| 1 |
On Wed, 2 May 2018 17:42:19 +0200 |
| 2 |
Michał Górny <mgorny@g.o> wrote: |
| 3 |
|
| 4 |
> Now that EAPI 7 is accepted and implemented in Portage, provide |
| 5 |
> the initial environment blacklist for coming EAPI 7 ebuilds. The list |
| 6 |
> is based on existing eclasses, xdg-utils mostly. Rationale is provided |
| 7 |
> in the comment above ENV_UNSET. |
| 8 |
|
| 9 |
It may be worth adding the values from perl-functions.eclass's |
| 10 |
perl_check_env() test, but I'd have to get a second opinion from |
| 11 |
dilfridge. |
| 12 |
|
| 13 |
The only real downside of leveraging this approach is if there's some |
| 14 |
unforseen but justifiable excuse to let these ENV vars blead |
| 15 |
*intentionally*, it might be a bit of work to subvert. |
| 16 |
|
| 17 |
( reading EAPI7's PMS docs says its a non-aggregated variable and none |
| 18 |
of the standard approaches like ENV_UNSET=" -FOO" are documented to |
| 19 |
work ) |
Archives