9

I am trying to login to MCA portal ( POST URL : http://www.mca.gov.in/mcafoportal/loginValidateUser.do )

I tried logging in with POSTMAN app on Google Chrome which works fine. However, it doesnt work either in PHP/Python. I am not able to login through PHP/Python

Here is the PHP Code :

$url="http://www.mca.gov.in/mcafoportal/loginValidateUser.do"; 

$post_fields = array();

$post_fields['userNamedenc']='hGJfsdnk`1t';
$post_fields['passwordenc']='675894242fa9c66939d9fcf4d5c39d1830f4ddb9';
$post_fields['accessCode'] = ""

$str = call_post_mca($url, $post_fields);
$str = str_replace(" ","",$str);   

$dom = new DOMDocument();
$dom->loadHTML($str);
$xpath = new DOMXPath($dom);

$input_id =  '//input[@id="login_accessCode"]/@value';
$input_val = $xpath->query($input_id)->item(0);
$input_val1 = $input_val->nodeValue;

$url="http://www.mca.gov.in/mcafoportal/loginValidateUser.do"; 

$post_fields['userNamedenc']='hGJfsdnk`1t';
$post_fields['passwordenc']='675894242fa9c66939d9fcf4d5c39d1830f4ddb9';
$post_fields['accessCode'] = $input_val1;  //New Accesscode 

function  call_post_mca($url, $params)
{   
    #$user_agent = getRandomUserAgent();
    $user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36";
    $str = "";
    foreach($params as $key=>$value)
    {
        $str = $str . "$key=$value" . "&";
    }
    $postData = rtrim($str, "&");

    $ch = curl_init();  

    curl_setopt($ch,CURLOPT_URL,$url);
    curl_setopt($ch,CURLOPT_RETURNTRANSFER,true);
    curl_setopt($ch,CURLOPT_HEADER, false); 
    #curl_setopt($ch, CURLOPT_CAINFO, DOC_ROOT . '/includes/cacert.pem');

    curl_setopt($ch, CURLOPT_POST, 1);
    curl_setopt($ch, CURLOPT_POSTFIELDS, $postData);   


    curl_setopt($ch,CURLOPT_USERAGENT, $user_agent);
    curl_setopt ($ch, CURLOPT_FOLLOWLOCATION, 1);
    curl_setopt ($ch, CURLOPT_REFERER, $url); 

    $cookie= DOC_ROOT . "/cookie.txt";
    curl_setopt ($ch, CURLOPT_COOKIEJAR, $cookie); 
    curl_setopt ($ch, CURLOPT_COOKIEFILE, $cookie); 

    $output=curl_exec($ch);

    curl_close($ch);
    return $output;

}

enter image description here

Any idea what is missing ?

Kiran
  • 6,858
  • 30
  • 95
  • 156
  • 2
    What error do you get? nor working is very generic – Tarun Lalwani Sep 23 '17 at 08:14
  • Yes @Tarun, I should be able to login through CURL. Its happening with POSTMAN, but with PHP/CURL, it is failing, it is still showing up the login page – Kiran Sep 23 '17 at 08:15
  • 1
    yes @Kiran, but let me ask you again : what error are you getting ? – YvesLeBorg Sep 23 '17 at 08:17
  • There is no error. It just doesnt login. It shows the login page again with no error. Since the username/pwd is correct, it should serve the dashboard page ( This is happening with postman) – Kiran Sep 23 '17 at 08:20
  • You're not using cURL to send POST correctly. Look at https://stackoverflow.com/questions/2138527/php-curl-http-post-sample-code – apokryfos Sep 23 '17 at 08:20
  • Also why do you keep saying PHP/Python, there's only PHP here – apokryfos Sep 23 '17 at 08:21
  • @apokryfos I tried with Python as well, and I am getting the same issue. It doesnt login – Kiran Sep 23 '17 at 08:22
  • If you want to log in in some page, you have to store the session ID within a cookie or someting like it. If not, your redirect to the next page will not have the correct data. Then you will get logged out in most cases. Check this out: https://stackoverflow.com/questions/10307744/how-to-login-in-with-curl-and-ssl-and-cookies – S.Gartmeier Sep 23 '17 at 09:33
  • 2
    This needs cookie handling, as well as you fetch the `login.do` first to get the new `accessCode` and then you need to submit the url, for which your code is still wrong – Tarun Lalwani Sep 25 '17 at 04:36
  • @TarunLalwani Yes, this is exactly how I am doing. I first get the accesscode from login.do and use this accesscode along with usernameenc and passwordenc to login to the portal. Updated the code I am using, can you please have a look – Kiran Sep 25 '17 at 05:40
  • @Kiran whats the **accessCode** here – RamaKrishna Sep 25 '17 at 12:06
  • @RamaKrishna, Accesscode is the value of the hidden variable loginaccesscode when you open the login page login.do in the MCA portal. Will update the code on how to obtain the accesscode – Kiran Sep 25 '17 at 12:08
  • there is a captcha while logging in what to do about it ? – RamaKrishna Sep 25 '17 at 12:54
  • @RamaKrishna, You dont have to pass the captcha, please check the attached screenshot, I am not passing this parameter, still I am able to login – Kiran Sep 25 '17 at 13:12
  • @Kiran I tried with postman with credentials you provided but unfortunately, it is showing me login page only. – RamaKrishna Sep 26 '17 at 06:34
  • Since its working on postman, can you export the request as a curl and paste it here. It will help anyone who wants to start. We can then see what worked and what didn't. – TheChetan Sep 29 '17 at 06:52
  • @Kiran, I may have a possible solution in PHP or python. Just working it out. Do wait for it :-) before accepting any other answer – Tarun Lalwani Oct 01 '17 at 14:07
  • @Kiran, just posted an answer. Have a look at it – Tarun Lalwani Oct 02 '17 at 06:52

