Sample code for 30+ languages & platforms
C++

AEAD AES 128-bit GCM

See more Encryption Examples

Demonstrates AES encryption using the Galois/Counter Mode (GCM). GCM is an authenticated encryption mode with "additional data" (often referred to as AEAD). GCM is a cipher mode that can be applied to any symmetric encryption algorithm with a 16-byte block size, such as AES and Twofish. In GCM mode, the block encryption algorithm is transformed into a stream encryption algorithm, and therefore no padding occurs (and the PaddingScheme property does not apply). The "additional data" (known as the AAD) does not get encrypted but plays a role in the computation of the resulting "authenticated tag".

Chilkat C++ Downloads

C++
#include <CkCrypt2.h>

void ChilkatSample(void)
    {
    bool success = false;

    // This example assumes the Chilkat API to have been previously unlocked.
    // See Global Unlock Sample for sample code.

    CkCrypt2 crypt;

    // Set the encryption algorithm to "AES"	
    crypt.put_CryptAlgorithm("aes");

    // Indicate that the Galois/Counter Mode (GCM) should be used:
    crypt.put_CipherMode("gcm");

    // KeyLength may be 128, 192, 256
    crypt.put_KeyLength(128);

    // This is the 128-bit AES secret key (in hex format)
    const char *K = "feffe9928665731c6d6a8f9467308308";

    // This is the 16-byte initialization vector:
    const char *IV = "cafebabefacedbaddecaf888";

    // This is the additional data to be used as input to the GCM AEAD algorithm,
    // but is not included in the output.  It plays a role in the computation of the
    // resulting authenticated tag.
    const char *AAD = "feedfacedeadbeeffeedfacedeadbeefabaddad2";

    // The plain-text bytes (in hex format) to be encrypted.
    const char *PT = "d9313225f88406e5a55909c5aff5269a86a7a9531534f7da2e4c303d8a318a721c3c0c95956809532fcf0e2449a6b525b16aedf5aa0de657ba637b39";

    // The expected cipher text (in hex format)
    const char *CT = "42831ec2217774244b7221b784d0d49ce3aa212f2c02a4e035c17e2329aca12e21d514b25466931c7d8f6a5aac84aa051ba30b396a0aac973d58e091";

    // The expected authenticated tag given the above inputs.
    const char *T = "5bc94fbc3221a5db94fae95ae7121a47";

    // Note: The above data are the values for test vector #4 from 
    // the PDF document at: http://csrc.nist.gov/groups/ST/toolkit/BCM/documents/proposedmodes/gcm/gcm-spec.pdf

    // EncodingMode specifies the encoding of the output for
    // encryption, and the input for decryption.
    // It may be "hex", "url", "base64", or "quoted-printable".
    crypt.put_EncodingMode("hex");

    // Set the secret key and IV
    crypt.SetEncodedIV(IV,"hex");
    crypt.SetEncodedKey(K,"hex");

    // Set the additional authenticated data (AAD)
    success = crypt.SetEncodedAad(AAD,"hex");

    // For the purpose of duplicating the test vectors, we are using the EncryptEncoded method.
    // This method decodes the input string according to the encoding specified by the EncodingMode
    // property, which in this case is "hex".  The decoded bytes are encrypted using the mode specified
    // by the CipherMode property.  The resulting
    // encrypted bytes are encoded (again using the encoding mode specified by EncodingMode),
    // and the result is returned.
    // <b>Note:</b> The CipherMode property sets the block mode of operation (gcm, cfb, cbc, ofb, ecb, etc.) 
    // for any of the Chilkat encryption/decryption methods (such as EncryptBytes, EncryptString, 
    // CkEncryptFile, etc.)   Just because GCM mode is demonstrated with EncryptEncoded/DecryptEncoded,
    // does not imply that GCM mode is specific to only these methods.
    const char *ctResult = crypt.encryptEncoded(PT);
    if (crypt.get_LastMethodSuccess() != true) {
        std::cout << crypt.lastErrorText() << "\r\n";
        return;
    }

    // Examine the result.  It should be the same (case insensitive) as our expected result:
    std::cout << "computed result: " << ctResult << "\r\n";
    std::cout << "expected result: " << CT << "\r\n";

    // Examine the authenticated tag. It should be the same (case insensitive) as our expected authenticated tag:
    const char *tResult = crypt.getEncodedAuthTag("hex");
    std::cout << "computed authTag: " << tResult << "\r\n";
    std::cout << "expected authTag: " << T << "\r\n";

    // -------------------------------------------------------------------------------------
    // Now let's GCM decrypt...
    // -------------------------------------------------------------------------------------

    // Before GCM decrypting, we must set the authenticated tag to the value that is expected.
    // The decryption will fail if the resulting authenticated tag is not equal (case insensitive) to 
    // the expected result.
    // Note: The return value of SetEncodedAuthTag indicates whether the string passed was a valid
    // representation of the encoding specified in the 2nd arg.
    success = crypt.SetEncodedAuthTag(T,"hex");

    // All of our properties (IV, secret key, cipher mode, and AAD) are already set from the code above...

    // So let's decrypt CT to and check to see if we get PT.
    const char *ptResult = crypt.decryptEncoded(CT);
    if (crypt.get_LastMethodSuccess() != true) {
        // Failed.  The resultant authenticated tag did not equal the expected authentication tag.
        std::cout << crypt.lastErrorText() << "\r\n";
        return;
    }

    // Examine the decrypted result.  It should be the same as our expected plaintext (case insensitive)
    std::cout << "plaintext decrypted: " << ptResult << "\r\n";
    std::cout << "plaintext expected:  " << PT << "\r\n";

    // Let's intentionally set the expected authenticated tag to an incorrect value. 
    // The decrypt operation should fail:
    const char *tInvalid = "ffaabbbc3221a5db94fae95ae7121a47";

    success = crypt.SetEncodedAuthTag(tInvalid,"hex");

    ptResult = crypt.decryptEncoded(CT);
    if (crypt.get_LastMethodSuccess() != true) {
        // Failed.  The resultant authenticated tag did not equal the expected authentication tag.
        std::cout << crypt.lastErrorText() << "\r\n";
    }
    }