Re: [PATCH 2/2] guix substitute-binary: Support the Signature field of a narinfo file.

[ludo@gnu.org (Ludovic Courtès)]

Nikita Karetnikov <nikita@karetnikov.org> skribis:

> The perfect is the enemy of the good.  So, may I push both patches to
> ‘core-updates’?

Yes, looks good to me!

I was about to say you could push to master, but since the new
substitute-binary will refuse any unsigned archives[*], it may be wiser
to push to a special branch so we can test it in real life before
merging into master.

WDYT?

We also need to adjust Hydra to use our ‘guix authenticate’.

[*] Actually, ‘alist->record’ will simply bork if the ‘Signature’ field
    is missing.  Eventually we may want graceful handling.

> From 26e98465deac592cbcdb9891f091d997e00eda9d Mon Sep 17 00:00:00 2001
> From: Nikita Karetnikov <nikita@karetnikov.org>
> Date: Wed, 12 Mar 2014 21:21:39 +0000
> Subject: [PATCH 1/2] nar: Clarify that 'assert-valid-signature' accepts a
>  string.
>
> * guix/nar.scm (assert-valid-signature): Turn a comment into a
>   docstring and improve the wording.

Actually docstrings are useless for internal defines, because they
cannot be accessed, so I usually use comments in those cases.

The wording fix is OK.

Thank you!

Ludo’.