* @file include/network.php
*/
-require_once("include/xml.php");
+use \Friendica\Core\Config;
+require_once("include/xml.php");
+require_once('include/Probe.php');
/**
* @brief Curl wrapper
- *
+ *
* 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.
- *
+ *
* @param string $url URL to fetch
* @param boolean $binary default false
* TRUE if asked to return binary results (file download)
* @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 $cookiejar Path to cookie jar file
- *
+ *
* @return string The fetched content
*/
function fetch_url($url,$binary = false, &$redirects = 0, $timeout = 0, $accept_content=Null, $cookiejar = 0) {
@curl_setopt($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);
+ }
if(x($opts,'headers')){
@curl_setopt($ch, CURLOPT_HTTPHEADER, $opts['headers']);
$check_cert = get_config('system','verifyssl');
@curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
- @curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
+ if ($check_cert) {
+ @curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 2);
+ }
$prx = get_config('system','proxy');
if(strlen($prx)) {
/**
* @brief Post request to $url
- *
+ *
* @param string $url URL to post
* @param mixed $params
* @param string $headers HTTP headers
* @param integer $redirects Recursion counter for internal use - default = 0
* @param integer $timeout The timeout in seconds, default system config value or 60 seconds
- *
+ *
* @return string The content
*/
function post_url($url,$params, $headers = null, &$redirects = 0, $timeout = 0) {
$check_cert = get_config('system','verifyssl');
curl_setopt($ch, CURLOPT_SSL_VERIFYPEER, (($check_cert) ? true : false));
- curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, (($check_cert) ? 2 : false));
+ if ($check_cert) {
+ @curl_setopt($ch, CURLOPT_SSL_VERIFYHOST, 2);
+ }
$prx = get_config('system','proxy');
if(strlen($prx)) {
curl_setopt($ch, CURLOPT_HTTPPROXYTUNNEL, 1);
killme();
}
-
/**
* @brief Send HTTP status header and exit.
*
* '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) {
}
-// Given an email style address, perform webfinger lookup and
-// return the resulting DFRN profile URL, or if no DFRN profile URL
-// is located, returns an OStatus subscription template (prefixed
-// with the string 'stat:' to identify it as on OStatus template).
-// If this isn't an email style address just return $webbie.
-// Return an empty string if email-style addresses but webfinger fails,
-// or if the resultant personal XRD doesn't contain a supported
-// subscription/friend-request attribute.
-
-// amended 7/9/2011 to return an hcard which could save potentially loading
-// a lengthy content page to scrape dfrn attributes
-
-function webfinger_dfrn($webbie,&$hcard) {
- if(! strstr($webbie,'@')) {
- return $webbie;
- }
- $profile_link = '';
-
- $links = webfinger($webbie);
- logger('webfinger_dfrn: ' . $webbie . ':' . print_r($links,true), LOGGER_DATA);
- if(count($links)) {
- foreach($links as $link) {
- if($link['@attributes']['rel'] === NAMESPACE_DFRN)
- $profile_link = $link['@attributes']['href'];
- if($link['@attributes']['rel'] === NAMESPACE_OSTATUSSUB)
- $profile_link = 'stat:' . $link['@attributes']['template'];
- if($link['@attributes']['rel'] === 'http://microformats.org/profile/hcard')
- $hcard = $link['@attributes']['href'];
- }
- }
- return $profile_link;
-}
-
-/**
- * @brief Perform webfinger lookup on an email style address
- *
- * @param string $webbi An email style address
- * @param boolean $debug
- *
- * @return array of link attributes from the personal XRD file
- * empty array on error/failure
- */
-function webfinger($webbie, $debug = false) {
- $host = '';
- if(strstr($webbie,'@')) {
- $host = substr($webbie,strpos($webbie,'@') + 1);
- }
- if(strlen($host)) {
- $tpl = fetch_lrdd_template($host);
- logger('webfinger: lrdd template: ' . $tpl);
- if(strlen($tpl)) {
- $pxrd = str_replace('{uri}', urlencode('acct:' . $webbie), $tpl);
- logger('webfinger: pxrd: ' . $pxrd);
- $links = fetch_xrd_links($pxrd);
- if(! count($links)) {
- // try with double slashes
- $pxrd = str_replace('{uri}', urlencode('acct://' . $webbie), $tpl);
- logger('webfinger: pxrd: ' . $pxrd);
- $links = fetch_xrd_links($pxrd);
- }
- return $links;
- }
- }
- return array();
-}
-
-function lrdd($uri, $debug = false) {
-
- $a = get_app();
-
- // default priority is host priority, host-meta first
-
- $priority = 'host';
-
- // All we have is an email address. Resource-priority is irrelevant
- // because our URI isn't directly resolvable.
