private function parsePage($url)
{
$content = @UTIL_HttpResource::getContents($url);
$matches = array();
preg_match('/<\\s*meta\\s*[^\\>]*?http-equiv=[\'"]content-type[\'"][^\\>]*?\\s*>/i', $content, $matches);
$meta = empty($matches[0]) ? null : $matches[0];
preg_match('/content=[\'"][^\'"]*?charset=([\\w-]+)(:[^\\w-][^\'"])*?[\'"]/i', $meta, $matches);
$encoding = empty($matches[1]) ? 'UTF-8' : $matches[1];
preg_match('/<\\s*title\\s*>([\\s\\S]*?)<\\s*\\/\\s*title\\s*>/i', $content, $matches);
$title = empty($matches[1]) ? null : mb_convert_encoding($matches[1], 'UTF-8', $encoding);
$matches = array();
$meta = "";
preg_match('/<\\s*meta\\s*[^\\>]*?name=[\'"]description[\'"][^\\>]*?\\s*>/i', $content, $matches);
$meta = empty($matches[0]) ? null : $matches[0];
$matches = array();
preg_match('/content=[\'"](.*?)[\'"]/i', $meta, $matches);
$description = empty($matches[1]) ? null : mb_convert_encoding($matches[1], 'UTF-8', $encoding);
$matches = array();
preg_match_all('/<\\s*img\\s*.*?src=[\'"](.+?)[\'"].*?>/i', $content, $matches);
$images = array();
foreach ($matches[1] as $img) {
$urlInfo = parse_url($url);
$imgInfo = parse_url($img);
if (empty($imgInfo['host'])) {
$imgDir = dirname($imgInfo['path']);
$urlScheme = empty($urlInfo['scheme']) ? '' : $urlInfo['scheme'] . '://';
$urlAddr = $urlScheme . $urlInfo['host'];
if (strpos($imgDir, '/') === 0) {
$img = $urlAddr . $imgInfo['path'];
} elseif (!empty($urlInfo['path'])) {
$pp = pathinfo($urlInfo['path']);
$urlPath = $pp['dirname'] . (empty($pp['extension']) ? $pp['basename'] . '/' : '');
$img = $urlAddr . $urlPath . $imgInfo['path'];
} else {
$img = $urlAddr . '/' . $imgInfo['path'];
}
}
$images[] = $img;
}
$firstImg = reset($images);
$firstImg = $firstImg ? $firstImg : null;
return array('type' => 'link', 'description' => UTIL_HtmlTag::escapeHtml($description), 'title' => UTIL_HtmlTag::escapeHtml($title), 'thumbnail_url' => $firstImg, 'allImages' => $images);
}
