| 1 |
On Mon, 2020-03-23 at 13:23 -0500, William Hubbs wrote: |
| 2 |
> Hey all, |
| 3 |
> |
| 4 |
> it has been brought to my attention that there have been several |
| 5 |
> backward-incompatible changes made to the python eclasses lately. |
| 6 |
> |
| 7 |
> It is true that everything in ::gentoo has been fixed along with the |
| 8 |
> changes to the eclasses; however, when a change like this goes into a |
| 9 |
> widely used eclass it breaks overlays with little to no notice; |
| 10 |
> especially since we do not require developers to be subscribed to this |
| 11 |
> mailing list. |
| 12 |
> |
| 13 |
> I do agree that overlay authors are on their own to fix things, but we need to |
| 14 |
> find a way to notify them when a breaking change is going into a widely |
| 15 |
> used eclass and give them time to adjust their ebuilds. |
| 16 |
> |
| 17 |
> If the old way of doing things cannot be supported |
| 18 |
> along side the new way the correct path forward is a new version of the |
| 19 |
> eclass then a lastrites on the old version. That would give overlay |
| 20 |
> authors time to switch to the new eclass. |
| 21 |
> |
| 22 |
> If the old and new way can be supported in the same code base, a |
| 23 |
> reasonable way forward is to allow both ways to exist while ::gentoo is |
| 24 |
> migrated to the new code path then do the equivalent of a lastrites for |
| 25 |
> the old code path so overlay authors can adjust their ebuilds. |
| 26 |
> |
| 27 |
> Thoughts? |
| 28 |
> |
| 29 |
> William |
| 30 |
> |
| 31 |
|
| 32 |
All of this was announced with a 3 month timeout, using the right channels. Have |
| 33 |
you checked all python-related eclasses changes submitted to the ML? In this |
| 34 |
case, eqawarn would not have been possible, because the change involved |
| 35 |
dereferencing a variable. |
| 36 |
|
| 37 |
Check the git-2 debacle: 6.5 years of deprecation, and still a bunch of overlays |
| 38 |
exploded. There comes a point where you just have to suck it up and move on. |
Archives