| 1 |
On 10:44 Wed 23 Mar , James Cloos wrote: |
| 2 |
> TC> So live with it. |
| 3 |
> |
| 4 |
> I cannot. It makes the eclass useless. |
| 5 |
> |
| 6 |
> I have almost 2 gigs of bare repo in distdirs/git-src. |
| 7 |
> |
| 8 |
> A forced re-download of all of that is just not possible! |
| 9 |
> |
| 10 |
> The existing distdir clones *MUST* continue to work. |
| 11 |
> |
| 12 |
> My applogies for not having looked for this kind of breakage in the new |
| 13 |
> eclass before now. The current git eclass finally got the submodules- |
| 14 |
> vs-normal stuff worked out some time ago; the possibility of going |
| 15 |
> backwards never occurred to me.... ☹ |
| 16 |
> |
| 17 |
> As someone who makes heavy use of live ebuilds, someone who will be |
| 18 |
> directly and severely affected by such a change, I have to beg you |
| 19 |
> to keep the current logic for submodule-less repos. |
| 20 |
|
| 21 |
I was discussing a couple of ideas on IRC with scarabeus yesterday but |
| 22 |
haven't had a chance to look into them in more detail yet: |
| 23 |
|
| 24 |
- Providing automatic migration from the old system to the new. This is |
| 25 |
the simplest approach to deal with your specific problem but still |
| 26 |
leaves separate codepaths for submodule and non-submodule repos. |
| 27 |
|
| 28 |
- Handling submodules in bare checkouts. Perhaps we could detect whether |
| 29 |
submodules are in use (need to find the right place/way in git), then |
| 30 |
grab them as separate bare checkouts that would eventually be cloned |
| 31 |
into TMPDIR by changing the repo location git looks for (again, need to |
| 32 |
sort out how in git). |
| 33 |
|
| 34 |
-- |
| 35 |
Thanks, |
| 36 |
Donnie |
| 37 |
|
| 38 |
Donnie Berkholz |
| 39 |
Sr. Developer, Gentoo Linux |
| 40 |
Blog: http://dberkholz.com |
Archives