1 |
Robin H. Johnson писал 2012-05-23 19:47: |
2 |
> On Wed, May 23, 2012 at 02:42:37PM +0200, Michael Weber wrote: |
3 |
>> i've looked at the blockers of "[TRACKER] portage migration to git" |
4 |
>> [1] and want to discuss "testing git-cvsserver" [2]. |
5 |
>> |
6 |
>> There are two proposed scenarios how to migrate the developers write |
7 |
>> access to the portage tree. |
8 |
> The primary reasons to continue to support CVS-style access via |
9 |
> git-cvsserver: |
10 |
> 1. Lightweight partial/subtree checkouts |
11 |
> - Git has implemented subtree checkouts, but they still bring down |
12 |
> a |
13 |
> fairly large packfile. |
14 |
> 2. Arches were Git repos are too heavy (Kumba wanted this for MIPS) |
15 |
|
16 |
Isnt git works with shallow clone? like |
17 |
# git clone --depth 1 <or any other desired value> |
18 |
git+ssh://gitrepo.uri::repo |
19 |
|
20 |
So you can clone in this manner and push changes back |
21 |
|
22 |
Also for depth = 1 pack size will be similar to gzipped rsync snapshot |
23 |
(~40M) |
24 |
|
25 |
|
26 |
> If we can get rid of #2, we're willing to live with #1. |
27 |
> |
28 |
>> "Clean cut" turns of cvs access on a given and announced timestamp, |
29 |
>> rsync-generation/updates is suspended (no input -> no changes), some |
30 |
>> magic scripts prepare the git repo (according to [3], some hours |
31 |
>> duration) and we all checkout the tree (might be some funny massive |
32 |
>> load). |
33 |
> 1. You will be given git bundles instead of being allowed to do |
34 |
> initial |
35 |
> clone. That way it's just a resumable HTTP download. |
36 |
> 2. rsync generation is NOT going away. Users will still be using it. |
37 |
|
38 |
-- |
39 |
Best Regards, |
40 |
Alexey 'Alexxy' Shvetsov |
41 |
Petersburg Nuclear Physics Institute, NRC Kurchatov Institute, |
42 |
Gatchina, Russia |
43 |
Department of Molecular and Radiation Biophysics |
44 |
Gentoo Team Ru |
45 |
Gentoo Linux Dev |
46 |
mailto:alexxyum@×××××.com |
47 |
mailto:alexxy@g.o |
48 |
mailto:alexxy@×××××××××××××.ru |