1 |
On Thu, Nov 15, 2012 at 10:44 PM, Michał Górny <mgorny@g.o> wrote: |
2 |
> Of course, a MIP would not be allowed to depend on a SIP. |
3 |
|
4 |
How big of a problem would this be? |
5 |
|
6 |
> To sum it up quickly, the aux variable solution: |
7 |
> |
8 |
> Advantages: |
9 |
> - painless for most of our users, |
10 |
> - explicit, transparent, |
11 |
> - simple and matching SIP->SIP and SIP->MIP deps. |
12 |
> |
13 |
> Disadvantages: |
14 |
> - a bit redundant, |
15 |
|
16 |
Given the requirement for enabling multiple implementation in |
17 |
PYTHON_TARGETS, which I feel strongly about, there has to be a choice |
18 |
which one of the implementations will be target for SINGLE_TARGET |
19 |
packages. Adding a variable for that, therefore, doesn't seem |
20 |
redundant. |
21 |
|
22 |
> - does not solve the issue for py3-only SIP packages, |
23 |
|
24 |
I guess this might become a problem in the future. Once we have both |
25 |
py2-only SIP packages and py3-only SIP packages, we have a bit of a |
26 |
problem. Maybe we should think more about that. |
27 |
|
28 |
> - may require explicit action when changing PYTHON_TARGETS. |
29 |
|
30 |
Yeah, I don't see a problem with this. In particular, it seems |
31 |
unlikely that many people will setup their system to exclude the |
32 |
default SINGLE_TARGET impl (2.7) from their TARGETS. |
33 |
|
34 |
Cheers, |
35 |
|
36 |
Dirkjan |