| 1 |
>>>>> On Mon, 29 May 2017, Michał Górny wrote: |
| 2 |
|
| 3 |
> On pon, 2017-05-29 at 20:00 +0200, Alexis Ballier wrote: |
| 4 |
>> Can you provide an efficient algorithm for the above syntax? That |
| 5 |
>> is, given a set of +/- useflags forced by user, output the set of |
| 6 |
>> effective useflags (or a rant if it is inconsistent). |
| 7 |
|
| 8 |
> I'd rather leave that to people who are good with algorithms. I find |
| 9 |
> the whole thing scary but I don't really see a sane alternative here. |
| 10 |
> Worst case, we have to figure out some arbitrary limitations to keep |
| 11 |
> things sane. |
| 12 |
|
| 13 |
IMHO the sanest alternative would be to restrict the syntax to USE |
| 14 |
conditional forms which have an obvious solution. One of the many |
| 15 |
problems of REQUIRED_USE is that it sometimes requires solving a |
| 16 |
Zebra Puzzle. |
| 17 |
|
| 18 |
Also, can we find a better name? Sorry for the bikeshedding at this |
| 19 |
early stage, but I believe that ENFORCED_USE can be easily confused |
| 20 |
with use.force in profiles. MAPPED_USE? USE_MAP? |
| 21 |
|
| 22 |
Ulrich |
Archives