Re: Bug 24410 - AES-GCM

On Thu, Feb 20, 2014 at 8:06 PM, Jim Schaad <ietf@augustcellars.com> wrote:

>
>
>
>
> *From:* Jim Schaad [mailto:ietf@augustcellars.com]
> *Sent:* Thursday, February 20, 2014 5:45 PM
> *To:* 'Mark Watson'; public-webcrypto@w3.org
> *Subject:* Bug 24410 - AES-GCM
>
>
>
> 1.       Section 18.13.1 has no text
>
Ok, I will fix this.


> 2.       I think that I have a problem with step #1 in the Encrypt
> operation description.  I believe the text should say "If any of the
> required members of ..." given that additionalData and tagLength can be
> absent without causing an error in the algorithm.  This may be an error in
> my knowledge of how WebIDL works however as absent and null may be the same
> thing.
>
IIUC, the "nullable" types (with the ?) mean that the parameter can be
present and take the value null. This is different from it being not
present.

However, it does seem natural for us to treat the case where it is not
present the same as the case where is it null, otherwise we force people to
include these properties explicitly with the value null.


> 3.       SP 800-38D is missing from the references section.
>
Ok.

https://www.w3.org/Bugs/Public/show_bug.cgi?id=24766 for all three.

>
>
>
>
> Jim
>
>
>
> I missed one more issue.
>
>
>
> When doing an import operation, one needs to match both "A128GCM" and
> "A128GCMKW".  In writing this I just also noted that the JSON document is
> using A128GCM not AES128GCM, we will need to do a double check on the
> string names that are defined in the JOSE document as they tend to try and
> omit letters when they can.
>