);
$contact['poco'] = str_replace('https:', 'http:', $contact['poco']);
}
if ((intval($new_policy) == SSL_POLICY_FULL || $new_policy === 'full') && strstr($contact['url'], 'http:')) {
$ssl_changed = true;
$contact['url'] = str_replace('http:', 'https:', $contact['url']);
$contact['request'] = str_replace('http:', 'https:', $contact['request']);
$contact['notify'] = str_replace('http:', 'https:', $contact['notify']);
$contact['poll'] = str_replace('http:', 'https:', $contact['poll']);
$contact['confirm'] = str_replace('http:', 'https:', $contact['confirm']);
$contact['poco'] = str_replace('http:', 'https:', $contact['poco']);
}
if ($ssl_changed) {
$fields = array('url' => $contact['url'], 'request' => $contact['request'],
'notify' => $contact['notify'], 'poll' => $contact['poll'],
'confirm' => $contact['confirm'], 'poco' => $contact['poco']);
dba::update('contact', $fields, array('id' => $contact['id']));
}
}
/**
* @brief Remove Google Analytics and other tracking platforms params from URL
*
2017-02-19 02:35:31 +01:00
* @param string $url Any user-submitted URL that may contain tracking params
* @return string The same URL stripped of tracking parameters
*/
function strip_tracking_query_params($url)
{
$urldata = parse_url($url);
if (is_string($urldata["query"])) {
$query = $urldata["query"];
parse_str($query, $querydata);
if (is_array($querydata)) {
foreach ($querydata as $param => $value) {
if (in_array(
$param,
array(
"utm_source", "utm_medium", "utm_term", "utm_content", "utm_campaign",
"wt_mc", "pk_campaign", "pk_kwd", "mc_cid", "mc_eid",
"fb_action_ids", "fb_action_types", "fb_ref",
"awesm", "wtrid",
"woo_campaign", "woo_source", "woo_medium", "woo_content", "woo_term")
)
) {
$pair = $param . "=" . urlencode($value);
$url = str_replace($pair, "", $url);
// Second try: if the url isn't encoded completely
$pair = $param . "=" . str_replace(" ", "+", $value);
$url = str_replace($pair, "", $url);
// Third try: Maybey the url isn't encoded at all
$pair = $param . "=" . $value;
$url = str_replace($pair, "", $url);
$url = str_replace(array("?&", "&&"), array("?", ""), $url);
}
}
}
if (substr($url, -1, 1) == "?") {
$url = substr($url, 0, -1);
}
}
return $url;
}
/**
* @brief Returns the original URL of the provided URL
*
* This function strips tracking query params and follows redirections, either
* through HTTP code or meta refresh tags. Stops after 10 redirections.
*
2017-02-19 02:39:16 +01:00
* @todo Remove the $fetchbody parameter that generates an extraneous HEAD request
*
* @see ParseUrl::getSiteinfo
*
* @param string $url A user-submitted URL
* @param int $depth The current redirection recursion level (internal)
* @param bool $fetchbody Wether to fetch the body or not after the HEAD requests
2017-02-19 02:39:16 +01:00
* @return string A canonical URL
*/
function original_url($url, $depth = 1, $fetchbody = false)
{
$a = get_app();
$url = strip_tracking_query_params($url);
if ($depth > 10) {
2015-03-07 23:14:26 +01:00
return($url);
}
2015-03-07 23:14:26 +01:00
$url = trim($url, "'");
2015-03-07 23:14:26 +01:00
$stamp1 = microtime(true);
2015-03-07 23:14:26 +01:00
$siteinfo = array();
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, 1);
curl_setopt($ch, CURLOPT_NOBODY, 1);
2015-03-07 23:14:26 +01:00
curl_setopt($ch, CURLOPT_TIMEOUT, 10);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_USERAGENT, $a->get_useragent());
2015-03-07 23:14:26 +01:00
$header = curl_exec($ch);
$curl_info = @curl_getinfo($ch);
$http_code = $curl_info['http_code'];
curl_close($ch);
$a->save_timestamp($stamp1, "network");
if ($http_code == 0)
return($url);
if ((($curl_info['http_code'] == "301") || ($curl_info['http_code'] == "302"))
&& (($curl_info['redirect_url'] != "") || ($curl_info['location'] != ""))
) {
if ($curl_info['redirect_url'] != "") {
2015-03-07 23:14:26 +01:00
return(original_url($curl_info['redirect_url'], ++$depth, $fetchbody));
} else {
2015-03-07 23:14:26 +01:00
return(original_url($curl_info['location'], ++$depth, $fetchbody));
}
2015-03-07 23:14:26 +01:00
}
// Check for redirects in the meta elements of the body if there are no redirects in the header.
