This commit is contained in:
James
2020-02-20 17:06:44 +00:00
commit 6a986a62c0
6 changed files with 2905 additions and 0 deletions

33
src/Crawler.php Normal file
View File

@@ -0,0 +1,33 @@
<?php
namespace JHodges\Sitemap;
use GuzzleHttp\Exception\RequestException;
use GuzzleHttp\RequestOptions;
use GuzzleHttp\Psr7\Uri;
use Psr\Http\Message\ResponseInterface;
use Psr\Http\Message\UriInterface;
use Spatie\Crawler\Crawler as SpatieCrawler;
use Spatie\Crawler\CrawlUrl;
use Spatie\Crawler\CrawlInternalUrls;
class Crawler{
public function Crawl($url){
$observer=new CrawlObserver();
SpatieCrawler::create([
RequestOptions::ALLOW_REDIRECTS => [
'track_redirects' => true,
]
])
//->setMaximumDepth(1)
->setCrawlObserver($observer)
->setCrawlProfile(new CrawlInternalUrls($url))
//->addToCrawlQueue( CrawlUrl::create(new Uri('https://hudevad.com/en/')) )
->startCrawling($url)
;
return $observer->results;
}
}