Restructure vectors into ciphers/hashes/hmac

[reaperhulk]

Also restructures the tests. Refs #184

[alex]

I'm not sure stuff is really fully consistent under this patch, there seems to be ciphers/Camellia/, ciphers/AES/KAT/, and hashes/SHABYTE/. I think we should try to tend towards only having one form, not sure what the components should be, but we should use algorithm/origin consistently.

[reaperhulk]

SHABYTE should definitely be changed (and split up for SHA-1/SHA-2 family). I was unsure what to do about the KAT/MMT vectors for AES so I punted on that one at the time. We could split them up by mode, which would be a bit nicer.

[alex]

Going by mode for ciphers sounds like a great idea.

On Thu, Oct 31, 2013 at 9:37 AM, Paul Kehrer notifications@github.comwrote:

SHABYTE should definitely be changed (and split up for SHA-1/SHA-2
family). I was unsure what to do about the KAT/MMT vectors for AES so I
punted on that one at the time. We could split them up by mode, which would
be a bit nicer.

—
Reply to this email directly or view it on GitHubhttps://github.com//pull/198#issuecomment-27502403
.

"I disapprove of what you say, but I will defend to the death your right to
say it." -- Evelyn Beatrice Hall (summarizing Voltaire)
"The people's good is the highest law." -- Cicero
GPG Key fingerprint: 125F 5C67 DFE9 4084

[dreid]

test_OpenSSL doesn't seem like a very good name anymore? Probably this should just be test_{mode}? That's at least what TestAESCTR.test_OpenSSL appears to have ended up as. Maybe they should even be moved to a single TestCamellia class?

[reaperhulk]

I think that makes sense. We could also collapse the test classes for the other ciphers to be consistent.

[dreid]

Want to rename these to be test_{mode} and collapse in a separate pull request?

[reaperhulk]

Up to you since you're reviewing. I'm fine with a separate PR.

[dreid]

Ok, lets rename the methods here, and then I'll merge it, and we can do the collapse in a separate pull request.