6 Answers6

5

The site does a redirect, so you need to add

CURLOPT_FOLLOWLOCATION => 1

to your options array. When in doubt with cURL, try

$status = curl_getinfo($curl);
echo json_encode($status, JSON_PRETTY_PRINT);

giving :

{
"url": "http:\/\/www.mca.gov.in\/mcafoportal\/loginValidateUser.do?userNamedenc=hGJfsdnk%601t&passwordenc=675894242fa9c66939d9fcf4d5c39d1830f4ddb9&accessCode=-825374456",
"content_type": "text\/plain",
"http_code": 302,
"header_size": 1560,
"request_size": 245,
"filetime": -1,
"ssl_verify_result": 0,
"redirect_count": 0,
"total_time": 1.298891,
"namelookup_time": 0.526375,
"connect_time": 0.999786,
"pretransfer_time": 0.999844,
"size_upload": 0,
"size_download": 0,
"speed_download": 0,
"speed_upload": 0,
"download_content_length": 0,
"upload_content_length": -1,
"starttransfer_time": 1.298875,
"redirect_time": 0,
"redirect_url": "http:\/\/www.mca.gov.in\/mcafoportal\/login.do",
"primary_ip": "115.114.108.120",
"certinfo": [],
"primary_port": 80,
"local_ip": "192.168.1.54",
"local_port": 62524
}

As you can see, you got a 302 redirect status, but a redirect_count was 0. After adding the option, i get:

{
"url": "http:\/\/www.mca.gov.in\/mcafoportal\/login.do",
"content_type": "text\/html;charset=ISO-8859-1",
"http_code": 200,
"header_size": 3131,
"request_size": 376,
"filetime": -1,
"ssl_verify_result": 0,
"redirect_count": 1,
"total_time": 2.383609,
"namelookup_time": 1.7e-5,
"connect_time": 1.7e-5,
"pretransfer_time": 4.4e-5,
"size_upload": 0,
"size_download": 42380,
"speed_download": 17779,
"speed_upload": 0,
"download_content_length": 42380,
"upload_content_length": -1,
"starttransfer_time": 0.30734,
"redirect_time": 0.915858,
"redirect_url": "",
"primary_ip": "14.140.191.120",
"certinfo": [],
"primary_port": 80,
"local_ip": "192.168.1.54",
"local_port": 62642
}

EDIT url encode the request parameters , and follow redirects

 $str = urlencode("userNamedenc=hGJfsdnk%601t&passwordenc=675894242fa9c66939d9fcf4d5c39d1830f4ddb9&accessCode=-825374456");
curl_setopt_array(
    $curl , array (
    CURLOPT_URL            => "http://www.mca.gov.in/mcafoportal/loginValidateUser.do" , // <- removed parameters here
    CURLOPT_RETURNTRANSFER => true ,
    CURLOPT_ENCODING       => "" ,
    CURLOPT_FOLLOWLOCATION => 1 ,
    CURLOPT_MAXREDIRS      => 10 ,
    CURLOPT_TIMEOUT        => 30 ,
    CURLOPT_HTTP_VERSION   => CURL_HTTP_VERSION_1_1 ,
    CURLOPT_CUSTOMREQUEST  => "POST" ,
    CURLOPT_POSTFIELDS     => $str,       // <- added this here
    CURLOPT_HTTPHEADER     => array (
        "cache-control: no-cache"
    ) ,
)
);
YvesLeBorg
  • 8,943
  • 8
  • 33
  • 46
  • I tried this option in the curl options, however it still throws up the login page. – Kiran Sep 23 '17 at 08:43
  • If it helps, the access code keeps changing so I need to fetch this first through first request and send the second request with the accesscode – Kiran Sep 23 '17 at 08:44
  • unfortunately the login page is bugged, and gives `HTTP 200 OK` even when login failed. it should have responded `HTTP 403 Forbidden`, someone should contact them and help them fix it. here's a more reliable way to detect login errors: `$domd = @DOMDocument::loadHTML ( $html ); $xp = new DOMXPath ( $domd ); $loginErrors = $xp->query ( '//ul[@class="errorMessage"]' ); if ($loginErrors->length > 0) { echo 'encountered following error(s) logging in: '; foreach ( $loginErrors as $err ) { echo $err->textContent, PHP_EOL; } die (); } else { echo "logged in successfully!"; }`. TL;DR: 200 OK != succes – hanshenrik Oct 01 '17 at 20:50
3

The simplest thing you can do, since you already have it working in POSTMAN, is to render out the PHP code in POSTMAN. Here is link about to get PHP code from POSTMAN. Then you can compare the POSTMAN example to your code.

