Twitter API with CURL

Question

OBS: THE COMPLETE CODE IS HERE, but I made this adjustment to receive the cookies

I am trying to read the cookie files to make the necessary requests to stop get the result I hope, see my code:

$tokens = ['_twitter_sess', 'ct0', 'auth_token'];

        foreach (glob($this->_cookieFile) as $file) {
            $pointer = fopen($file, 'r');
        while (!feof($pointer)) {
        $line = fgets($pointer);

        foreach ($tokens as $token) {
          if (stripos($line, $token) !== false) {
            var_dump($line);
            curl_setopt_array($request, [
                                CURLOPT_COOKIEFILE  => $line,
                                CURLOPT_HTTPHEADER          => [
                          'origin: https://twitter.com',
                          'authorization: Bearer ' . $bearer,
                          'x-csrf-token: ' . $line,
                          'referer: https://twitter.com/',
                          'x-twitter-auth-type: OAuth2Session',
                          'x-twitter-active-user: yes',
                        ],
                         CURLOPT_POSTFIELDS => http_build_query([
                              'challenges_passed'   => false,
                              'handles_challenges'  => 1,
                              'include_blocked_by'  => true,
                              'include_blocking'    => true,
                              'include_can_dm'      => true,
                              'include_followed_by' => true,
                              'include_mute_edge'   => true,
                              'skip_status'         => true,
                              'user_id'             => Session::get('username'),
                                ], '', '&', PHP_QUERY_RFC3986),
                            ]
                        );
            break;
          }
        }
        }
        }

But when giving a var_dump in $line, I receive as return:

I need to read each line in one array associativo or read all cookies at once.

So how do I return from mine cURL get back to me:

'{"errors":[{"code":220,"message":"Your credentials do not allow access to this resource."}]}' (length=92)

Answer 1

If you really want an array, which by me doesn’t make any sense and seems to me to be one that this issue is an "XY problem". But anyway, just "explode" for each \t.

So, supposing it’s in string:

$cookie = '# Netscape HTTP Cookie File

.example.org    TRUE    /   FALSE   1507286655  remember_me true
.example.org    TRUE    /   FALSE   1507286655  APISID  DijdSAOAjgwijnhFMndsjiejfdSDNSgfsikasASIfgijsowITITeoknsd
.example.org    TRUE    /   FALSE   1507286655  static_files    iy1aBf1JhQR';


$array = [];

foreach(array_slice(explode("\n", $cookie), 2) as $linha){

    $colunas = explode("\t", $linha);
    $array[$colunas[5]] = $colunas[6];

}

Upshot (var_dump($array)):

array(3) { ["remember_me"]=> string(5) "true " ["APISID"]=> string(58) "DijdSAOAjgwijnhFMndsjiejfdSDNSgfsikasASIfgijsowITITeoknsd " ["static_files"]=> string(11) "iy1aBf1JhQR" }

---

The idea is simple, ignore the first two lines (array_slice(..., 2)) then for each line break by tab (\n) and then get the name and value information.

---

'Cause I think that’s totally wrong:

1. You have native features for this on Curl, all features that use Curl’s cookie management engine handle this Netscape format well.
2. You have how to use the CURLOPT_HEADERFUNCTION to filter cookies, if you are making a request to save such cookies.
3. You can also use the CURLINFO_COOKIELIST to obtain the same information, filter and save it, leaving only the data that matters.
4. You’re storing unnecessary data, worsening performance by having to filter it through every run, with no benefit.