-
- if(strstr($uri,'@')) {
- return(webfinger($uri));
- }
-
- // get the host meta file
-
- $host = @parse_url($uri);
-
- if($host) {
- $url = ((x($host,'scheme')) ? $host['scheme'] : 'http') . '://';
- $url .= $host['host'] . '/.well-known/host-meta' ;
- }
- else
- return array();
-
- logger('lrdd: constructed url: ' . $url);
-
- $xml = fetch_url($url);
-
- $headers = $a->get_curl_headers();
-
- if (! $xml)
- return array();
-
- logger('lrdd: host_meta: ' . $xml, LOGGER_DATA);
-
- if(! stristr($xml,'<xrd'))
- return array();
-
- $h = parse_xml_string($xml);
- if(! $h)
- return array();
-
- $arr = xml::element_to_array($h);
-
- if(isset($arr['xrd']['property'])) {
- $property = $arr['crd']['property'];
- if(! isset($property[0]))
- $properties = array($property);
- else
- $properties = $property;
- foreach($properties as $prop)
- if((string) $prop['@attributes'] === 'http://lrdd.net/priority/resource')
- $priority = 'resource';
- }
-
- // save the links in case we need them
-
- $links = array();
-
- if(isset($arr['xrd']['link'])) {
- $link = $arr['xrd']['link'];
- if(! isset($link[0]))
- $links = array($link);
- else
- $links = $link;
- }
-
- // do we have a template or href?
-
- if(count($links)) {
- foreach($links as $link) {
- if($link['@attributes']['rel'] && attribute_contains($link['@attributes']['rel'],'lrdd')) {
- if(x($link['@attributes'],'template'))
- $tpl = $link['@attributes']['template'];
- elseif(x($link['@attributes'],'href'))
- $href = $link['@attributes']['href'];
- }
- }
- }
-
- if((! isset($tpl)) || (! strpos($tpl,'{uri}')))
- $tpl = '';
-
- if($priority === 'host') {
- if(strlen($tpl))
- $pxrd = str_replace('{uri}', urlencode($uri), $tpl);
- elseif(isset($href))
- $pxrd = $href;
- if(isset($pxrd)) {
- logger('lrdd: (host priority) pxrd: ' . $pxrd);
- $links = fetch_xrd_links($pxrd);
- return $links;
- }
-
- $lines = explode("\n",$headers);
- if(count($lines)) {
- foreach($lines as $line) {
- if((stristr($line,'link:')) && preg_match('/<([^>].*)>.*rel\=[\'\"]lrdd[\'\"]/',$line,$matches)) {
- return(fetch_xrd_links($matches[1]));
- break;
- }
- }
- }
- }
-
-
- // priority 'resource'
-
-
- $html = fetch_url($uri);
- $headers = $a->get_curl_headers();
- logger('lrdd: headers=' . $headers, LOGGER_DEBUG);
-
- // don't try and parse raw xml as html
- if(! strstr($html,'<?xml')) {
- require_once('library/HTML5/Parser.php');
-
- try {
- $dom = HTML5_Parser::parse($html);
- } catch (DOMException $e) {
- logger('lrdd: parse error: ' . $e);
- }
-
- if(isset($dom) && $dom) {
- $items = $dom->getElementsByTagName('link');
- foreach($items as $item) {
- $x = $item->getAttribute('rel');
- if($x == "lrdd") {
- $pagelink = $item->getAttribute('href');
- break;
- }
- }
- }
- }
-
- if(isset($pagelink))
- return(fetch_xrd_links($pagelink));
-
- // next look in HTTP headers
-
- $lines = explode("\n",$headers);
- if(count($lines)) {
- foreach($lines as $line) {
- /// @TODO Alter the following regex to support multiple relations (space separated)
- if((stristr($line,'link:')) && preg_match('/<([^>].*)>.*rel\=[\'\"]lrdd[\'\"]/',$line,$matches)) {
- $pagelink = $matches[1];
- break;
- }
- // don't try and run feeds through the html5 parser
- if(stristr($line,'content-type:') && ((stristr($line,'application/atom+xml')) || (stristr($line,'application/rss+xml'))))
- return array();
- if(stristr($html,'<rss') || stristr($html,'<feed'))
- return array();
- }
- }
-
- if(isset($pagelink))
- return(fetch_xrd_links($pagelink));
-
- // If we haven't found any links, return the host xrd links (which we have already fetched)
-
- if(isset($links))
- return $links;
-
- return array();
-
-}
-
-// Given a host name, locate the LRDD template from that
-// host. Returns the LRDD template or an empty string on
-// error/failure.
