| 1 |
On Wed, 12 Jul 2017 15:00:30 +1000 |
| 2 |
"Sam Jorna (wraeth)" <wraeth@g.o> wrote: |
| 3 |
> |
| 4 |
> My point was that --unmerge is not intended to be dependency-aware. |
| 5 |
> --depclean is. As far as I can tell, that is the point others have |
| 6 |
> been trying to make as well, when pointing out the differences |
| 7 |
> between -c and -C. |
| 8 |
|
| 9 |
Sure but that is easily addressed. |
| 10 |
|
| 11 |
How does emerge know a package is in system profile or a set? |
| 12 |
Similar methods or others can be used to determine if a user installed |
| 13 |
a package. |
| 14 |
|
| 15 |
In a clean scenario, with a world file that ONLY contains stuff the |
| 16 |
user merged directly. Then emerge could simply check that file, against |
| 17 |
the stuff it already checks now. |
| 18 |
|
| 19 |
Is it in system? |
| 20 |
Is it in a set? |
| 21 |
Is it in world? |
| 22 |
If no to all, its a dep, warn! |
| 23 |
|
| 24 |
It is really NOT complex, nor does it require complex depgraph or any |
| 25 |
of the function of --depclean/-c. |
| 26 |
|
| 27 |
Thus I say once again, mentioning anything to do with depclean is not |
| 28 |
relevant and side tracks the discussion. There is no need. If you did |
| 29 |
want to actually see if any deps existed. All you need is to see if |
| 30 |
there is 1 installed package that needs the one a user is trying to |
| 31 |
install. No need for a complete depgrah etc. |
| 32 |
|
| 33 |
There are several ways to go about this that are not complex. |
| 34 |
|
| 35 |
-- |
| 36 |
William L. Thomson Jr. |
Archives