<?php

$curl = curl_init();

curl_setopt_array($curl, array(
  CURLOPT_URL => "http://www.mca.gov.in/mcafoportal/loginValidateUser.do?userNamedenc=hGJfsdnk%601t&passwordenc=675894242fa9c66939d9fcf4d5c39d1830f4ddb9&accessCode=",
  CURLOPT_RETURNTRANSFER => true,
  CURLOPT_ENCODING => "",
  CURLOPT_MAXREDIRS => 10,
  CURLOPT_TIMEOUT => 30,
  CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
  CURLOPT_CUSTOMREQUEST => "POST",
  CURLOPT_HTTPHEADER => array(
    "cache-control: no-cache",
    "postman-token: b54abdc0-17be-f38f-9aba-dbf8f007de99"
  ),
));

$response = curl_exec($curl);
$err = curl_error($curl);

curl_close($curl);

if ($err) {
  echo "cURL Error #:" . $err;
} else {
  echo $response;
}

What is immediately popping out to me is this 'hGJfsdnk`1t'. The backward quote can be an escape character '`'. This could very well be throwing an error where error handling redirects back to the login page. POSTMAN likely has something built in to render out the escape character to 'hGJfsdnk%601t'. Thus, this works in POSTMAN, but not in your code.

Here is the status of this request:

{
"url": "http:\/\/www.mca.gov.in\/mcafoportal\/login.do",
"content_type": "text\/html;charset=ISO-8859-1",
"http_code": 200,
"header_size": 3020,
"request_size": 821,
"filetime": -1,
"ssl_verify_result": 0,
"redirect_count": 1,
"total_time": 2.920125,
"namelookup_time": 8.2e-5,
"connect_time": 8.7e-5,
"pretransfer_time": 0.000181,
"size_upload": 0,
"size_download": 42381,
"speed_download": 14513,
"speed_upload": 0,
"download_content_length": -1,
"upload_content_length": -1,
"starttransfer_time": 0.320995,
"redirect_time": 2.084554,
"redirect_url": "",
"primary_ip": "115.114.108.120",
"certinfo": [],
"primary_port": 80,
"local_ip": "192.168.1.3",
"local_port": 45086
}

Here is shows the successful login.

Jed Lynch
  • 1,457
  • 14
  • 9
  • nope, unfortunately the login page is bugged, and gives `HTTP 200 OK` even when login failed. it *should* have responded `HTTP 403 Forbidden`, someone should contact them and help them fix it. here's a more reliable way to detect login errors: `$domd = @DOMDocument::loadHTML ( $html ); $xp = new DOMXPath ( $domd ); $loginErrors = $xp->query ( '//ul[@class="errorMessage"]' ); if ($loginErrors->length > 0) { echo 'encountered following error(s) logging in: '; foreach ( $loginErrors as $err ) { echo $err->textContent, PHP_EOL; } die (); } else { echo "logged in successfully!"; }` – hanshenrik Oct 01 '17 at 13:46
  • It could be that the home page is bugged, but this does return with "Welcome Guest" at the top, just as POSTMAN did. – Jed Lynch Oct 01 '17 at 17:51
  • and when you log in with the credentials OP provided, it says `Welcome Kiran` (not Guest). try the code in my answer if you wanna see it for yourself, `php foo.php | tee foo.html`, solve the captcha, and open foo.html in your browser – hanshenrik Oct 01 '17 at 20:09
  • here's a screenshot from when i run it, save the html, and open it in the browser: https://image.prntscr.com/image/_AsB_0J6TLOFSZuvQdjyNg.png – hanshenrik Oct 01 '17 at 20:26
3

Successful login through code

This is honestly one of weird sites I have seen in a long time. First thing was to know how it works. So I decided to use chrome and see what happens when we login with wrong data

Site

Observations:

  • Blanks username and passwords fields
  • Generates SHA1 hashes of username and password fields and sets then in userNamedenc and respectively
  • We can override username and password directly in JavaScript and login to your account just by overriding the details from console.
  • There are lot of different request which generates cookies but none of them look any useful

So the approach to solve the issue was to follow below steps

  • Get the login url login.do
  • Fetch the form details from the response for the access code
  • Submit the form to loginValidateUser.do

The form sends below parameters

Form Submit

Now one interesting part of the same is below post data

displayCaptcha:true
userEnteredCaptcha:strrty

If we override the displayCaptcha to false then captcha is no more needed. So a wonderful bypass

displayCaptcha: false

Next was to code all of the above in PHP, but the site seemed so weird that many of the attempts failed. So finally I realized that we need to take it a bit closer to the browser login and also I felt delays between calls are needed

<?php

    require_once("curl.php");

    $curl = new CURL();
    $default_headers = Array(
        "Accept" => "text/html,application/xhtml+xml,application/xml;q=0.9,image/webp,image/apng,*/*;q=0.8",
        "Accept-Encoding" => "deflate",
        "Accept-Language" => "en-US,en;q=0.8",
        "Cache-Control" => "no-cache",
        "Connection" => "keep-alive",
        "DNT" => "1",
        "Pragma" => "no-cache",
        "Referer" => "http://www.mca.gov.in/mcafoportal/login.do",
        "Upgrade-Insecure-Requests" => "1"
    );

    // Get the login page 
    $curl
        ->followlocation(0)
        ->cookieejar("")
        ->verbose(1)
        ->get("http://www.mca.gov.in/mcafoportal/login.do")
        ->header($default_headers)
        ->useragent("Mozilla/5.0 (Macintosh; Intel Mac OS X 10_12_6) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/61.0.3163.100 Safari/537.36")
        ->execute();


