1 |
On Fri, Aug 24, 2018 at 10:37 AM Virgil Dupras <vdupras@g.o> wrote: |
2 |
> |
3 |
> On Fri, 24 Aug 2018 10:27:01 -0400 |
4 |
> Mike Gilbert <floppym@g.o> wrote: |
5 |
> |
6 |
> > On Fri, Aug 24, 2018 at 9:23 AM Kent Fredric <kentnl@g.o> |
7 |
> > wrote: |
8 |
> > > |
9 |
> > > On Tue, 21 Aug 2018 22:29:29 -0400 |
10 |
> > > Mike Gilbert <floppym@g.o> wrote: |
11 |
> > > |
12 |
> > > > Setting RESTRICT="!test? ( test )" is generally sufficient. |
13 |
> > > |
14 |
> > > But that would require setting that virtually *everything* that has |
15 |
> > > both tests, and required dependencies for tests. |
16 |
> > > |
17 |
> > > Which, in my experience, is practically everything with tests. |
18 |
> > > |
19 |
> > > To the point it seems like that should be the *default* mechanic, |
20 |
> > > not a requirement that everyone pay not to have a randomly broken |
21 |
> > > package. |
22 |
> > |
23 |
> > If you want to define behavior that can be relied upon in ebuilds, it |
24 |
> > should be specified in PMS. PMS does not define any meaning for the |
25 |
> > "test" USE flag. |
26 |
> > |
27 |
> |
28 |
> Which is the easiest path, updating the PMS or adding RESTRICT="!test? |
29 |
> ( test )" to thousands of ebuilds? I don't see how we can realistically |
30 |
> hope for every developer to cooperate in making sure that their ebuilds |
31 |
> behave properly in "USE=-test" situation. |
32 |
|
33 |
Updates to PMS happen infrequently, and generally only introduce |
34 |
behavioral changes in new EAPIs. |
35 |
|
36 |
Adding RESTRICT to ebuilds does not need to happen overnight; setting |
37 |
FEATURES=test still does the right thing for most people, assuming |
38 |
they haven't done something stupid like setting USE=-test in |
39 |
make.conf. |