if (!$fetchbody) {
return(original_url($url, ++$depth, true));
}
// if the file is too large then exit
if ($curl_info["download_content_length"] > 1000000) {
return($url);
}
// if it isn't a HTML file then exit
if (($curl_info["content_type"] != "") && !strstr(strtolower($curl_info["content_type"]), "html")) {
return($url);
}
2015-03-07 23:14:26 +01:00
$stamp1 = microtime(true);
$ch = curl_init();
curl_setopt($ch, CURLOPT_URL, $url);
curl_setopt($ch, CURLOPT_HEADER, 0);
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());
$body = curl_exec($ch);
curl_close($ch);
2015-03-07 23:14:26 +01:00
$a->save_timestamp($stamp1, "network");
if (trim($body) == "") {
return($url);
}
// Check for redirect in meta elements
2015-03-07 23:14:26 +01:00
$doc = new DOMDocument();
@$doc->loadHTML($body);
$xpath = new DomXPath($doc);
$list = $xpath->query("//meta[@content]");
foreach ($list as $node) {
$attr = array();
if ($node->attributes->length) {
foreach ($node->attributes as $attribute) {
2015-03-07 23:14:26 +01:00
$attr[$attribute->name] = $attribute->value;
}
}
2015-03-07 23:14:26 +01:00
if (@$attr["http-equiv"] == 'refresh') {
$path = $attr["content"];
$pathinfo = explode(";", $path);
$content = "";
foreach ($pathinfo as $value) {
if (substr(strtolower($value), 0, 4) == "url=") {
2015-03-07 23:14:26 +01:00
return(original_url(substr($value, 4), ++$depth));
}
}
2015-03-07 23:14:26 +01:00
}
}
return($url);
}
function short_link($url)
{
require_once 'library/slinky.php';
$slinky = new Slinky($url);
$yourls_url = Config::get('yourls', 'url1');
if ($yourls_url) {
$yourls_username = Config::get('yourls', 'username1');
$yourls_password = Config::get('yourls', 'password1');
$yourls_ssl = Config::get('yourls', 'ssl1');
$yourls = new Slinky_YourLS();
$yourls->set('username', $yourls_username);
$yourls->set('password', $yourls_password);
$yourls->set('ssl', $yourls_ssl);
$yourls->set('yourls-url', $yourls_url);
2016-11-14 06:44:18 +01:00
$slinky->set_cascade(array($yourls, new Slinky_Ur1ca(), new Slinky_TinyURL()));
} else {
// setup a cascade of shortening services
// try to get a short link from these services
// in the order ur1.ca, tinyurl
2016-11-14 06:44:18 +01:00
$slinky->set_cascade(array(new Slinky_Ur1ca(), new Slinky_TinyURL()));
}
return $slinky->short();
}
/**
* @brief Encodes content to json
2017-02-18 04:32:33 +01:00
*
* This function encodes an array to json format
* and adds an application/json HTTP header to the output.
* After finishing the process is getting killed.
*
2016-04-14 17:05:19 +02:00
* @param array $x The input content
*/
function json_return_and_die($x)
{
header("content-type: application/json");
echo json_encode($x);
killme();
}
/**
* @brief Find the matching part between two url
*
* @param string $url1
* @param string $url2
* @return string The matching part
*/
function matching_url($url1, $url2)
{
if (($url1 == "") || ($url2 == "")) {
return "";
}
$url1 = normalise_link($url1);
$url2 = normalise_link($url2);
$parts1 = parse_url($url1);
$parts2 = parse_url($url2);
if (!isset($parts1["host"]) || !isset($parts2["host"])) {
return "";
}
if ($parts1["scheme"] != $parts2["scheme"]) {
return "";
}
if ($parts1["host"] != $parts2["host"]) {
return "";
}
if ($parts1["port"] != $parts2["port"]) {
return "";
}
$match = $parts1["scheme"]."://".$parts1["host"];
if ($parts1["port"]) {
$match .= ":".$parts1["port"];
}
$pathparts1 = explode("/", $parts1["path"]);
$pathparts2 = explode("/", $parts2["path"]);
$i = 0;
$path = "";
do {
$path1 = $pathparts1[$i];
$path2 = $pathparts2[$i];
if ($path1 == $path2) {
$path .= $path1."/";
}
} while (($path1 == $path2) && ($i++ <= count($pathparts1)));
$match .= $path;
return normalise_link($match);
}
/**
* @brief Glue url parts together
*
* @param array $parsed URL parts
*
* @return string The glued URL
*/
function unParseUrl($parsed)
{
$get = function ($key) use ($parsed) {
return isset($parsed[$key]) ? $parsed[$key] : null;
};
$pass = $get('pass');
$user = $get('user');
$userinfo = $pass !== null ? "$user:$pass" : $user;
$port = $get('port');
$scheme = $get('scheme');
$query = $get('query');
$fragment = $get('fragment');
$authority = ($userinfo !== null ? $userinfo."@" : '') .
$get('host') .
($port ? ":$port" : '');
return (strlen($scheme) ? $scheme.":" : '') .
(strlen($authority) ? "//".$authority : '') .
$get('path') .
(strlen($query) ? "?".$query : '') .
(strlen($fragment) ? "#".$fragment : '');
}