/**
* Examine a page at {@link $Url} for title, description & images.
*
* Be sure to check the resultant array for any Exceptions that occurred while retrieving the page.
*
* @param string $url The url to examine.
* @param integer $timeout How long to allow for this request.
* Default Garden.SocketTimeout or 1, 0 to never timeout. Default is 0.
* @param bool $sendCookies Whether or not to send browser cookies with the request.
* @return array Returns an array containing Url, Title, Description, Images (array) and Exception
* (if there were problems retrieving the page).
*/
function fetchPageInfo($url, $timeout = 3, $sendCookies = false)
{
$PageInfo = array('Url' => $url, 'Title' => '', 'Description' => '', 'Images' => array(), 'Exception' => false);
try {
// Make sure the URL is valid.
$urlParts = parse_url($url);
if ($urlParts === false || !in_array(val('scheme', $urlParts), array('http', 'https'))) {
throw new Exception('Invalid URL.', 400);
}
if (!defined('HDOM_TYPE_ELEMENT')) {
require_once PATH_LIBRARY . '/vendors/simplehtmldom/simple_html_dom.php';
}
$Request = new ProxyRequest();
$PageHtml = $Request->Request(array('URL' => $url, 'Timeout' => $timeout, 'Cookies' => $sendCookies));
if (!$Request->status()) {
throw new Exception('Couldn\'t connect to host.', 400);
}
$Dom = str_get_html($PageHtml);
if (!$Dom) {
throw new Exception('Failed to load page for parsing.');
}
// FIRST PASS: Look for open graph title, desc, images
$PageInfo['Title'] = domGetContent($Dom, 'meta[property=og:title]');
Trace('Getting og:description');
$PageInfo['Description'] = domGetContent($Dom, 'meta[property=og:description]');
foreach ($Dom->find('meta[property=og:image]') as $Image) {
if (isset($Image->content)) {
$PageInfo['Images'][] = $Image->content;
}
}
// SECOND PASS: Look in the page for title, desc, images
if ($PageInfo['Title'] == '') {
$PageInfo['Title'] = $Dom->find('title', 0)->plaintext;
}
if ($PageInfo['Description'] == '') {
Trace('Getting meta description');
$PageInfo['Description'] = domGetContent($Dom, 'meta[name=description]');
}
// THIRD PASS: Look in the page contents
if ($PageInfo['Description'] == '') {
foreach ($Dom->find('p') as $element) {
Trace('Looking at p for description.');
if (strlen($element->plaintext) > 150) {
$PageInfo['Description'] = $element->plaintext;
break;
}
}
if (strlen($PageInfo['Description']) > 400) {
$PageInfo['Description'] = SliceParagraph($PageInfo['Description'], 400);
}
}
// Final: Still nothing? remove limitations
if ($PageInfo['Description'] == '') {
foreach ($Dom->find('p') as $element) {
Trace('Looking at p for description (no restrictions)');
if (trim($element->plaintext) != '') {
$PageInfo['Description'] = $element->plaintext;
break;
}
}
}
// Page Images
if (count($PageInfo['Images']) == 0) {
$Images = domGetImages($Dom, $url);
$PageInfo['Images'] = array_values($Images);
}
$PageInfo['Title'] = htmlEntityDecode($PageInfo['Title']);
$PageInfo['Description'] = htmlEntityDecode($PageInfo['Description']);
} catch (Exception $ex) {
$PageInfo['Exception'] = $ex->getMessage();
}
return $PageInfo;
}
/**
* Examine a page at {@link $Url} for title, description & images.
*
* Be sure to check the resultant array for any Exceptions that occurred while retrieving the page.
*
* @param string $url The url to examine.
* @param integer $timeout How long to allow for this request.
* Default Garden.SocketTimeout or 1, 0 to never timeout. Default is 0.
* @param bool $sendCookies Whether or not to send browser cookies with the request.
* @return array Returns an array containing Url, Title, Description, Images (array) and Exception
* (if there were problems retrieving the page).
*/
function fetchPageInfo($url, $timeout = 3, $sendCookies = false)
{
$PageInfo = array('Url' => $url, 'Title' => '', 'Description' => '', 'Images' => array(), 'Exception' => false);
try {
// Make sure the URL is valid.
$urlParts = parse_url($url);
if ($urlParts === false || !in_array(val('scheme', $urlParts), array('http', 'https'))) {
throw new Exception('Invalid URL.', 400);
}
$Request = new ProxyRequest();
$PageHtml = $Request->Request(array('URL' => $url, 'Timeout' => $timeout, 'Cookies' => $sendCookies, 'Redirects' => true));
if (!$Request->status()) {
throw new Exception('Couldn\'t connect to host.', 400);
}
$Dom = pQuery::parseStr($PageHtml);
if (!$Dom) {
throw new Exception('Failed to load page for parsing.');
}
// FIRST PASS: Look for open graph title, desc, images
$PageInfo['Title'] = domGetContent($Dom, 'meta[property="og:title"]');
Trace('Getting og:description');
$PageInfo['Description'] = domGetContent($Dom, 'meta[property="og:description"]');
foreach ($Dom->query('meta[property="og:image"]') as $Image) {
if ($Image->attr('content')) {
$PageInfo['Images'][] = $Image->attr('content');
}
}
// SECOND PASS: Look in the page for title, desc, images
if ($PageInfo['Title'] == '') {
$PageInfo['Title'] = $Dom->query('title')->text();
}
if ($PageInfo['Description'] == '') {
Trace('Getting meta description');
$PageInfo['Description'] = domGetContent($Dom, 'meta[name="description"]');
}
// THIRD PASS: Look in the page contents
if ($PageInfo['Description'] == '') {
foreach ($Dom->query('p') as $element) {
Trace('Looking at p for description.');
if (strlen($element->plaintext) > 150) {
$PageInfo['Description'] = $element->text();
break;
}
}
if (strlen($PageInfo['Description']) > 400) {
$PageInfo['Description'] = SliceParagraph($PageInfo['Description'], 400);
}
}
// Final: Still nothing? remove limitations
if ($PageInfo['Description'] == '') {
foreach ($Dom->query('p') as $element) {
Trace('Looking at p for description (no restrictions)');
if (trim($element->text()) != '') {
$PageInfo['Description'] = $element->text();
break;
}
}
}
// Page Images
if (count($PageInfo['Images']) == 0) {
$Images = domGetImages($Dom, $url);
$PageInfo['Images'] = array_values($Images);
}
$PageInfo['Title'] = htmlEntityDecode($PageInfo['Title']);
$PageInfo['Description'] = htmlEntityDecode($PageInfo['Description']);
} catch (Exception $ex) {
$PageInfo['Exception'] = $ex->getMessage();
}
return $PageInfo;
}