-
-function fetch_lrdd_template($host) {
- $tpl = '';
-
- $url1 = 'https://' . $host . '/.well-known/host-meta' ;
- $url2 = 'http://' . $host . '/.well-known/host-meta' ;
- $links = fetch_xrd_links($url1);
- logger('fetch_lrdd_template from: ' . $url1);
- logger('template (https): ' . print_r($links,true));
- if(! count($links)) {
- logger('fetch_lrdd_template from: ' . $url2);
- $links = fetch_xrd_links($url2);
- logger('template (http): ' . print_r($links,true));
- }
- if(count($links)) {
- foreach($links as $link)
- if($link['@attributes']['rel'] && $link['@attributes']['rel'] === 'lrdd' && (!$link['@attributes']['type'] || $link['@attributes']['type'] === 'application/xrd+xml'))
- $tpl = $link['@attributes']['template'];
- }
- if(! strpos($tpl,'{uri}'))
- $tpl = '';
- return $tpl;
-}
-
-/**
- * @brief Given a URL, retrieve the page as an XRD document.
- *
- * @param string $url An url
- * @return array of links
- * return empty array on error/failure
- */
-function fetch_xrd_links($url) {
-
- $xrd_timeout = intval(get_config('system','xrd_timeout'));
- $redirects = 0;
- $xml = fetch_url($url,false,$redirects,(($xrd_timeout) ? $xrd_timeout : 20), "application/xrd+xml");
-
- logger('fetch_xrd_links: ' . $xml, LOGGER_DATA);
-
- if ((! $xml) || (! stristr($xml,'<xrd')))
- return array();
-
- // fix diaspora's bad xml
- $xml = str_replace(array('href="','"/>'),array('href="','"/>'),$xml);
-
- $h = parse_xml_string($xml);
- if(! $h)
- return array();
-
- $arr = xml::element_to_array($h);
-
- $links = array();
-
- if(isset($arr['xrd']['link'])) {
- $link = $arr['xrd']['link'];
- if(! isset($link[0]))
- $links = array($link);
- else
- $links = $link;
- }
- if(isset($arr['xrd']['alias'])) {
- $alias = $arr['xrd']['alias'];
- if(! isset($alias[0]))
- $aliases = array($alias);
- else
- $aliases = $alias;
- if(is_array($aliases) && count($aliases)) {
- foreach($aliases as $alias) {
- $links[]['@attributes'] = array('rel' => 'alias' , 'href' => $alias);
- }
- }
- }
-
- logger('fetch_xrd_links: ' . print_r($links,true), LOGGER_DATA);
-
- return $links;
-
-}
-
/**
* @brief Check URL to se if ts's real
- *
+ *
* 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)
- *
+ *
* @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(get_config('system','disable_url_validation'))
return true;
+
// no naked subdomains (allow localhost for tests)
if(strpos($url,'.') === false && strpos($url,'/localhost/') === false)
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(($h) && (dns_get_record($h['host'], DNS_A + DNS_CNAME + DNS_PTR) || filter_var($h['host'], FILTER_VALIDATE_IP) )) {
+ if((is_array($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
- *
+ *
* @param string $addr The email address
* @return boolean True if it's a valid email address, false if it's not
*/
/**
* @brief Check if URL is allowed
- *
+ *
* Check $url against our list of allowed sites,
* wildcards allowed. If allowed_sites is unset return true;
- *
+ *
* @param string $url URL which get tested
* @return boolean True if url is allowed otherwise return false
*/
/**
* @brief Check if email address is allowed to register here.
- *
+ *
* Compare against our list (wildcards allowed).
