| 1 |
> Personally, I use it because it gives me extra checks that a package |
| 2 |
> is working correctly before I merge into my system and potentially |
| 3 |
> break things. Since I am mostly using non-x86 this becomes more |
| 4 |
> likely. |
| 5 |
|
| 6 |
So peace of mind is your motivation? |
| 7 |
|
| 8 |
> Btw, if we (as users) run into a package that fails maketest, is opening |
| 9 |
> a bug the right thing to do? |
| 10 |
|
| 11 |
I can't speak for all packages, but for GNOME packages, we'll take |
| 12 |
patches you can provide that fix "make test" errors, but we don't have |
| 13 |
time to look into fixing the tests ourselves. |
| 14 |
|
| 15 |
Unfortunately, "make test" isn't uniformly adhered to upstream, and in |
| 16 |
my experience, the tests that are conducted with it are often outdated, |
| 17 |
and/or unmaintained. In these cases, "make test" doesn't provide any |
| 18 |
indication of the state of the software - if it's |
| 19 |
runnable/testable/linkable/however the tests are conducted. |
| 20 |
|
| 21 |
This said, I'm not aware of (any?) Gentoo policy/stance on this, if |
| 22 |
anyone can fill me in on that, please do. |
| 23 |
|
| 24 |
Mike Gardiner |
| 25 |
(Obz) |
| 26 |
|
| 27 |
|
| 28 |
|
| 29 |
|
| 30 |
-- |
| 31 |
gentoo-dev@g.o mailing list |
Archives