    // Save the postfileds and access code as we would need them later for the POST field
    $post = $curl->loadInputFieldsFromResponse()
        ->updatePostParameter(array(
            "displayCaptcha" => "false",
            "userNamedenc" => "hGJfsdnk`1t",
            "passwordenc" => "675894242fa9c66939d9fcf4d5c39d1830f4ddb9",
            "userName" => "",
            "Cert" => ""))
        ->referrer("http://www.mca.gov.in/mcafoportal/login.do")
        ->removePostParameters(
            Array("dscBasedLoginFlag", "maxresults", "fe", "query", "SelectCert", "newUserRegistration")
        );

    $postfields = $curl->getPostFields();

    var_dump($postfields);


    // Access some dummy URLs to make it look like browser
    $curl
        ->get("http://www.mca.gov.in/mcafoportal/js/global.js")->header($default_headers)->execute()->sleep(2)
        ->get("http://www.mca.gov.in/mcafoportal/js/loginValidations.js")->header($default_headers)->execute()->sleep(2)
        ->get("http://www.mca.gov.in/mcafoportal/css/layout.css")->header($default_headers)->execute()->sleep(2)
        ->get("http://www.mca.gov.in/mcafoportal/img/bullet.png")->header($default_headers)->execute()->sleep(2)
        ->get("http://www.mca.gov.in/mcafoportal/getCapchaImage.do")->header($default_headers)->execute()->sleep(2);


    // POST to the login form the postfields saved earlier
    $curl
        ->sleep(20)
        ->header($default_headers)
        ->postfield($postfields)
        ->referrer("http://www.mca.gov.in/mcafoportal/login.do")
        ->post("http://www.mca.gov.in/mcafoportal/loginValidateUser.do")
        ->execute(false)
        ->sleep(3)
        ->get("http://www.mca.gov.in/mcafoportal/login.do")
        ->header($default_headers)
        ->execute(true);

    // Get the response from last GET of login.do
    $curl->getResponseText($output);

    //Check if user name is present in the output or not
    if (stripos($output, "Kiran") > 0) {
        echo "Hurray!!!! Login succeeded";
    } else {
        echo "Login failed please retry after sometime";
    }

After running the code it works few times and few times it doesn't. My observations

  • Only one login is allowed at a time. So not sure if others were using the login when I was testing
  • Without delays it would fail most of the time
  • There is no obvious reason when it fails to login except the site doing something on server side to block the request

Successful login

The reusable curl.php I created and used for chaining methods is below

<?php

class CURL
{
    protected $ch;
    protected $postfields;

    public function getPostFields() {
        return $this->postfields;
    }

    public function newpost()
    {
        $this->postfields = array();
        return $this;
    }

    public function addPostFields($key, $value)
    {
        $this->postfields[$key]=$value;
        return $this;
    }

    public function __construct()
    {
        $ch       = curl_init();
        $this->ch = $ch;
        $this->get()->followlocation()->retuntransfer(); //->connectiontimeout(20)->timeout(10);
    }

    function url($url)
    {
        curl_setopt($this->ch, CURLOPT_URL, $url);
        return $this;
    }

    function verbose($value = true)
    {
        curl_setopt($this->ch, CURLOPT_VERBOSE, $value);
        return $this;
    }

    function post($url='')
    {
        if ($url !== '')
            $this->url($url);
        curl_setopt($this->ch, CURLOPT_POST, count($this->postfields));
        curl_setopt($this->ch, CURLOPT_POSTFIELDS, http_build_query($this->postfields));
        return $this;
    }

    function postfield($fields)
    {
        if (is_array($fields)){
            $this->postfields = $fields;
        }
        return $this;
    }

    function close()
    {
        curl_close($this->ch);
        return $this;
    }

    function cookieejar($cjar)
    {
        curl_setopt($this->ch, CURLOPT_COOKIEJAR, $cjar);
        return $this;
    }

    function cookieefile($cfile)
    {
        curl_setopt($this->ch, CURLOPT_COOKIEFILE, $cfile);
        return $this;
    }

    function followlocation($follow = 1)
    {
        curl_setopt($this->ch, CURLOPT_FOLLOWLOCATION, $follow);
        return $this;
    }

    function loadInputFieldsFromResponse($response ='')
    {
        if ($response)
            $doc = $response;
        else
            $doc = $this->lastCurlRes;


        /* @var $doc DOMDocument */
        //simplexml_load_string($data)
        $this->getResponseDoc($doc);
        $this->postfields = array();

        foreach ($doc->getElementsByTagName('input') as $elem) {
            /* @var $elem DomNode */
            $name = $elem->getAttribute('name');
//            if (!$name)
//                $name = $elem->getAttribute('id');
            if ($name)
                $this->postfields[$name] = $elem->getAttribute("value");

        }

        return $this;
    }

    function retuntransfer($transfer=1)
    {
        curl_setopt($this->ch, CURLOPT_RETURNTRANSFER, $transfer);
        return $this;
    }

    function connectiontimeout($connectiontimeout)
    {
        curl_setopt($this->ch, CURLOPT_CONNECTTIMEOUT, $connectiontimeout);
        return $this;
    }

    function timeout($timeout)
    {
        curl_setopt($this->ch, CURLOPT_TIMEOUT, $timeout);
        return $this;
    }
    function useragent($useragent)
    {
        curl_setopt($this->ch, CURLOPT_USERAGENT, $useragent);
        return $this;
    }

    function referrer($referrer)
    {
        curl_setopt($this->ch, CURLOPT_REFERER, $referrer);
        return $this;
    }

    function getCURL()
    {
        return $this->ch;
    }

    protected $lastCurlRes;
    protected $lastCurlResInfo;

