| 1 |
On 3 October 2010 13:28, Michał Górny <mgorny@g.o> wrote: |
| 2 |
> Hello, |
| 3 |
> |
| 4 |
> I would like to propose a new attempt at Manifest signatures. Instead |
| 5 |
> of using a single per-Manifest signature, we would keep separate |
| 6 |
> signatures for each of the files, as an additional (optional) hash |
| 7 |
> type. |
| 8 |
> |
| 9 |
> |
| 10 |
> Motivation |
| 11 |
> ---------- |
| 12 |
> The current signing approach gives all the responsibility for Manifest |
| 13 |
> signature to the developer who committed last update to the ebuild |
| 14 |
> directory regardless of the actual commit significance. |
| 15 |
> |
| 16 |
> Consider the following: Dev A is the primary package maintainer. He/she |
| 17 |
> reviewed all the ebuilds and committed a signed Manifest. Then Dev B |
| 18 |
> performs a slight cleanup of the ebuild directory. He/she modifies |
| 19 |
> metadata.xml a little and/or removes an old ebuild. |
| 20 |
> |
| 21 |
> The actual ebuilds weren't modified -- yet Dev B has to sign all |
| 22 |
> of them once again. Moreover, if Dev B doesn't use Manifest signing, |
| 23 |
> the signature from Dev A is lost. |
| 24 |
|
| 25 |
If we make the GPG signatures mandatory at some point of time, that |
| 26 |
addresses the second of your concerns. I do not understand why the |
| 27 |
first a problem - could you clarify? |
| 28 |
|
| 29 |
Cheers, |
| 30 |
-- |
| 31 |
Arun Raghavan |
| 32 |
http://arunraghavan.net/ |
| 33 |
(Ford_Prefect | Gentoo) & (arunsr | GNOME) |
Archives