I'm using the following php script to get search results from Google.
include("simple_html_dom.php");
include("random-user-agent.php");
$query = 'facebook';
$curl = curl_init();
curl_setopt($curl, CURLOPT_URL, 'http://www.google.com/search?q='.$query.'');
#curl_setopt($curl, CURLOPT_RETURNTRANSFER, 1);
curl_setopt($curl, CURLOPT_FOLLOWLOCATION, TRUE);
curl_setopt($curl, CURLOPT_USERAGENT,random_user_agent());
$str = curl_exec($curl);
curl_close($curl);
$html= str_get_html($str);
$i = 0;
foreach($html->find('li[class=g]') as $element) {
foreach($element->find('h3') as $item)
{
$title[$i] = ''.$item->plaintext.'' ;
}
$i++;
}
print_r($title);
When this script runs in a cronjob (with 5 sec sleep) I receive a warning from Google and have to fill in a captcha (obvious). I always thought that using curl and a random user agent can avoid this. What is the correct solution?
A better way to avoid captcha is to set a randomized sleep between 3-6 seconds per request.
Best solution is to use proxies.