<?php
namespace App\Services\Scrapers;

use Goutte\Client;

class ThePressScraper
{
    protected $client;

    protected $content;
    protected $author;
    protected $topics = [];
    
    public function __construct()
    {
        $this->client = new Client;
    }

    public function extract($url)
    {

        $crawler = $this->client->request('GET', $url);

        $title = $crawler->filter('h1')->first()->text();


        $image = $crawler->filter('.article-image img')->first()->attr('src');

        $crawler->filter('article p')->each(function ($node) {
            $this->content[] = preg_replace("/[a-zA-Z]/","",$node->text());
        });

        $crawler->filter('.article-tags a')->each(function ($node) {
           
            $this->topics[] = [
                "name" => $node->text(),
                "slug" => str_replace("https://thepress.mv/", "", $node->attr('href'))
            ];
        });

        if($crawler->filter(".author-details strong")->count() == 1)
        {
            $this->author = $crawler->filter('.author-details strong')->first()->text();
        }

        //Remove all the alphabets from string
        //preg_replace("/[a-zA-Z]/", "",$string);
       return [
            'source'    => 'The Press',
            'title'      => $title,
            'og_title'   => $crawler->filter('meta[property*="og:title"]')->first()->attr('content'),
            'image'      => $image,
            'content'    => $this->content,
            'url'        => $url,
            'date'       => $crawler->filter('.article-header .datetime')->first()->text(),
            'guid'       => str_replace("https://thepress.mv/","",$url),
            'author'     => $this->author,
            'topics'       => $this->topics
        ];
    }
}