function parse_event($h)
{
require_once 'include/Scrape.php';
require_once 'library/HTMLPurifier.auto.php';
require_once 'include/html2bbcode';
$h = '<html><body>' . $h . '</body></html>';
$ret = array();
try {
$dom = HTML5_Parser::parse($h);
} catch (DOMException $e) {
logger('parse_event: parse error: ' . $e);
}
if (!$dom) {
return $ret;
}
$items = $dom->getElementsByTagName('*');
foreach ($items as $item) {
if (attribute_contains($item->getAttribute('class'), 'vevent')) {
$level2 = $item->getElementsByTagName('*');
foreach ($level2 as $x) {
if (attribute_contains($x->getAttribute('class'), 'dtstart') && $x->getAttribute('title')) {
$ret['start'] = $x->getAttribute('title');
if (!strpos($ret['start'], 'Z')) {
$ret['adjust'] = true;
}
}
if (attribute_contains($x->getAttribute('class'), 'dtend') && $x->getAttribute('title')) {
$ret['finish'] = $x->getAttribute('title');
}
if (attribute_contains($x->getAttribute('class'), 'description')) {
$ret['desc'] = $x->textContent;
}
if (attribute_contains($x->getAttribute('class'), 'location')) {
$ret['location'] = $x->textContent;
}
}
}
}
// sanitise
if (x($ret, 'desc') && (strpos($ret['desc'], '<') !== false || strpos($ret['desc'], '>') !== false)) {
$config = HTMLPurifier_Config::createDefault();
$config->set('Cache.DefinitionImpl', null);
$purifier = new HTMLPurifier($config);
$ret['desc'] = html2bbcode($purifier->purify($ret['desc']));
}
if (x($ret, 'location') && (strpos($ret['location'], '<') !== false || strpos($ret['location'], '>') !== false)) {
$config = HTMLPurifier_Config::createDefault();
$config->set('Cache.DefinitionImpl', null);
$purifier = new HTMLPurifier($config);
$ret['location'] = html2bbcode($purifier->purify($ret['location']));
}
if (x($ret, 'start')) {
$ret['start'] = datetime_convert('UTC', 'UTC', $ret['start']);
}
if (x($ret, 'finish')) {
$ret['finish'] = datetime_convert('UTC', 'UTC', $ret['finish']);
}
return $ret;
}
function scrape_dfrn($url)
{
$ret = array();
$s = fetch_url($url);
if (!$s) {
return $ret;
}
$dom = HTML5_Parser::parse($s);
if (!$dom) {
return $ret;
}
$items = $dom->getElementsByTagName('link');
// get DFRN link elements
foreach ($items as $item) {
$x = $item->getAttribute('rel');
if (substr($x, 0, 5) == "dfrn-") {
$ret[$x] = $item->getAttribute('href');
}
}
// Pull out hCard profile elements
$items = $dom->getElementsByTagName('*');
foreach ($items as $item) {
if (attribute_contains($item->getAttribute('class'), 'vcard')) {
$level2 = $item->getElementsByTagName('*');
foreach ($level2 as $x) {
if (attribute_contains($x->getAttribute('class'), 'fn')) {
$ret['fn'] = $x->textContent;
}
if (attribute_contains($x->getAttribute('class'), 'photo')) {
$ret['photo'] = $x->getAttribute('src');
}
if (attribute_contains($x->getAttribute('class'), 'key')) {
$ret['key'] = $x->textContent;
}
}
}
}
return $ret;
}
function scrape_vcard($url)
{
$a = get_app();
$ret = array();
logger('scrape_vcard: url=' . $url);
$x = z_fetch_url($url);
if (!$x['success']) {
return $ret;
}
$s = $x['body'];
if (!$s) {
return $ret;
}
$headers = $x['header'];
$lines = explode("\n", $headers);
if (count($lines)) {
foreach ($lines as $line) {
// don't try and run feeds through the html5 parser
if (stristr($line, 'content-type:') && (stristr($line, 'application/atom+xml') || stristr($line, 'application/rss+xml'))) {
return ret;
}
}
}
try {
$dom = HTML5_Parser::parse($s);
} catch (DOMException $e) {
logger('scrape_vcard: parse error: ' . $e);
}
if (!$dom) {
return $ret;
}
// Pull out hCard profile elements
$largest_photo = 0;
$items = $dom->getElementsByTagName('*');
foreach ($items as $item) {
if (attribute_contains($item->getAttribute('class'), 'vcard')) {
$level2 = $item->getElementsByTagName('*');
foreach ($level2 as $x) {
if (attribute_contains($x->getAttribute('class'), 'fn')) {
$ret['fn'] = $x->textContent;
}
if (attribute_contains($x->getAttribute('class'), 'photo') || attribute_contains($x->getAttribute('class'), 'avatar')) {
$size = intval($x->getAttribute('width'));
if ($size > $largest_photo || !$largest_photo) {
$ret['photo'] = $x->getAttribute('src');
$largest_photo = $size;
}
}
if (attribute_contains($x->getAttribute('class'), 'nickname') || attribute_contains($x->getAttribute('class'), 'uid')) {
$ret['nick'] = $x->textContent;
}
}
}
}
return $ret;
}
function lrdd($uri)
{
$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);
$h = parse_xml_string($xml);
if (!$h) {
return array();
}
$arr = convert_xml_element_to_array($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: ' . $e);
}
if ($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 $links;
}
return array();
}
function scrape_dfrn($url, $max_nodes = 3500)
{
$minNodes = 100;
//Lets do at least 100 nodes per type.