    function get($url = '')
    {
        if ($url !== '')
            $this->url($url);
        curl_setopt($this->ch, CURLOPT_POST, 0);
        curl_setopt($this->ch, CURLOPT_HTTPGET, true);
        return $this;
    }

    function sleep($seconds){
        sleep($seconds);
        return $this;
    }

    function execute($output=false)
    {
        $this->lastCurlRes = curl_exec($this->ch);

        if ($output == true)
        {
            echo "Response is \n " . $this->lastCurlRes;
            file_put_contents("out.html", $this->lastCurlRes);
        }
        $this->lastCurlResInfo = curl_getinfo($this->ch);
        $this->postfields = array();
        return $this;
    }

    function header($headers)
    {
        //curl_setopt($this->ch, CURLOPT_HEADER, true);
        curl_setopt($this->ch, CURLOPT_HTTPHEADER, $headers);
        return $this;
    }
    function getResponseText(&$text){
        $text = $this->lastCurlRes;
        return $this;
    }


    /*
     *
    * @param DOMDocument $doc
    *
    *
    */
    function getResponseDoc(&$doc){
        $doc = new DOMDocument();
        libxml_use_internal_errors(false);
        libxml_disable_entity_loader();
        @$doc->loadHTML($this->lastCurlRes);
        return $this;
    }

    function removePostParameters($keys) {
        if (!is_array($keys))
            $keys = Array($keys);

        foreach ($keys as $key){
            if (array_key_exists($key, $this->postfields))
                unset($this->postfields[$key]);
        }

        return $this;
    }

    function keepPostParameters($keys) {
        $delete = Array();
        foreach ($this->postfields as $key=>$value){
            if (!in_array($key, $keys)){
                array_push($delete, $key);
            }
        }

        foreach ($delete as $key) {
            unset($this->postfields[$key]);
        }

        return $this;
    }

    function updatePostParameter($postarray, $encoded=false)
    {
        if (is_array($postarray))
        {
            foreach ($postarray as $key => $value) {
                if (is_null($value))
                    unset($this->postfields[$key]);
                else
                    $this->postfields[$key] = $value;
            }}
        elseif (is_string($postarray))
        {
            $parr = preg_split("/&/",$postarray);
            foreach ($parr as $postvalue) {
                if (($index = strpos($postvalue, "=")) != false)
                {
                    $key = substr($postvalue, 0,$index);
                    $value = substr($postvalue, $index + 1);
                    if ($encoded)
                        $this->postfields[$key]=urldecode($value);
                    else
                        $this->postfields[$key]=$value;
                }
                else
                    $this->postfields[$postvalue] = "";
            }


        }

        return $this;
    }

    function getResponseXml(){
        //SimpleXMLElement('<INPUT/>')->asXML();
    }

    function SSLVerifyPeer($verify=false)
    {
        curl_setopt($this->ch, CURLOPT_SSL_VERIFYPEER, $verify);
        return $this;
    }
}

?>
Tarun Lalwani
  • 124,930
  • 8
  • 149
  • 214
  • Thank you so much. Appreciate you taking time. It gave me the insights and I will try your code and see if this helps. – Kiran Oct 02 '17 at 13:32
1

@yvesleborg and @tarun-lalwani gave the right hints. You need to take care of the cookies and the redirects. But nevertheless it was not working always for me. I guess the site operator requires some timeout between the two requests.

I rewrote your code a little bit to play around with it. mycurl.php:

function my_curl_init() {
    $url="http://www.mca.gov.in/mcafoportal/loginValidateUser.do"; 
    $user_agent = "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/60.0.3112.113 Safari/537.36";

    $ch = curl_init();  

    curl_setopt($ch, CURLOPT_USERAGENT, $user_agent);
    curl_setopt($ch, CURLOPT_URL, $url);
    curl_setopt($ch, CURLOPT_FOLLOWLOCATION, 1);
    curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);

    return $ch;
}

/*
 * first call in order to get accessCode and sessionCookie
 */
$ch = my_curl_init();
curl_setopt($ch, CURLOPT_COOKIEJAR, __DIR__ . "/cookie.txt"); // else cookielist is empty

$output = curl_exec($ch);

file_put_contents(__DIR__ . '/loginValidateUser.html', $output);

// save cookie info
$cookielist = curl_getinfo($ch, CURLINFO_COOKIELIST);
//print_r($cookielist);

curl_close($ch);

// parse accessCode from output
$re = '/\<input.*name="accessCode".*value="([-0-9]+)"/';
preg_match_all($re, $output, $matches, PREG_SET_ORDER, 0);
if ($matches) {
    $accessCode = $matches[0][1];

    // debug
    echo "accessCode: $accessCode" . PHP_EOL;   


    /*
     * second call in order to login
     */ 

    $post_fields = array(
        'userNamedenc' => 'hGJfsdnk`1t',
        'passwordenc'  => '675894242fa9c66939d9fcf4d5c39d1830f4ddb9',
        'accessCode'   => $accessCode
    );

    $cookiedata = preg_split('/\s+/', $cookielist[0]);
    $session_cookie = $cookiedata[5] . '=' . $cookiedata[6];

    // debug
    echo "sessionCookie: $session_cookie" . PHP_EOL;
    file_put_contents(__DIR__ . '/cookie2.txt', $session_cookie);

    /* 
     * !!! pause !!!
     */  
    sleep(20);

    // debug
    echo "curl -v -L -X POST -b '$session_cookie;' --data 'userNamedenc=hGJfsdnk`1t&passwordenc=675894242fa9c66939d9fcf4d5c39d1830f4ddb9&accessCode=$accessCode' http://www.mca.gov.in/mcafoportal/loginValidateUser.do > loginValidateUser2.html";

    $ch = my_curl_init();  

    curl_setopt($ch, CURLOPT_POST, 1);
    curl_setopt($ch, CURLOPT_POSTFIELDS, $post_fields);   

    curl_setopt($ch, CURLOPT_COOKIE, $session_cookie);

    $output = curl_exec($ch);

    file_put_contents(__DIR__ . '/loginValidateUser2.html', $output);

    curl_close($ch);
}

The script issues two request to the website. The output of the first one is used to read the accessCode and to store the session cookie. Then after a little break the second one is issued using the accessCode and session information together with the login credentials.

