1 |
Dnia 2014-05-05, o godz. 11:02:33 |
2 |
Ulrich Mueller <ulm@g.o> napisał(a): |
3 |
|
4 |
> >>>>> On Mon, 5 May 2014, Michał Górny wrote: |
5 |
> |
6 |
> > Three quick patches for review: |
7 |
> |
8 |
> > 1. adds multilib_get_enabled_abi_pairs() as a replacement for |
9 |
> > multilib_get_enabled_abis(). The latter returned just the value |
10 |
> > of ${ABI}, the new function returns ${use_flag}:${ABI} pairs. |
11 |
> |
12 |
> > e.g.: |
13 |
> |
14 |
> > multilib_get_enabled_abis: x86 amd64 |
15 |
> > multilib_get_enabled_abi_pairs: abi_x86_32:x86 abi_x86_64:amd64 |
16 |
> |
17 |
> These will get included in the pathname of the build dir, right? |
18 |
> |
19 |
> If yes, are you sure that all upstream build systems can cope with |
20 |
> the colon in path names? I'd rather suggest to stay within the POSIX |
21 |
> portable filename character set (i.e. [A-Za-z0-9._-]). |
22 |
|
23 |
I was wondering about that in the morning but thought mostly of Windows |
24 |
and decided we don't need to care. But now that I think of it, I can |
25 |
think of ${PATH} and similar colon-separated variables. |
26 |
|
27 |
I should probably use '.' then, since it's both more distinct than '-' |
28 |
and disallowed in USE flags. Possibly stripping ':*' in multibuild could |
29 |
result in more elegant paths but I don't want to introduce any other |
30 |
special rules to follow. |
31 |
|
32 |
Which makes me think that I need to check that no ebuild assumes |
33 |
${S%/}-${ABI}. |
34 |
|
35 |
-- |
36 |
Best regards, |
37 |
Michał Górny |