namespace phinde\LinkExtractor;
use phinde\LinkInfo;
+use phinde\Helper;
class Html
{
$dx = new \DOMXPath($doc);
- $meta = $dx->evaluate('/html/head/meta[@name="robots" and @value]')
+ $xbase = $dx->evaluate('/html/head/base[@href]')->item(0);
+ if ($xbase) {
+ $base = $base->resolve(
+ $xbase->attributes->getNamedItem('href')->textContent
+ );
+ }
+
+ $meta = $dx->evaluate('/html/head/meta[@name="robots" and @content]')
->item(0);
if ($meta) {
- $robots = $meta->attributes->getNamedItem('value')->textContent;
+ $robots = $meta->attributes->getNamedItem('content')->textContent;
foreach (explode(',', $robots) as $value) {
if (trim($value) == 'nofollow') {
//we shall not follow the links
$links = $dx->evaluate('//a');
//FIXME: link rel, img, video
- $alreadySeen = array();
+ $alreadySeen = array($url => true);
foreach ($links as $link) {
- $linkTitle = $link->textContent;
+ $linkTitle = Helper::sanitizeTitle($link->textContent);
$href = '';
foreach ($link->attributes as $attribute) {
if ($attribute->name == 'href') {