Added VNC standard DES authentication

[mcclymont]

I'm working with Xen servers that have VNC authentication, found myself writing this to get my custom packer plugin working.

So far I have tested this against Xen, and RealVNC

A reference to explain the "non-RFC-documented" behaviour of VNC authentication:
http://bytecrafter.blogspot.com.au/2010/09/des-encryption-as-used-in-vnc.html

[kward]

Would love to see this merged.

[mcclymont]

I believe all the comments of @vtolstov have been addressed. Let me know if it's preferred to squash my commits together

[kward]

FYI, I've added a pull request for mcclymont/go-vnc that adds an additional unit test for the VNC Authentication. (mcclymont#1)

[mcclymont]

I rebased onto master, squashing the 2 commits together, and force pushed. I didn't realise that @vtolstov and I had made comments directly on the commits rather than on the PR - they are no longer visible.

@kward, I'm finding the concept of a PR on a PR difficult, since I can't tell what the repo owner @mitchellh would prefer in terms of functionality and style. I'm also not convinced of the utility of the integration test with a literal hex code challenge and response. Therefore I would prefer to keep my PR simpler, how it is now.

[kward]

The unit test demonstrates that the bit rotation code works properly with
various length passwords, which also proves that the 0 padding is working.
The point is that these are challenge/response values that properly worked
against a real server. If the code were to be changed (e.g. removing the
bit rotation code because it isn't trusted), the code would break.

On Mon, May 4, 2015 at 2:06 PM, Chris McClymont notifications@github.com
wrote:

I rebased onto master, squashing the 2 commits together, and force pushed.
I didn't realise that @vtolstov https://github.com/vtolstov and I had
made comments directly on the commits rather than on the PR - they are no
longer visible.

@kward https://github.com/kward, I'm finding the concept of a PR on a
PR difficult, since I can't tell what the repo owner @mitchellh
https://github.com/mitchellh would prefer in terms of functionality and
style. I'm also not convinced of the utility of the integration test with a
literal hex code challenge and response. Therefore I would prefer to keep
my PR simpler, how it is now.

—
Reply to this email directly or view it on GitHub
#7 (comment).

Kate Ward kate.ward@forestent.com

[vtolstov]

nice, @mitchellh please merge..