1 |
On Sat, 2019-08-17 at 09:42 -0400, Michael Orlitzky wrote: |
2 |
> On 8/17/19 4:54 AM, Michał Górny wrote: |
3 |
> > On Sat, 2019-08-17 at 10:52 +0200, Ulrich Mueller wrote: |
4 |
> > > Shouldn't there be a blocker against dev-vcs/gitolite{,-gentoo} |
5 |
> > > (and vice versa)? These packages cannot be installed at the same time, |
6 |
> > > and I guess that a direct blocker would result in a friendlier error |
7 |
> > > message than REQUIRED_USE magic in acct-user/git. |
8 |
> > > |
9 |
> > |
10 |
> > You're probably right. I'll update the patches to add mutual blockers |
11 |
> > everywhere. |
12 |
> > |
13 |
> |
14 |
> I discovered on the pull request that "git" is just the name that the |
15 |
> upstream examples use, but isn't required for either gitolite or gitea: |
16 |
> |
17 |
> * https://docs.gitea.io/en-us/install-from-binary/ |
18 |
> |
19 |
> * https://gitolite.com/gitolite/concepts#the-hosting-user |
20 |
> |
21 |
> For example, on Fedora and Debian, gitolite uses a "gitolite" user. We |
22 |
> should also be able to use "gitea" for gitea, meaning that the two |
23 |
> packages don't have to block each other, and that they don't have to |
24 |
> fight over the same username. |
25 |
> |
26 |
> I realize we'd have to tell people how to rename the account to support |
27 |
> upgrades -- but is there some other reason to keep the shared "git" name? |
28 |
|
29 |
The argument I've been told is that users expect 'git@...' to work |
30 |
as remote URI on their boxes. They don't want users to bind the URI to |
31 |
specific implementation. |
32 |
|
33 |
-- |
34 |
Best regards, |
35 |
Michał Górny |