1 |
On Sun, 12 Apr 2020 10:43:07 +0200 |
2 |
Agostino Sarubbo <ago@g.o> wrote: |
3 |
|
4 |
> If you work on the stabilization workflow you may have noticed that: |
5 |
> |
6 |
> - There are people that rant if you don't run src_test against their packages; |
7 |
> - There are people that rant if you open a test failure bug against their |
8 |
> packages and you block the stabilization. |
9 |
> |
10 |
> So, unless there will be a clear policy about that, I'd suggest to introduce a |
11 |
> way to establish if a package is expected to pass src_test or not. |
12 |
> |
13 |
> A simple way to achieve this goal would be: |
14 |
> introduce a new bugzilla custom flag (like "Runtime testing required" we |
15 |
> already have) maybe called "src_test pass" or similar, that, by default is yes |
16 |
> and the maintainer should set it to no when needed. |
17 |
> |
18 |
> In case of 'yes', the arch team member must compile with FEATURE="test" and he |
19 |
> is allowed to block the stabilization in case of test-failure. |
20 |
> |
21 |
> In case there will be a test-failure there are two choices: |
22 |
> 1) The maintainer fixes the test failure; |
23 |
> 2) The maintainer does not want to take care, so he can simply remove the |
24 |
> blocker and set "src_test pass" to no. |
25 |
> |
26 |
> Both of the above are fine for me. |
27 |
> |
28 |
> Obviously, if there are already test-failure bugs open, the flag "src_test |
29 |
> pass" should be set to 'no' when the stabilization bugs is filled. |
30 |
> |
31 |
> I'm sure that this way would help to avoid wasting time on both sides. |
32 |
|
33 |
Expecting the package to be marked stable while the tests fail seems |
34 |
like a very odd stance to take. I would politely request that you flag |
35 |
this up and if you get pushback but understandably don't want to fight |
36 |
it then just add RESTRICT="test" yourselves, maybe filtered on a |
37 |
particular USE flag if it only happens in specific situations. I think |
38 |
the QA team would back you up here? |
39 |
|
40 |
I understand the comments about only a few minor tests failing but if |
41 |
the maintainer doesn't want to skip the entire suite then it is their |
42 |
responsibility to limit the set of tests run by whatever means. |
43 |
Unfortunately there is no universal method of doing this. |
44 |
|
45 |
-- |
46 |
James Le Cuirot (chewi) |
47 |
Gentoo Linux Developer |