php – The translator script only translates one word



I found a translator script , it uses version v2 (exactly what is in the API specification)

It seems that everything in the script corresponds to the Bing API at the moment, but this script can only translate one word.

If you write more, it doesn't translate, I can't understand why – can anyone explain?

This script has a minimum of code if you compare it with what I posted above

Please help me figure it out!


$your_client_id = '0750facb-bb29-43b4-a629-42603c2efbea';
$your_secret = 'I3GDuMftmn4pTeUTiJgJp0VGNbEkfDbBgwFDC7eb6E0';

    class BingTranslation
        public $clientID;
        public $clientSecret;

        public function __construct($cid, $secret)
            $this->clientID = $cid;
            $this->clientSecret = $secret;

        public function get_access_token()
            //if access token is not expired and is stored in COOKIE
                return $_COOKIE['bing_access_token'];

            // Get a 10-minute access token for Microsoft Translator API.
            $url = '';
            $postParams = 'grant_type=client_credentials&client_id='.urlencode($this->clientID).

            $ch = curl_init();
            curl_setopt($ch, CURLOPT_URL, $url); 
            curl_setopt($ch, CURLOPT_POSTFIELDS, $postParams);
            curl_setopt($ch, CURLOPT_RETURNTRANSFER, TRUE);  
            $rsp = curl_exec($ch); 
            $rsp = json_decode($rsp);
            $access_token = $rsp->access_token;

            setcookie('bing_access_token', $access_token, $rsp->expires_in);

            return $access_token;

        public function translate($word, $from, $to)
            $access_token = $this->get_access_token();
            $url = ''.$word.'&from='.$from.'&to='.$to;

            $ch = curl_init();
            curl_setopt($ch, CURLOPT_URL, $url); 
            curl_setopt($ch, CURLOPT_HTTPHEADER, array('Authorization:bearer '.$access_token));
            curl_setopt($ch, CURLOPT_RETURNTRANSFER, TRUE);  
            $rsp = curl_exec($ch); 

            preg_match_all('/<string (.*?)>(.*?)<\/string>/s', $rsp, $matches);
            return $matches[2][0];

    //usage example
    $bt = new BingTranslation($your_client_id, $your_secret);
    echo $bt->translate('home', 'en', 'ru');


As long as you send one word in Latin, your HTTP request is correct. But this is nothing more than an accident.

Send a request with Cyrillic content – this will already be an RFC violation, but not fatal. In the harsh outside world, servers have to forgive this. And here is a space – it completely breaks the structure of the request and the request is obtained:

GET /V2/Http.svc/Translate?text=привет мир&from=ru&to=en HTTP/1.1

Of course, they can no longer answer anything other than HTTP 400. It is very sad that you do not check curl_errno (although not curl_errno for this problem, but will help with other network problems) or curl_getinfo .

A valid URL must be URL-encoded.

$url = ''.urlencode($word).'&from='.urlencode($from).'&to='.urlencode($to);

Or not to pervert, but to form the correct query string using a native function:

$url = ''.http_build_query([
            'text' => $word,
            'from' => $from,
            'to' => $to,
Scroll to Top