test crawler

kaigokensaku.mhlw.go.jp.php

<?php

namespace App\Console\Commands;

use Goutte\Client;
use Illuminate\Console\Command;

class Crawler extends Command
{
    /**
     * The name and signature of the console command.
     *
     * @var string
     */
    protected $signature = 'crawler:run';

    /**
     * The console command description.
     *
     * @var string
     */
    protected $description = 'Command description';

    /**
     * Create a new command instance.
     *
     * @return void
     */
    public function __construct()
    {
        parent::__construct();
    }

    /**
     * Execute the console command.
     *
     * @return int
     */
    public function handle()
    {
        $url = 'https://www.kaigokensaku.mhlw.go.jp/13/index.php?action_kouhyou_pref_seikatu_search_list_list=true&PrefCd=13&SearchType=life';

        $client = new Client();
        $params = ['method' => 'pager', 'action_kouhyou_pref_seikatu_search_list_result' => 'true', 'PrefCd' => '13', 'OriPrefCd' => '13', 'SearchType' => 'life', 'p_offset' => '10', 'p_count' => '2', 'p_sort' => '0', 'p_order' => '0', 'p_total' => '75', 'p_sort_name' => '26', 'p_order_name' => '1'];
        $crawler = $client->request('POST', $url, $params);
        //get body tag
        $body = $crawler->filter('body')->first();
        if ($body) {
        //get first script yag
            $script = $body->filter('script')->first();
            if ($script) {
            // decode html
                $string = html_entity_decode($script->text());
                $startStr = 'var aryJigyosyo = [];';
                $endStr = '// 検索結果';
                //position of start string
                $startPos = strpos($string, $startStr);
                //start substr index
                $startIndex = $startPos + strlen($startStr) + 1;
                //end post
                $endPos = strpos($string, $endStr);
                // substr (data string lenght)
                $length = $endPos - $startIndex - 1;
                $str = trim(substr($string, $startIndex, $length));
                
                // replace to make json string
                $str = str_replace('aryJigyosyo.push(', '', $str);
                $str = str_replace(');', ',', $str);
                $str = str_replace('\'', '"', $str);
                $str = '['.trim($str, ',').']';
                // decode json string to array
                try {
                    $arr = json_decode($str, true);
                    dd($arr);
                } catch (\Exception $e) {
                    dd($e);
                }
            }
        }

        return 0;
    }
}