| 1 |
On 3/6/12 11:17 PM, Thomas Kahle wrote: |
| 2 |
> the x86 team has many ruby keywording bugs in the queue and it is |
| 3 |
> hard to keep up because testing them can be a pain. There are tons |
| 4 |
> of circular deps with USE="test" or USE="doc", there are the |
| 5 |
> different ruby interpreters. |
| 6 |
|
| 7 |
Yeah. Maintainers, if you get delays on arch-related bugs it's often |
| 8 |
because of pain like this. |
| 9 |
|
| 10 |
> Also the inter-bug dependencies are often not resolved correctly, |
| 11 |
> that is the to be keyworded package depends on non-keyworded stuff |
| 12 |
> not listed in the bug. |
| 13 |
|
| 14 |
And this is even worse. Please check things with repoman before filing |
| 15 |
bugs. You can even write automated scripts at least for the "check |
| 16 |
whether we got all deps right" part. |
| 17 |
|
| 18 |
> All of these packages are already keyworded ~amd64. Ruby is an |
| 19 |
> interpreted language, I don't see any point in having every arch team |
| 20 |
> do the testing for every small package. Could the ruby team add ~x86 |
| 21 |
> themselves after testing on ~amd64, or are there compelling reasons |
| 22 |
> to not do this? |
| 23 |
|
| 24 |
It's trivial to set up x86 chroot on amd64 box, so I can't imagine |
| 25 |
what's preventing people from creating such chroot, doing the testing |
| 26 |
and keywording themselves. |
Archives