| 1 |
Roy Bamford wrote: |
| 2 |
> What about the case where the new EAPI breaks backwards compatibility |
| 3 |
> with existing package managers, as would be the case with glep 55? |
| 4 |
> |
| 5 |
> Its quite true that such changes can be introduced after a wait and |
| 6 |
> only upset late adoptors. By implementing the key feature of glep 55, |
| 7 |
> which is making the EAPI known to the PM without sourcing the ebuild, |
| 8 |
> we would only need one last wait to introduce new features in this |
| 9 |
> way in later EAPIs.PMs would then know the EAPI in advance and ignore |
| 10 |
> ebuilds using EAPIs they don't understand. |
| 11 |
|
| 12 |
But still then the special case I mentioned comes in. Newer version |
| 13 |
migrated to newer EAPI. Urgent bump for security reasons is necessary. |
| 14 |
Backporting the ebuild is necessary. Not that likely, but iirc we had |
| 15 |
that special case for EAPI-2. |
| 16 |
|
| 17 |
Putting in a wait for 4 or 8 weeks or whatever doesn't cost us anything |
| 18 |
but does simplify things and gives us a clear deployment process. |
| 19 |
|
| 20 |
I do know we have developers with nil interest in our stable branch, but |
| 21 |
we also have users heavily relying on our stable branch. |
| 22 |
|
| 23 |
Tobias |
Archives