| 1 |
Michał Górny <mgorny@g.o> wrote: |
| 2 |
>> Maybe this could be solved by having two kinds of revisions: |
| 3 |
>> - One would rebuild all as usually (for example, -r1...) |
| 4 |
>> - The other one would only regenerate VDB and wouldn't change the |
| 5 |
>> installed files (for example, -r1.1) |
| 6 |
> |
| 7 |
> I'm afraid it couldn't. The major problem is not knowing *when* to |
| 8 |
> migrate metadata, portage usually gets that right. The problem is in |
| 9 |
> getting the correct output which is often near to impossible. |
| 10 |
|
| 11 |
Could you explain where you see here a problem with -r1.1 |
| 12 |
which is not caused as well with -r2? |
| 13 |
|
| 14 |
The only difference should be that when revbumping -r1 to -r1.1 |
| 15 |
there is actually no recompilation done (and perhaps the |
| 16 |
PF and PR variables are treated differently) - everything else |
| 17 |
should be exactly the same as for current revbumps. |
| 18 |
|
| 19 |
And once more, this is only one of the several possibilities |
| 20 |
how to tell portage that actually no compilation is necessary: |
| 21 |
Some other metadata/variable/whatever might be used as well. |
| 22 |
|
| 23 |
The idea is to act "as usual", just to skip unnecessary phases... |
Archives