1 |
I forgot to mention that (a) is what the Ruby team has been doing up |
2 |
to now -- it feels a bit more cumbersome in some cases, but it's |
3 |
definitely easier to spot the problems from the start than finding |
4 |
them months after adding the package of the tree. |
5 |
|
6 |
Especially if you change your mind and decide that you want to add the |
7 |
dependency _after_ the package has been keyworded by half the arches |
8 |
out there. |
9 |
Diego Elio Pettenò — Flameeyes |
10 |
flameeyes@×××××××××.eu — http://blog.flameeyes.eu/ |
11 |
|
12 |
|
13 |
On Sun, Jan 6, 2013 at 1:50 PM, hasufell <hasufell@g.o> wrote: |
14 |
> I agree with "a". |
15 |
> A problem with "b" is: the user might install one of those "optional |
16 |
> dependencies" later, but that will not trigger a rebuild of the other |
17 |
> package and another run through the test phase. |
18 |
> I would find "c" a bit confusing. |
19 |
> |
20 |
> The most elegant way would probably be to trigger a remerge of package |
21 |
> a, when you want to emerge package b which is also an optional |
22 |
> dependency of package a (in case package a has a test phase ofc). But I |
23 |
> don't see a clean and easy way to do that. |
24 |
> |
25 |
> On 01/06/2013 01:28 PM, Diego Elio Pettenò wrote: |
26 |
>> Go for a. The widest and more consistent the testing, the better. |
27 |
>> |
28 |
>> Otherwise the day after tomorrow you'll get a bug from me that with |
29 |
>> $foo installed, $bar fails tests. |
30 |
>> Diego Elio Pettenò — Flameeyes |
31 |
>> flameeyes@×××××××××.eu — http://blog.flameeyes.eu/ |
32 |
>> |
33 |
>> |
34 |
>> On Sun, Jan 6, 2013 at 11:38 AM, Michał Górny <mgorny@g.o> wrote: |
35 |
>>> Hello, |
36 |
>>> |
37 |
>>> There are some Python packages which have a bunch of optional tests |
38 |
>>> utilizing external packages. For example, the dev-python/logilab-common |
39 |
>>> runs a few additional tests if dev-python/egenix-mx-base is installed; |
40 |
>>> if the package is not installed, it just skips those tests. |
41 |
>>> |
42 |
>>> Those tests can't be really considered 'heavy' or in any way suggesting |
43 |
>>> use of an additional USE flag. |
44 |
>>> |
45 |
>>> Do you believe that the ebuilds should: |
46 |
>>> |
47 |
>>> a) depend on all optional test dependencies conditionally to USE=test, |
48 |
>>> therefore always requesting the widest (and consistent) testing, |
49 |
>>> |
50 |
>>> b) not depend on the optional test dependencies, resulting in less |
51 |
>>> dependencies for most users but also a bit inconsistent test |
52 |
>>> experience, |
53 |
>>> |
54 |
>>> c) put the optional test dependencies behind an additional USE flag? |
55 |
>>> |
56 |
>>> -- |
57 |
>>> Best regards, |
58 |
>>> Michał Górny |
59 |
>> |
60 |
> |
61 |
> |