<?php

namespace App\Services\Scrapers;

use Goutte\Client;
use Illuminate\Support\Carbon;

class JazeeraScraper
{
    protected $client;

    protected $title;
    protected $content;
    protected $topics = [];
    protected $author = "unknown";

    public function __construct()
    {
        $this->client = new Client;
    }

    public function extract($url, $date)
    {

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

        $crawler->filter('.single-body--content p')->each(function ($node) {
            $this->content[] = $node->text();
        });

        $crawler->filter('.entry-tags a')->each(function ($node) {
           if(!preg_match('/[^A-Za-z0-9-]/', basename($node->attr('href'))))
           {
            $this->topics[] = [
                "name" => $node->text(),
                "slug" => basename($node->attr('href'))
            ];
           }
            
        });


        if ($crawler->filter('a[class*="entry-author__name"]')->count() == 1) {
            $this->author = $crawler->filter('a[class*="entry-author__name"]')->first()->text();
        }
        //Remove all the alphabets from string
        //preg_replace("/[a-zA-Z]/", "",$string);
        return [
            'source'    => 'Jazeera',
            'title'      => $crawler->filter('h1')->first()->text(),
            'og_title'   => $crawler->filter('meta[property*="og:title"]')->first()->attr('content'),
            'image'      => $crawler->filter(".single-entry-thumb img")->first()->attr('data-src'),
            'content'    => $this->content,
            'url'        => strtok($url,'?'),
            'date'       => Carbon::parse($date)->format("Y-m-d H:i:s"),
            'guid'       =>basename(strtok($url,'?')),
            'author'     => $this->author,
            'topics'       => $this->topics ? : [
                [
                    "name" => "ވަކި މަޢުލޫއެއް ނޭންގެ",
                    "slug" => "no-specific-topic"
                ]
            ]
        ];
    }
}