| 1 |
>>>>> On Mon, 20 Mar 2017, Alexis Ballier wrote: |
| 2 |
|
| 3 |
> What makes me wonder more are the proposed solutions: So far the |
| 4 |
> only proposals I've seen are either inlining *all* the code or |
| 5 |
> moving *all* the code into an eclass. Having a quick look at |
| 6 |
> autoconf, it seems to me an intermediate solution would work |
| 7 |
> perfectly fine for the above goals/rules: Put main.eblit into an |
| 8 |
> eclass. The loading code then would access $FILESDIR only in src_* |
| 9 |
> phases. This would likely work better for all parties and would |
| 10 |
> allow to focus on better specifying this gray area of PMS instead. |
| 11 |
|
| 12 |
But is it desirable as a goal, that all packages in the tree use |
| 13 |
regular eclasses, but two packages (autoconf and glibc) use something |
| 14 |
else that is a "grey area"? |
| 15 |
|
| 16 |
Also, can somebody please point me to the discussion that preceded the |
| 17 |
introduction of eblits. AFAICS they appeared sometime in 2007, but I |
| 18 |
cannot find anything in our mailing list archives. |
| 19 |
|
| 20 |
Ulrich |
Archives