1 |
On Sun, 2020-07-26 at 14:02 +0200, Marek Szuba wrote: |
2 |
> Dear all, |
3 |
> |
4 |
> sword-module.eclass needs some changes, not in the least because it |
5 |
> assumes versioned source files are uploaded directly to the Gentoo |
6 |
> mirror network. The complete list of proposed changes is as follows: |
7 |
> |
8 |
> - add eclassdoc blocks; |
9 |
> - add the optional variable SWORD_MINIMUM_VERSION to specify the lowest |
10 |
> version of app-text/sword supported by the module at hand; |
11 |
> - check if SWORD_MODULE is set and not null, and die otherwise; |
12 |
> - change the default SRC_URI from mirror://gentoo to my Web space on |
13 |
> dev.gentoo.org; |
14 |
> - remove redundant declarations of HOMEPAGE and IUSE; |
15 |
> - app-arch/unzip is now in BDEPEND rather than DEPEND; |
16 |
> - as a consequence of the above, enforce the use of EAPI-7 in ebuilds |
17 |
> inheriting this eclass; |
18 |
> - remove redundant references to ${S} from doins() calls. |
19 |
> |
20 |
> A diff is attached. |
21 |
> |
22 |
> This is the first time I change anything other than documentation in any |
23 |
> eclass so I would very, very much appreciate your comments. In |
24 |
> particular, do we really need a revision bump? I initially decided to |
25 |
> err on the side of caution because I would rather not have the SRC_URI |
26 |
> change break all the app-dicts/sword-* modules, that said Portage |
27 |
> usually (always?) queries Gentoo mirrors before attempting to fetch |
28 |
> directly from SRC_URI so maybe it is harmless after all? |
29 |
> |
30 |
|
31 |
Please use git-send-email, so that we can reply in context without |
32 |
jumping through hoops. |
33 |
|
34 |
-- |
35 |
Best regards, |
36 |
Michał Górny |