IDP-616

[Cantor, Scott]

On 2/26/15, 6:01 PM, "Tom Zeller" <tzeller at dragonacea.biz> wrote:

>On Thu, Feb 26, 2015 at 4:55 PM, Cantor, Scott <cantor.2 at osu.edu> wrote:
>> If I have until Monday at least, I can give it a try.
>
>Fine with me.

Alright. It's not even self-evident yet that it's a real layering problem. 
The actual key information comes from a KeyStrategy inner bean that I 
could pull out of the DataSealer and inject somewhere else at the same 
time and that already has an API for getting the default key (and it 
returns the version/ID of that key).

I think all that's missing is a way to know what the key used to encrypt 
the previous blob was, and I can probably do something to the DataSealer 
to expose that with a new API for the unseal() call. If they don't match, 
I could hopefully just tag the record dirty and be done with it.

-- Scott