| 1 |
On Tue, Apr 29, 2003 at 05:41:39PM +0200, duchier@×××××××××.de wrote: |
| 2 |
> Patrick Kursawe <phosphan@g.o> writes: |
| 3 |
> |
| 4 |
> >> Calculating world dependencies ...done! |
| 5 |
> >> [ebuild UD] app-text/tetex-1.0.7-r12 [2.0-r1] |
| 6 |
> > |
| 7 |
> > Tetex 2.0-r1 has been removed from portage. It looks like emerge |
| 8 |
> > tries to be on the safe side and rather downgrade than keep |
| 9 |
> > a version it does not know about. If you upgrade to tetex-2.0.2, |
| 10 |
> > emerge -U should no longer try to downgrade. |
| 11 |
> |
| 12 |
> Thank you very much! That explains it and your suggestion fixed the |
| 13 |
> problem. However, is this really the _right_ behaviour? |
| 14 |
> |
| 15 |
> o portage actually knows something about tetex 2.0-r1 since it is |
| 16 |
> emerged |
| 17 |
> |
| 18 |
> o --upgradeonly promises to not downgrade: regardless of whether |
| 19 |
> there is currently an ebuild for the installed version, the |
| 20 |
> version is known, and IMHO that fixes the semantics of |
| 21 |
> downgrading |
| 22 |
> |
| 23 |
> o maybe --pretend could provide an indication that no ebuild for |
| 24 |
> the currently installed version can be found |
| 25 |
> |
| 26 |
> Comments? |
| 27 |
|
| 28 |
Nicholas, are you following this thread? |
| 29 |
|
| 30 |
Patrick |
Archives