| 1 |
>>>>> On Tue, 22 Jul 2014, Martin Vaeth wrote: |
| 2 |
|
| 3 |
> PF has to be filled correctly, of course: |
| 4 |
> The versions foo-1 and foo-1-r0 are identical according to PMS |
| 5 |
> and should thus lead to the same $PF. |
| 6 |
|
| 7 |
This is not so. These versions are equal in comparision, so they |
| 8 |
cannot be in the tree at the same time. However, PF will be different |
| 9 |
for them. |
| 10 |
|
| 11 |
Note that this also occurs elsewhere. For example, versions 1.0.2 and |
| 12 |
1.000.2 cound as equal, but again PF will be different. |
| 13 |
|
| 14 |
> If this is currently not happening in portage, this is a bug |
| 15 |
> which should be fixed, independently of whether minor revisions |
| 16 |
> are introduced or not. |
| 17 |
|
| 18 |
Portage behaves as PMS specifies. PF is derived from the ebuild's |
| 19 |
name, without any "normalisation" to a canonical version. |
| 20 |
|
| 21 |
Ulrich |
Archives