- *
+ *
* @param type $email
* @return boolean False if not allowed, true if allowed
* or if allowed list is not configured
function avatar_img($email) {
- $a = get_app();
-
$avatar['size'] = 175;
$avatar['email'] = $email;
$avatar['url'] = '';
call_hooks('avatar_lookup', $avatar);
- if(! $avatar['success'])
- $avatar['url'] = $a->get_baseurl() . '/images/person-175.jpg';
+ if (! $avatar['success']) {
+ $avatar['url'] = App::get_baseurl() . '/images/person-175.jpg';
+ }
logger('Avatar: ' . $avatar['email'] . ' ' . $avatar['url'], LOGGER_DEBUG);
return $avatar['url'];
libxml_use_internal_errors(true);
$x = @simplexml_load_string($s2);
- if(! $x) {
+ if (! $x) {
logger('libxml: parse: error: ' . $s2, 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);
+ }
libxml_clear_errors();
}
return $x;
function scale_external_images($srctext, $include_link = true, $scale_replace = false) {
// Suppress "view full size"
- if (intval(get_config('system','no_view_full_size')))
+ if (intval(get_config('system','no_view_full_size'))) {
$include_link = false;
+ }
$a = get_app();
$matches = null;
$c = preg_match_all('/\[img.*?\](.*?)\[\/img\]/ism',$s,$matches,PREG_SET_ORDER);
- if($c) {
+ if ($c) {
require_once('include/Photo.php');
- foreach($matches as $mtch) {
+ foreach ($matches as $mtch) {
logger('scale_external_image: ' . $mtch[1]);
- $hostname = str_replace('www.','',substr($a->get_baseurl(),strpos($a->get_baseurl(),'://')+3));
- if(stristr($mtch[1],$hostname))
+ $hostname = str_replace('www.','',substr(App::get_baseurl(),strpos(App::get_baseurl(),'://')+3));
+ if (stristr($mtch[1],$hostname)) {
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)
+ if ($scale_replace) {
$scaled = str_replace($scale_replace[0], $scale_replace[1], $mtch[1]);
- else
+ } else {
$scaled = $mtch[1];
- $i = @fetch_url($scaled);
- if(! $i)
+ }
+ $i = fetch_url($scaled);
+ if (! $i) {
return $srctext;
+ }
// guess mimetype from headers or filename
$type = guess_image_type($mtch[1],true);
- if($i) {
+ if ($i) {
$ph = new Photo($i, $type);
- if($ph->is_valid()) {
+ if ($ph->is_valid()) {
$orig_width = $ph->getWidth();
$orig_height = $ph->getHeight();
- if($orig_width > 640 || $orig_height > 640) {
+ if ($orig_width > 640 || $orig_height > 640) {
$ph->scaleImage(640);
$new_width = $ph->getWidth();
function fix_contact_ssl_policy(&$contact,$new_policy) {
$ssl_changed = false;
- if((intval($new_policy) == SSL_POLICY_SELFSIGN || $new_policy === 'self') && strstr($contact['url'],'https:')) {
+ 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']);
$contact['poco'] = str_replace('https:','http:',$contact['poco']);
}
- if((intval($new_policy) == SSL_POLICY_FULL || $new_policy === 'full') && strstr($contact['url'],'http:')) {
+ 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']);
$contact['poco'] = str_replace('http:','https:',$contact['poco']);
}
- if($ssl_changed) {
- q("update contact set
- url = '%s',
- request = '%s',
- notify = '%s',
- poll = '%s',
- confirm = '%s',
- poco = '%s'
- where id = %d limit 1",
+ if ($ssl_changed) {
+ q("UPDATE `contact` SET
+ `url` = '%s',
+ `request` = '%s',
+ `notify` = '%s',
+ `poll` = '%s',
+ `confirm` = '%s',
+ `poco` = '%s'
+ WHERE `id` = %d LIMIT 1",
dbesc($contact['url']),
dbesc($contact['request']),
dbesc($contact['notify']),
}
}
-function original_url($url, $depth=1, $fetchbody = false) {
-
- $a = get_app();
-
- // Remove Analytics Data from Google and other tracking platforms
+/**
+ * @brief Remove Google Analytics and other tracking platforms params from URL
+ *
+ * @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 (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);
+ $pair = $param . "=" . urlencode($value);
$url = str_replace($pair, "", $url);
// Second try: if the url isn't encoded completely
- $pair = $param."=".str_replace(" ", "+", $value);
+ $pair = $param . "=" . str_replace(" ", "+", $value);
$url = str_replace($pair, "", $url);
// Third try: Maybey the url isn't encoded at all
- $pair = $param."=".$value;
+ $pair = $param . "=" . $value;
$url = str_replace($pair, "", $url);
$url = str_replace(array("?&", "&&"), array("?", ""), $url);
}
+ }
+ }
- if (substr($url, -1, 1) == "?")
+ 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.
+ *
+ * @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 $fetchbody Wether to fetch the body or not after the HEAD requests
+ * @return string A canonical URL
+ */
+function original_url($url, $depth = 1, $fetchbody = false) {
+ $a = get_app();
+
+ $url = strip_tracking_query_params($url);
+
if ($depth > 10)
return($url);
$yourls->set('password', $yourls_password);
$yourls->set('ssl', $yourls_ssl);
$yourls->set('yourls-url', $yourls_url);
- $slinky->set_cascade( array($yourls, new Slinky_UR1ca(), new Slinky_Trim(), new Slinky_IsGd(), new Slinky_TinyURL()));
+ $slinky->set_cascade(array($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, trim, id.gd, tinyurl
- $slinky->set_cascade(array(new Slinky_UR1ca(), new Slinky_Trim(), new Slinky_IsGd(), new Slinky_TinyURL()));
+ // in the order ur1.ca, tinyurl
+ $slinky->set_cascade(array(new Slinky_Ur1ca(), new Slinky_TinyURL()));
}
return $slinky->short();
}
/**
* @brief Encodes content to json
- *
+ *
* 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.