1 |
On Sat, 20 Sep 2008 22:05:43 +0300 |
2 |
Petteri Räty <betelgeuse@g.o> wrote: |
3 |
|
4 |
> Alexis Ballier kirjoitti: |
5 |
> > Hi, |
6 |
> > |
7 |
> >> When EAPI 2 goes live built_with_use should probably die for most |
8 |
> >> cases. |
9 |
> > |
10 |
> > I don't understand here: you mean die like being removed or die like |
11 |
> > the die call in an ebuild? If I understood correctly the following |
12 |
> > it should be the latter. |
13 |
> > |
14 |
> |
15 |
> Well we could go with either but if there are valid use cases we need |
16 |
> to go with the latter. |
17 |
> |
18 |
> >> Are there valid use cases for built_with_use that are not |
19 |
> >> covered by the use deps in EAPI 2? |
20 |
> > |
21 |
> > I can think of checks like: |
22 |
> > - foo is a dep/rdep of bar |
23 |
> > - foo has a "plugin like" architecture |
24 |
> > - bar will "work" with minimal foo |
25 |
> > - most people will expect some features in bar that come with foo's |
26 |
> > plugins |
27 |
> > - we might want to display warnings for the most useful features |
28 |
> > - having useflags in bar for each of foo's useflags doesn't seem |
29 |
> > wise |
30 |
> > |
31 |
> > |
32 |
> > Ok that's not really a nice example but I can't think of anything |
33 |
> > better at the moment. |
34 |
> > |
35 |
> > By the way, how will the --missing option of built_with_use be |
36 |
> > handled by eapi 2? |
37 |
> > |
38 |
> > Alexis. |
39 |
> |
40 |
> Why should it be any different than what it is now? |
41 |
|
42 |
i still don't get what you're saying. using built_with_use should die |
43 |
as in cause the ebuild to exit with an error code? or that we should |
44 |
stop using it for checking use flags? |
45 |
|
46 |
there are plenty of reasons for using built_with_use that have nothing |
47 |
to do with dependencies. i use it in the wxwidgets eclass to determine |
48 |
what possible configuration tuple(s) the user can have installed. |
49 |
toolchain.eclass uses it for some cross-compile logic. in many cases |
50 |
it's simply used to display elog messages conditionally. |
51 |
|
52 |
so, please don't make it "die" or i will be ":(" |
53 |
|
54 |
|
55 |
-- |
56 |
gcc-porting, by design, by neglect |
57 |
treecleaner, for a fact or just for effect |
58 |
wxwidgets @ gentoo EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662 |