| 1 |
Dnia 2013-11-11, o godz. 13:38:56 |
| 2 |
Sergey Popov <pinkbyte@g.o> napisał(a): |
| 3 |
|
| 4 |
> 11.11.2013 13:32, Manuel Rüger wrote: |
| 5 |
> > Hi, |
| 6 |
> > |
| 7 |
> > I recently noticed it twice, that it seems to be common practice to |
| 8 |
> > remove a package without using the methods described in [1], but just |
| 9 |
> > dropping it from cvs. |
| 10 |
> > |
| 11 |
> > From my observations packages removed without last-rites could be |
| 12 |
> > characterized by this: |
| 13 |
> > |
| 14 |
> > - it was a dependency of another package |
| 15 |
> > - this package dropped / incorporated the dependency |
| 16 |
> > - no other packages depend on it |
| 17 |
> > - there are possible forks or updates, but maintainer doesn't care^W^W |
| 18 |
> > has no interest |
| 19 |
> |
| 20 |
> +1, this should be documented IMO. I last-rite |
| 21 |
> games-strategy/seven-kingdoms-data recently without sending notice, |
| 22 |
> cause last versions of games-strategy/seven-kingdoms includes all of |
| 23 |
> it's data. |
| 24 |
|
| 25 |
How hard would it be to send proper last rites for that package and add |
| 26 |
it to package.mask explaining the move? |
| 27 |
|
| 28 |
Silent removals do us no good. The only valid reason to remove |
| 29 |
a package without lastriting it is when it is package-moved with proper |
| 30 |
'updates' entry. However, that won't work for package merges, so |
| 31 |
the usual lastriting procedure applies. |
| 32 |
|
| 33 |
Overlays are just one of the potential issues. Another issue is users |
| 34 |
who ended up with that package in @world. If it were masked, they would |
| 35 |
know why they need to remove it. Now, they will just get awful blockers. |
| 36 |
|
| 37 |
-- |
| 38 |
Best regards, |
| 39 |
Michał Górny |
Archives