1 |
On 24/02/13 17:53, Michał Górny wrote: |
2 |
>> I still try to use plain ebuilds without |
3 |
>> inheritting autotools-utils.eclass as I usually don't need it, probably |
4 |
>> others do the same and refuse to have to inherit it only for multilib |
5 |
>> support :/ How do you plan to solve this problem? |
6 |
> |
7 |
> You generally have two options on doing multilib builds: either using |
8 |
> out-of-source builds or in-source builds. If you decide on the latter, |
9 |
> you unnecessarily waste users' time and disk space to create two more |
10 |
> copies of sources. I don't think we should go this way. |
11 |
> |
12 |
> If you decide on out-of-source builds, you basically need proper |
13 |
> src_{configure,compile,test,install} and that's what autotools-utils |
14 |
> does. Plus: |
15 |
> |
16 |
> - patch applying and autoreconf in src_prepare() -- which are |
17 |
> completely optional, you are free to write your own src_prepare(). |
18 |
> If you wanted to apply patches by hand, you'd need to write |
19 |
> src_prepare() anyway. |
20 |
|
21 |
It's that "Plus" part that is my problem with autotools-multilib.eclass |
22 |
currently, it adds EXPORT_FUNCTIONS of src_prepare() from |
23 |
autotools-utils.eclass which is irrelevant to the autotools-multilib.eclass |
24 |
adds just another eclass/phase function to worry about for inherit order |
25 |
|
26 |
> |
27 |
> - prune_libtool_files in src_install() which most people want to do |
28 |
> anyway, so that doesn't hurt -- and the pkg-config dep is going to |
29 |
> be removed, by the patch I sent already. |
30 |
|
31 |
but lacks a way to pass arguments to prune_libtool_files, like --all, |
32 |
since prune_libtool_files isn't that smart it gets it right everytime |
33 |
i propably prefer to stick to manually calling it with or without --all |
34 |
and well, this is not related to the multilib conversion so it shouldn't |
35 |
be executed anyway |
36 |
|
37 |
> - adding libtool args for shared/static libs if IUSE=static-libs -- |
38 |
> which I wanted to remove but people considered it useful. |
39 |
|
40 |
if it's not related to the multilib conversion, it shouldn't be executed... |
41 |
|
42 |
> |
43 |
>> I would also like to hear why that people refuses to use |
44 |
>> autotools-utils.eclass... because I don't have a strong opinion on this |
45 |
>> topic |
46 |
> |
47 |
> Well, the major argument was similar to yours -- why we should use |
48 |
> an eclass if default PMS functions work. But in the multilib case, they |
49 |
> do not work by design anymore. |
50 |
> |
51 |
|
52 |
src_prepare() seems to be adequate from PMS for the multilib conversion |