| 1 |
On 01/20/2015 12:15 PM, Ciaran McCreesh wrote: |
| 2 |
> On Tue, 20 Jan 2015 12:12:32 -0800 |
| 3 |
> Zac Medico <zmedico@g.o> wrote: |
| 4 |
>>>> Regardless of whether or not (or how) we choose to apply |
| 5 |
>>>> REQUIRED_USE to various cases, I think we should keep REQUIRED_USE |
| 6 |
>>>> around, since having a machine-readable representation of these |
| 7 |
>>>> constraints can potentially be extremely useful to dependency |
| 8 |
>>>> resolvers. |
| 9 |
>>> |
| 10 |
>>> [evidence needed] |
| 11 |
>> |
| 12 |
>> REQUIRED_USE is already successfully used to apply constraints that |
| 13 |
>> would otherwise need to be enforced by executing shell code in |
| 14 |
>> pkg_pretend. I think it's obvious that the declarative and |
| 15 |
>> machine-readable nature of REQUIRED_USE make if preferable to using |
| 16 |
>> pkg_pretend in many cases. |
| 17 |
> |
| 18 |
> It's not obvious at all. The evidence so far suggests that good human |
| 19 |
> readable messages are important, and that if we're ever going to get |
| 20 |
> automated resolution then we'll need a new not-SAT-based format anyway. |
| 21 |
|
| 22 |
Nothing stops us from using pkg_pretend right now. Since REQUIRED_USE |
| 23 |
remains useful, I would suggest that we keep it around until we have |
| 24 |
implemented a replacement that allows for automated resolution. |
| 25 |
-- |
| 26 |
Thanks, |
| 27 |
Zac |
Archives