1 |
>>>>> On Tue, 7 Oct 2014, Michał Górny wrote: |
2 |
|
3 |
> Dnia 2014-10-07, o godz. 09:57:32 |
4 |
> Dirkjan Ochtman <djc@g.o> napisał(a): |
5 |
|
6 |
>> On Mon, Oct 6, 2014 at 10:48 AM, Michał Górny <mgorny@g.o> wrote: |
7 |
>> > I was thinking of prepending the old history via 'git replace' in |
8 |
>> > the gitweb repo to allow looking back. However, that idea has the |
9 |
>> > downside that users would be confused by having past commits in |
10 |
>> > gitweb yet not in clones. |
11 |
>> > |
12 |
>> > Kent improved my idea suggesting that we use a separate repo for |
13 |
>> > that 'complete history' view. That is, we would have three repos: |
14 |
>> > |
15 |
>> > 1. history.git -- with past CVS history up to conversion, |
16 |
>> > |
17 |
>> > 2. dev.git -- with history starting with conversion, |
18 |
>> > |
19 |
>> > 3. joined-history.git -- dev.git with 'git replace' for |
20 |
>> > history.git, that is the complete history including both pre- and |
21 |
>> > post-conversion commits. |
22 |
>> |
23 |
>> Can we do that without requiring the git replace stuff? E.g. have |
24 |
>> dev.git be a shallow clone of a joined-history.git? |
25 |
|
26 |
> No, I don't think we can push to a shallow repo. Additionally, this |
27 |
> brings back all the issues I mentioned in the first mail. |
28 |
|
29 |
But dev (= master) and history could be two branches in the same |
30 |
repository? So we wouldn't need three repos? |
31 |
|
32 |
Ulrich |