I tested it with PHP5.6 from a terminal (php -f mycurl.php). The script debugs all necessary information, outputs a curl command you could use in a terminal and logs the HTML and cookie information to some files in the same folder like the script.

Running the script too often doesn't work. The login won't work. So take your time and wait some minutes between your tries. Or change your IP ;)

Hope it helps.

simne7
  • 151
  • 1
  • 1
  • 10
  • weird, in my tests, you get the following error if you try to log in without solving the captcha: `Enter valid Letters shown`, ps you can check for login errors like this: `$domd = @DOMDocument::loadHTML ( $html ); $xp = new DOMXPath ( $domd ); $loginErrors = $xp->query ( '//ul[@class="errorMessage"]' ); if ($loginErrors->length > 0) { echo 'encountered following error(s) logging in: '; foreach ( $loginErrors as $err ) { echo $err->textContent, PHP_EOL; } die (); } else { echo "logged in successfully!"; }` – hanshenrik Oct 01 '17 at 13:41
  • Strange enough. I tested it again and the file `loginValidateUser2.html` my script above produces contains `Welcome Kiran`. So it seems like that the captcha is not mandatory (at least no for me...) in order to log in programatically. – simne7 Oct 02 '17 at 04:51
0

Replication of the Problem

I did the same thing in Postman as you did your screenshot but was not able to log in:

enter image description here

The only difference I can see is that your request had cookies, which I suspect is why you were able to log in without all the other input fields. And it seems like there are quite a number of input fields:

enter image description here

Using Postman

So, I used postman intercept to capture all the fields used during a login, including the captcha and access code and I was able to login:

enter image description here

Update 1

I've found out that once you've solved a captcha to login, after you've logged out you may login again without displayCaptcha and userEnteredCaptcha in your form data, provided that you use the same cookie as the one you've used to login successfully. You just need to get a valid accessCode from the login page.

aljo f
  • 2,002
  • 20
  • 22
  • I also suggest you change your password after this. – aljo f Sep 27 '17 at 05:09
  • This does not look like the solution. Can you please provide the code that you were able to login ? – Kiran Sep 27 '17 at 15:31
  • Sorry, the point was I was only able to login when I provided the accessCode and captcha. If you need the code check https://pastebin.com/GjLYGqPf . But please note I had to manually put the accessCode and captcha value in the script to make it work. If you need it to be automatic you'll need a captcha solver, like http://www.captchatronix.com/api.php – aljo f Sep 27 '17 at 16:04
  • This code didnt work actually. Did it work on your setup ? – Kiran Sep 28 '17 at 03:42
  • Yes it did, but like I mentioned, the only way it worked is if I: 1. manually entered the accessCode and captcha in the script, and 2. I copied the cookie from my browser (where I looked at the captcha) to cookies.txt. – aljo f Sep 28 '17 at 05:29
0

it doesnt work either in PHP/Python that is (as others have already pointed out) because you were using your browser's existing cookie session, which had already solved the captcha. clear your browser cookies, get a fresh cookie session, and DO NOT SOLVE THE CAPTCHA, and Postman won't be able to log in either. Any idea what is missing ? several things, among them, several post login parameters (browserFlag, loginType,__checkbox_dscBasedLoginFlag, and many more), also your encoding loop here is bugged $str = $str . "$key=$value" . "&"; , it pretty much only works as long as both keys and values only contain [a-zA-Z0-9] characters, and since your userNamedenc contains a grave accent character, your encoding loop is insufficient. a fixed loop would be

foreach($params as $key=>$value){
    $str = $str . urlencode($key)."=".urlencode($value) . "&";
}
$str=substr($str,0,-1);

, but this is exactly why we have the http_build_query function, that entire loop and the following trim can be replaced with this single line:

$str=http_build_query($params);

, also, seems you're trying to login without a pre-existing cookie session, that's not going to work. when you do a GET request to the login page, you get a cookie, and a unique captcha, the captcha answer is tied to your cookie session, and needs to be solved before you attempt to login, you also provide no code to deal with the captcha. also, when parsing the "userName" input element, it will default to "Enter Username", which is emtied with javascript and replaced with userNamedenc, you must replicate this in PHP, also, it will have an input element named "dscBasedLoginFlag", which is removed with javascript, you must also do this part in php, also it has an input element named "Cert", which has a default value, but this value is cleared with javascript, do the same in php, and an input element named "newUserRegistration", which is removed with javascript, do that,

here's what you should do: make a GET request to the login page, save the cookie session and make sure to provide it for all further requests, and parse out all the login form's elements and add them to your login request (but be careful, there is 2x form inputs, 1 belong to the search bar, only parse the children of the login form, don't mix the 2), and remember to clear/remove the special input tags to emulate the javascript, as described above, then make a GET request to the captcha url, make sure to provide the session cookie, solve the captcha, then make the final login request, with the captcha answer, and userNamedenc and passwordenc and all the other elements parsed out from the login page... that should work. now, solving the captcha programmatically, the captha doesn't look too hard, cracking it probably can be automated, but until someone actually does that, you can use Deathbycaptcha to do it for you, however note that it isn't a free service.

