1 |
On 03/04/10 23:11, Brian Harring wrote: |
2 |
> Random sidenote, anyone looked at using an alternate vcs to do the |
3 |
> work, then proxy it back? Specifically thinking of workflow like svk |
4 |
> (or in this case hg cvs, |
5 |
> https://wiki.mozilla.org/Using_Mercurial_locally_with_CVS ). The |
6 |
> reason I ask is that via building the work up outside of cvs, then |
7 |
> proxying the add/remove/modifications back into it, it should be |
8 |
> possible to minimize the window of cvs breakage down to bare minimum |
9 |
> while still getting the same level of QA validation for the changes. |
10 |
|
11 |
Assuming the person using such a tool |
12 |
|
13 |
- is fluent in using such tool |
14 |
(to at least compensate extra abstraction. with plain CVS you |
15 |
at least know for sure what's happening) |
16 |
|
17 |
- is manually doing extra commits for manifest fixes |
18 |
(like "repoman commit" would do for him otherwise) |
19 |
|
20 |
|
21 |
|
22 |
Sebastian |