1 |
On Fri, 4 Nov 2005 01:19:35 +0900 |
2 |
Jason Stubbs <jstubbs@g.o> wrote: |
3 |
|
4 |
> package.env would be a list of "<atom> <file> [<file> ...]" |
5 |
... |
6 |
> With a couple of small modifications to emerge to check FEATURES |
7 |
> for "buildpkg" after the call to setcpv() is done rather than |
8 |
> doing it once globally, this would also cover TGL's BUILD_PKGS |
9 |
> addition too. |
10 |
|
11 |
Not if env files are only selectable by strict depatoms. I mean, |
12 |
sure this syntax is perfect for *DEPEND strings, and is fine for |
13 |
package.{use,mask,unmask} (although the randomness of |
14 |
best_match_to_list() is rather annoying). But for package.keywords, |
15 |
it is already sub-optimal imho (i run ~arch so i don't care |
16 |
much, but for instance i often see people on forums who list |
17 |
some whole categories there), and it would be too for the generic |
18 |
package.env. For instance, people who develop gnome stuffs might |
19 |
want to use the debug env for gnome-*/* packages. As for |
20 |
the buildpkg FEATURES flag, it would be a real pita if i had to list |
21 |
all packages matched by my current BUILD_PKGS spec (just look at |
22 |
the examples i've put in my email on that topic to get the idea). |
23 |
|
24 |
Since being able to list several env file on a same line doesn't |
25 |
sounds like a must have feature to me, i would much prefer a |
26 |
package.env format of that kind: |
27 |
<rule> [<rule> ...] <envfile> |
28 |
where <rule> would be similar to what i've defined for BUILD_PKGS |
29 |
(with addition of full versioned dep atoms, which is a trivial |
30 |
change to my code). And if a package happens to match the rules |
31 |
lists of several lines, then the corresponding env files would all |
32 |
be sourced, in the order of the said lines. I can try to implement |
33 |
that if you agree on the idea. |
34 |
|
35 |
|
36 |
My second concern is about unsupported variables (some of the |
37 |
FEATURES flags, some of the *DIR locations, etc.): i think they |
38 |
should be somehow blacklisted, to avoid crazy breakages/bugreports |
39 |
(btw, a quick look on the variables listed in make.conf.example |
40 |
made me realize it was not that easy to write an accurate |
41 |
blacklist, which tends to confirm it's really needed). |
42 |
|
43 |
-- |
44 |
TGL. |
45 |
-- |
46 |
gentoo-portage-dev@g.o mailing list |