here's a fully tested, working example implementation, using my hhb_curl library (from https://github.com/divinity76/hhb_.inc.php/blob/master/hhb_.inc.php ), and the Deathbycaptcha api:

<?php
declare(strict_types = 1);
header ( "content-type: text/plain;charset=utf8" );
require_once ('hhb_.inc.php');
const DEATHBYCATPCHA_USERNAME = '?';
const DEATHBYCAPTCHA_PASSWORD = '?';
$hc = new hhb_curl ( '', true );
$hc->setopt(CURLOPT_TIMEOUT,20);// im on a really slow net atm :(
$html = $hc->exec ( 'http://www.mca.gov.in/mcafoportal/login.do' )->getResponseBody (); // cookie session etc
$domd = @DOMDocument::loadHTML ( $html );
$inputs = getDOMDocumentFormInputs ( $domd, true ) ['login'];
$params = [ ];
foreach ( $inputs as $tmp ) {
    $params [$tmp->getAttribute ( "name" )] = $tmp->getAttribute ( "value" );
}
assert ( isset ( $params ['userNamedenc'] ), 'username input not found??' );
assert ( isset ( $params ['passwordenc'] ), 'passwordenc input not found??' );
$params ['userName'] = ''; // defaults to "Enter Username", cleared with javascript
unset ( $params ['dscBasedLoginFlag'] ); // removed with javascript
$params ['Cert'] = ''; // cleared to emptystring with javascript
unset ( $params ['newUserRegistration'] ); // removed with javascript
unset ( $params ['SelectCert'] ); // removed with javascript
$params ['userNamedenc'] = 'hGJfsdnk`1t';
$params ['passwordenc'] = '675894242fa9c66939d9fcf4d5c39d1830f4ddb9';
echo 'parsed login parameters: ';
var_dump ( $params );
$captchaRaw = $hc->exec ( 'http://www.mca.gov.in/mcafoportal/getCapchaImage.do' )->getResponseBody ();
$params ['userEnteredCaptcha'] = solve_captcha2 ( $captchaRaw );
// now actually logging in.
$html = $hc->setopt_array ( array (
        CURLOPT_POST => true,
        CURLOPT_POSTFIELDS => http_build_query ( $params ) 
) )->exec ( 'http://www.mca.gov.in/mcafoportal/loginValidateUser.do' )->getResponseBody ();
var_dump ( $hc->getStdErr (), $hc->getStdOut () ); // printing debug data
$domd = @DOMDocument::loadHTML ( $html );
$xp = new DOMXPath ( $domd );
$loginErrors = $xp->query ( '//ul[@class="errorMessage"]' );
if ($loginErrors->length > 0) {
    echo 'encountered following error(s) logging in: ';
    foreach ( $loginErrors as $err ) {
        echo $err->textContent, PHP_EOL;
    }
    die ();
}
echo "logged in successfully!";
/**
 * solves the captcha manually, by doing: echo ANSWER>captcha.txt
 *
 * @param string $raw_image
 *          raw image bytes
 * @return string answer
 */
function solve_captcha2(string $raw_image): string {
    $imagepath = getcwd () . DIRECTORY_SEPARATOR . 'captcha.png';
    $answerpath = getcwd () . DIRECTORY_SEPARATOR . 'captcha.txt';
    @unlink ( $imagepath );
    @unlink ( 'captcha.txt' );
    file_put_contents ( $imagepath, $raw_image );
    echo 'the captcha is saved in ' . $imagepath . PHP_EOL;
    echo ' waiting for you to solve it by doing: echo ANSWER>' . $answerpath, PHP_EOL;
    while ( true ) {
        sleep ( 1 );
        if (file_exists ( $answerpath )) {
            $answer = trim ( file_get_contents ( $answerpath ) );
            echo 'solved: ' . $answer, PHP_EOL;
            return $answer;
        }
    }
}
function solve_captcha(string $raw_image): string {
    echo 'solving captcha, hang on, with DEATBYCAPTCHA this usually takes between 10 and 20 seconds.';
    {
        // unfortunately, CURLFile requires a filename, it wont accept a string, so make a file of it
        $tmpfileh = tmpfile ();
        fwrite ( $tmpfileh, $raw_image ); // TODO: error checking (incomplete write or whatever)
        $tmpfile = stream_get_meta_data ( $tmpfileh ) ['uri'];
    }
    $hc = new hhb_curl ( '', true );
    $hc->setopt_array ( array (
            CURLOPT_URL => 'http://api.dbcapi.me/api/captcha',
            CURLOPT_POSTFIELDS => array (
                    'username' => DEATHBYCATPCHA_USERNAME,
                    'password' => DEATHBYCAPTCHA_PASSWORD,
                    'captchafile' => new CURLFile ( $tmpfile, 'image/png', 'captcha.png' ) 
            ) 
    ) )->exec ();
    fclose ( $tmpfileh ); // when tmpfile() is fclosed(), its also implicitly deleted.
    $statusurl = $hc->getinfo ( CURLINFO_EFFECTIVE_URL ); // status url is given in a http 300x redirect, which hhb_curl auto-follows
    while ( true ) {
        // wait for captcha to be solved.
        sleep ( 10 );
        echo '.';
        $json = $hc->setopt_array ( array (
                CURLOPT_HTTPHEADER => array (
                        'Accept: application/json' 
                ),
                CURLOPT_HTTPGET => true 
        ) )->exec ()->getResponseBody ();
        $parsed = json_decode ( $json, false );
        if (! empty ( $parsed->captcha )) {
            echo 'captcha solved!: ' . $parsed->captcha, PHP_EOL;
            return $parsed->captcha;
        }
    }
}
function getDOMDocumentFormInputs(\DOMDocument $domd, bool $getOnlyFirstMatches = false): array {
    // :DOMNodeList?
    $forms = $domd->getElementsByTagName ( 'form' );
    $parsedForms = array ();
    $isDescendantOf = function (\DOMNode $decendant, \DOMNode $ele): bool {
        $parent = $decendant;
        while ( NULL !== ($parent = $parent->parentNode) ) {
            if ($parent === $ele) {
                return true;
            }
        }
        return false;
    };
    // i can't use array_merge on DOMNodeLists :(
    $merged = function () use (&$domd): array {
        $ret = array ();
        foreach ( $domd->getElementsByTagName ( "input" ) as $input ) {
            $ret [] = $input;
        }
        foreach ( $domd->getElementsByTagName ( "textarea" ) as $textarea ) {
            $ret [] = $textarea;
        }
        foreach ( $domd->getElementsByTagName ( "button" ) as $button ) {
            $ret [] = $button;
        }
        return $ret;
    };
    $merged = $merged ();
    foreach ( $forms as $form ) {
        $inputs = function () use (&$domd, &$form, &$isDescendantOf, &$merged): array {
            $ret = array ();
            foreach ( $merged as $input ) {
                // hhb_var_dump ( $input->getAttribute ( "name" ), $input->getAttribute ( "id" ) );
                if ($input->hasAttribute ( "disabled" )) {
                    // ignore disabled elements?
                    continue;
                }
                $name = $input->getAttribute ( "name" );
                if ($name === '') {
                    // echo "inputs with no name are ignored when submitted by mainstream browsers (presumably because of specs)... follow suite?", PHP_EOL;
                    continue;
                }
                if (! $isDescendantOf ( $input, $form ) && $form->getAttribute ( "id" ) !== '' && $input->getAttribute ( "form" ) !== $form->getAttribute ( "id" )) {
                    // echo "this input does not belong to this form.", PHP_EOL;
                    continue;
                }
                if (! array_key_exists ( $name, $ret )) {
                    $ret [$name] = array (
                            $input 
                    );
                } else {
                    $ret [$name] [] = $input;
                }
            }
            return $ret;
        };
        $inputs = $inputs (); // sorry about that, Eclipse gets unstable on IIFE syntax.
        $hasName = true;
        $name = $form->getAttribute ( "id" );
        if ($name === '') {
            $name = $form->getAttribute ( "name" );
            if ($name === '') {
                $hasName = false;
            }
        }
        if (! $hasName) {
            $parsedForms [] = array (
                    $inputs 
            );
        } else {
            if (! array_key_exists ( $name, $parsedForms )) {
                $parsedForms [$name] = array (
                        $inputs 
                );
            } else {
                $parsedForms [$name] [] = $tmp;
            }
        }
    }
    unset ( $form, $tmp, $hasName, $name, $i, $input );
    if ($getOnlyFirstMatches) {
        foreach ( $parsedForms as $key => $val ) {
            $parsedForms [$key] = $val [0];
        }
        unset ( $key, $val );
        foreach ( $parsedForms as $key1 => $val1 ) {
            foreach ( $val1 as $key2 => $val2 ) {
                $parsedForms [$key1] [$key2] = $val2 [0];
            }
        }
    }
    return $parsedForms;
}

