Encode URL from C#, decrypt URL in PHP (extra characters being added somehow)

I have read so many topics on this very subject by now, I can't understand where the issue could possibly lie. I am encrypting part of a URL from a C# winform application. I then want to read in the URL using php and decrypt the url (all using base-64). I do have some code to shrae:

Code to encrypt URL (C#):

public static string Base64Encode(string str)
{
    byte[] encbuff = Encoding.UTF8.GetBytes(str);
    return System.Web.HttpServerUtility.UrlTokenEncode(encbuff);
}

Decrypt a section of the URL:

Base64Encode("CND0311J4S68CCU Ver. F.0BHPQOEM - f");

Returns:

Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY1

Code 开发者_开发问答to decrypt URL (PHP):

echo base64_decode("Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY1");

Returns:

CND0311J4S68CCU Ver. F.0BHPQOEM - f5

So, where is the extra "5" at the end of the return coming from? I cannot figure this out for the life of me, quite frustrating as you could imagine.

I appreciate any help with this - as well as any suggestions!

Thank you,

Evan

---

"CND0311J4S68CCU Ver. F.0BHPQOEM - f"

encoded as base64 is not:

Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY1

Probably something else is adding the 1 at the end, because

echo base64_decode("Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY");

gives you what you're looking for. And that something adding it is in fact System.Web.HttpServerUtility.UrlTokenEncode. The issue is the following (from MSDN):

This (System.Web.HttpServerUtility.UrlTokenEncode) does not use a standard encoding. It encodes with - and _ characters which is standard. It also removes the = signs. But rather than simply removing them (they're not necessary to decode the string), it replaces them with a digit (0, 1, 2) indicating the number of = signs that were removed.

So go for it (Demo):

<?php

$urltoken = "Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY1";
echo urltoken_decode($urltoken);

function urltoken_decode($token)
{
    return base64_decode(substr($token, 0, -1));
}

The function is pretty rough and could be improved to actually deal with it more specifically (Demo2):

function urltoken_decode($token)
{
    $len = strlen($token);
    if (!$len)
       return $token;

    $digit = $token[$len-1];
    if (!in_array($digit, range(0,2)))
    {
        throw InvalidArgumentException(sprintf('Invalid end digit (%s).', $digit));
    }
    return base64_decode(substr($token, 0, -1));
}

---

Interesting - when I encode "CND0311J4S68CCU Ver. F.0BHPQOEM - f" at this site (using JavaScript) it encodes as "Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY". If I decode "Q05EMDMxMUo0UzY4Q0NVIFZlci4gRi4wQkhQUU9FTSAtIGY1" I get the extra 5 on the end. Apparently the problem is on the encoding end - maybe try outputting the value you're encoding just before it's passed to the base64 encoder to make sure it's exactly what you think it is?