| 1 |
On Sun, Aug 26, 2018 at 3:28 AM Zac Medico <zmedico@g.o> wrote: |
| 2 |
> |
| 3 |
> On 08/25/2018 07:11 PM, Andrew Savchenko wrote: |
| 4 |
> > On Sat, 25 Aug 2018 14:24:02 -0400 Mike Gilbert wrote: |
| 5 |
> >> On Sat, Aug 25, 2018 at 1:41 AM Andrew Savchenko <bircoph@g.o> wrote: |
| 6 |
> >>> |
| 7 |
> >>> On Fri, 24 Aug 2018 14:24:06 -0400 Mike Gilbert wrote: |
| 8 |
> >>>> --- |
| 9 |
> >>>> general-concepts/dependencies/text.xml | 38 ++++++++++++++++++++++++++ |
| 10 |
> >>>> 1 file changed, 38 insertions(+) |
| 11 |
> >>>> |
| 12 |
> >>>> diff --git a/general-concepts/dependencies/text.xml b/general-concepts/dependencies/text.xml |
| 13 |
> >>>> index 2f10380..64be9dc 100644 |
| 14 |
> >>>> --- a/general-concepts/dependencies/text.xml |
| 15 |
> >>>> +++ b/general-concepts/dependencies/text.xml |
| 16 |
> >>>> @@ -578,6 +578,44 @@ valid. |
| 17 |
> >>>> </body> |
| 18 |
> >>>> </section> |
| 19 |
> >>>> |
| 20 |
> >>>> +<section> |
| 21 |
> >>>> +<title>Test Dependencies</title> |
| 22 |
> >>>> +<body> |
| 23 |
> >>>> + |
| 24 |
> >>>> +<p> |
| 25 |
> >>>> +Packages often have optional dependencies that are needed only when running |
| 26 |
> >>>> +tests. These should be specified in DEPEND behind a USE flag. Often, the |
| 27 |
> >>>> +'test' USE flag is used for this purpose. |
| 28 |
> >>>> +</p> |
| 29 |
> >>>> + |
| 30 |
> >>>> +<p> |
| 31 |
> >>>> +Since testing will likely fail when test dependencies are not installed, the |
| 32 |
> >>>> +test phase should be disabled in this case. This may be accomplished via USE |
| 33 |
> >>>> +conditionals in the RESTRICT variable. |
| 34 |
> >>>> +</p> |
| 35 |
> >>>> + |
| 36 |
> >>>> +<p> |
| 37 |
> >>>> +If other optional features must be enabled/disabled when testing, REQUIRED_USE |
| 38 |
> >>>> +may be set to express this. |
| 39 |
> >>>> +</p> |
| 40 |
> >>>> + |
| 41 |
> >>>> +<codesample lang="ebuild"> |
| 42 |
> >>>> +# Define some USE flags |
| 43 |
> >>>> +IUSE="debug test" |
| 44 |
> >>>> + |
| 45 |
> >>>> +# Disable test phase when test USE flag is disabled |
| 46 |
> >>>> +RESTRICT="!test? ( test )" |
| 47 |
> >>> |
| 48 |
> >>> I do not understand why we need this useless code. If test USE flag |
| 49 |
> >>> is disabled, tests must be disabled as well. It is PM's job and |
| 50 |
> >>> there is no need to put this obvious stuff into each ebuild with |
| 51 |
> >>> tests and extra deps. I see no reason to support running src_test() |
| 52 |
> >>> with USE="-test". |
| 53 |
> >> |
| 54 |
> >> PMS does not specify that behavior (skipping src_test with USE=-test). |
| 55 |
> >> It is better to define the requrement explicitly rather than relying |
| 56 |
> >> on a Portage-specific behavior. |
| 57 |
> > |
| 58 |
> > Then PMS should be fixed. Putting useless code in thousands |
| 59 |
> > of ebuilds due to bureaucratic reasons is ridiculous. Having strict |
| 60 |
> > conformance to the PMS is good, but common sense should still be |
| 61 |
> > considered. |
| 62 |
> |
| 63 |
> Since PMS doesn't specify the behavior of FEATURES, I suppose we could |
| 64 |
> make FEATURES=test imply RESTRICT="!test? ( test )". Would there be any |
| 65 |
> drawbacks to that? |
| 66 |
|
| 67 |
I think that's a nice idea from a usability standpoint, at least while |
| 68 |
most ebuilds in the tree do not set it explicitly. |
| 69 |
|
| 70 |
I still think adding an explicit RESTRICT="!test? ( test )" to ebuilds |
| 71 |
is a good idea though. My fear is that certain people would start |
| 72 |
rejecting patches or reverting commits as "useless changes". |
Archives