example usage: in terminal, write php foo.php | tee test.html, after a few seconds it will say something like:

the captcha is saved in /home/captcha.png
 waiting for you to solve it by doing: echo ANSWER>/home/captcha.txt

then look at the captcha in /home/captcha.png , solve it, and write in another terminal: echo ANSWER>/home/captcha.txt, now the script will log in, and dump the logged in html in test.html, which you can open in your browser, to confirm that it actually logged in, screenshot when i run it: https://image.prntscr.com/image/_AsB_0J6TLOFSZuvQdjyNg.png

also note that i made 2 captcha solver functions, 1 use the deathbycaptcha api, and wont work until you provide a valid and credited deathbycaptcha account on line 5 and 6, which is not free, the other 1, solve_captcha2, asks you to solve the captcha yourself, and tells you where the captcha image is saved (so you can go have a look at it), and what command line argument to write, to provide it with the answer. just replace solve_captcha with solve_captcha2 on line 28, to solve it manually, and vise-versa. the script is fully tested with solve_captcha2, but the deathbycaptcha solver is untested, as my deathbycatpcha account is empty (if you would like to make a donation so i can actually test it, send 7 dollars to paypal account divinity76@gmail.com with a link to this thread, and i will buy the cheapest deathbycaptcha credit pack and actually test it)

  • disclaimer: i am not associated with deathbycaptcha in any way (except that i was a customer of theirs a couple of years back), and this post was not sponsored.
hanshenrik
  • 15,263
  • 3
  • 28
  • 61