1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA1 |
3 |
|
4 |
Marius Mauch wrote: |
5 |
>>> "Marijn Schouten (hkBst)" <hkBst@g.o>: |
6 |
>>>> I imagine there are a lot more cases where the simple on/off |
7 |
>>>> system we have now is suboptimal. I could be wrong of course. |
8 |
>>>> Please comment. |
9 |
>>> This key=value systems sounds interesting. Another use could be |
10 |
>>> the choice between xulrunner, seamonkey, firefox. |
11 |
> |
12 |
> We already have this with USE_EXPAND. Not exactly the same syntax, but |
13 |
> I don't see a terrible problem in that, and we don't have to fix all |
14 |
> three trillion related tools to handle it. Unless you can come up |
15 |
> with a case that can't be handled with USE_EXPAND. |
16 |
|
17 |
No, USE_EXPAND is only a way to abbreviate use flags with a common substring |
18 |
in their name, such as "impl_guile impl_sbcl impl_clisp" which could be |
19 |
encoded interchangeably as either |
20 |
|
21 |
# without USE_EXPAND |
22 |
IUSE="impl_guile impl_sbcl impl_clisp" |
23 |
|
24 |
or |
25 |
|
26 |
# with USE_EXPAND |
27 |
for impl in IMPL; do IUSE+="impl_${impl}"; done |
28 |
|
29 |
but the effect is that there are 3 use flags with a total of 2^3=8 |
30 |
combinations, while really something with exactly 3 combinations is needed: |
31 |
|
32 |
IUSE="impl" |
33 |
|
34 |
case ${impl} in |
35 |
guile) #use guile as backend |
36 |
sbcl) #use sbcl as backend |
37 |
clisp) #use clisp as backend |
38 |
|
39 |
or something like that. |
40 |
|
41 |
Marijn |
42 |
|
43 |
- -- |
44 |
Marijn Schouten (hkBst), Gentoo Lisp project |
45 |
<http://www.gentoo.org/proj/en/lisp/>, #gentoo-lisp on FreeNode |
46 |
-----BEGIN PGP SIGNATURE----- |
47 |
Version: GnuPG v2.0.7 (GNU/Linux) |
48 |
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org |
49 |
|
50 |
iD8DBQFHChhAp/VmCx0OL2wRArxWAKCWfciDl5XihPOoiI/01J3DjGGpqgCdFJxV |
51 |
9n89OMcqxqD4JqFTPDGt12o= |
52 |
=njyU |
53 |
-----END PGP SIGNATURE----- |
54 |
-- |
55 |
gentoo-dev@g.o mailing list |