Re: More inter samples

Takeshi Imamura wrote:

> >> Attached is a revised version of the samples sent before. I mainly
> >> update the Diffie-Hellman examples to accommodate the newest changes in
> >> the specification.
>
> I confirmed that our implementation succeeded in decrypting most of your
> examples except ones using DH and RSA-OAEP with SHA256 and SHA512.  I also
> confirmed that it failed to decrypt your bad-algorithm example because of
> wrong key length.

Thanks! As for the DH and RSA-OAEP with SHA256 and SHA512, I am wondering if
you
couldn't decrypt them or you haven't tested them.

> However, I found that it succeeded in decrypting your
> bad-type example.  That is reasonable to me because the decryptor is not
> required to perform validation on the serialized XML and hence our
> implementation does not.  Should we include this example in test vectors?

My intention is that if you do the decrypt and replace, the type information
should be needed.
In other words, it should cause you trouble when you replace the EncryptedData
element with
the decrypted data if the the type attribute is not set correctly.

>
>
> >Thank you! I've linked this from [1]. Would you mind sending a report
> then?
> >(Yes and No down the column)? For most of the application and processing
> >features, there's no interop, but self-reporting. But we could do interop
> >on bits of the spec other than the crypto:
> >
> >1. We could interop testing on the encoding requirements. (Have a document
> >in an exotic encoding with an encrypted element which is obviously in
> >UTF-8).
> >2. The Decryption Transform.
> >
> >And of course, I want to make sure performance is acceptable, but I don't
> >have a metric. What scenarios are folks looking at, 3 encryptions in a
> 1Meg
> >files?

I will send a report soon. I will also add more samples for anything else
interopable
in the future.

Thanks.

Jiandong Guo
Phaos Technology Corp.

Received on Monday, 11 March 2002 10:26:53 UTC