$timeout = 10;
//Timeout will affect batch processing.
//Try and cheat our way into faster profiles.
if (strpos($url, 'tab=profile') === false) {
$url .= (strpos($url, '?') > 0 ? '&' : '?') . 'tab=profile';
}
$scrape_start = microtime(true);
$ret = array();
$s = fetch_url($url, $timeout);
$scrape_fetch_end = microtime(true);
if (!$s) {
return $ret;
}
$dom = HTML5_Parser::parse($s);
if (!$dom) {
return $ret;
}
$items = $dom->getElementsByTagName('meta');
// get DFRN link elements
$nodes_left = max(intval($max_nodes), $minNodes);
$targets = array('hide', 'comm', 'tags');
$targets_left = count($targets);
foreach ($items as $item) {
$x = $item->getAttribute('name');
if ($x == 'dfrn-global-visibility') {
$z = strtolower(trim($item->getAttribute('content')));
if ($z != 'true') {
$ret['hide'] = 1;
}
if ($z === 'false') {
$ret['explicit-hide'] = 1;
}
$targets_left = pop_scrape_target($targets, 'hide');
}
if ($x == 'friendika.community' || $x == 'friendica.community') {
$z = strtolower(trim($item->getAttribute('content')));
if ($z == 'true') {
$ret['comm'] = 1;
}
$targets_left = pop_scrape_target($targets, 'comm');
}
if ($x == 'keywords') {
$z = str_replace(',', ' ', strtolower(trim($item->getAttribute('content'))));
if (strlen($z)) {
$ret['tags'] = $z;
}
$targets_left = pop_scrape_target($targets, 'tags');
}
$nodes_left--;
if ($nodes_left <= 0 || $targets_left <= 0) {
break;
}
}
$items = $dom->getElementsByTagName('link');
// get DFRN link elements
$nodes_left = max(intval($max_nodes), $minNodes);
foreach ($items as $item) {
$x = $item->getAttribute('rel');
if (substr($x, 0, 5) == "dfrn-") {
$ret[$x] = $item->getAttribute('href');
}
$nodes_left--;
if ($nodes_left <= 0) {
break;
}
}
// Pull out hCard profile elements
$nodes_left = max(intval($max_nodes), $minNodes);
$items = $dom->getElementsByTagName('*');
$targets = array('fn', 'pdesc', 'photo', 'key', 'locality', 'region', 'postal-code', 'country-name', 'gender', 'marital');
$targets_left = count($targets);
foreach ($items as $item) {
if (attribute_contains($item->getAttribute('class'), 'vcard')) {
$level2 = $item->getElementsByTagName('*');
foreach ($level2 as $x) {
if (attribute_contains($x->getAttribute('class'), 'fn')) {
$ret['fn'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'fn');
}
if (attribute_contains($x->getAttribute('class'), 'title')) {
$ret['pdesc'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'pdesc');
}
if (attribute_contains($x->getAttribute('class'), 'photo')) {
$ret['photo'] = $x->getAttribute('src');
$targets_left = pop_scrape_target($targets, 'photo');
}
if (attribute_contains($x->getAttribute('class'), 'key')) {
$ret['key'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'key');
}
if (attribute_contains($x->getAttribute('class'), 'locality')) {
$ret['locality'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'locality');
}
if (attribute_contains($x->getAttribute('class'), 'region')) {
$ret['region'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'region');
}
if (attribute_contains($x->getAttribute('class'), 'postal-code')) {
$ret['postal-code'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'postal-code');
}
if (attribute_contains($x->getAttribute('class'), 'country-name')) {
$ret['country-name'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'country-name');
}
if (attribute_contains($x->getAttribute('class'), 'x-gender')) {
$ret['gender'] = $x->textContent;
$targets_left = pop_scrape_target($targets, 'gender');
}
}
}
if (attribute_contains($item->getAttribute('class'), 'marital-text')) {
$ret['marital'] = $item->textContent;
$targets_left = pop_scrape_target($targets, 'marital');
}
$nodes_left--;
if ($nodes_left <= 0 || $targets_left <= 0) {
break;
}
}
$scrape_end = microtime(true);
$fetch_time = round(($scrape_fetch_end - $scrape_start) * 1000);
$scrape_time = round(($scrape_end - $scrape_fetch_end) * 1000);
$ret['_timings'] = array('fetch' => $fetch_time, 'scrape' => $scrape_time);
return $ret;
}
/**
* test input with special chars
*/
public function testAttributeContainsSpecialChars()
{
$testAttr = "--... %\$ä() /(=?}";
$this->assertFalse(attribute_contains($testAttr, "class2"));
}