1 |
Hi, |
2 |
On Sun, 9 Oct 2005 18:19:50 +0000 |
3 |
Jerry McBride <mcbrides9@×××××××.net> wrote: |
4 |
|
5 |
> |
6 |
> I'm getting ready to make a post on bugs.gentoo.org to ask for a |
7 |
> couple of feature enhancements for portage's emerge routine. I |
8 |
> thought I'd ask here first, to see how everyone feels about them and |
9 |
> if there's too much negative response I'll just drop the idea. |
10 |
> |
11 |
> Two new features I'd like to see added are; |
12 |
> |
13 |
> 1- In /etc/make.conf add a new key to FEATURES that would cause |
14 |
> portage to cleanup the DISTDIR folder if everything emerges cleanly. |
15 |
> Perhaps "CLEANDIST" or "DISTCLEAN" or something similar would |
16 |
> suffice. I never, ever keep the source files after a successfully |
17 |
> emerge session and it would be nice if portage would take care of the |
18 |
> dirty work for me.... automatically. |
19 |
> |
20 |
Think this is already in Portage (as FEATURE) and is called |
21 |
"distclean", haven't tried it though - remove source after install. |
22 |
> 2- We have "emerge world" that covers everything already installed |
23 |
> and "emerge sys" that covers all the system related stuff.... How |
24 |
> about "emerge apps" that would ONLY cover things NOT in "emerge |
25 |
> sys"? Face it, it'd be nice.... Right now, I have to resort to extra |
26 |
> steps to "emerge apps" and well... it'd be a nicer chore for portage |
27 |
> to handle. |
28 |
> |
29 |
What about e.g."emerge -ev gnome", think this will work, specially for |
30 |
meta-packages. Tried it - too many packages, catches deps of deps etc. |
31 |
> 3- Make a list of hooks available to hackers, like myself. I know I |
32 |
> can implement the above features, but the emerge source is huge and |
33 |
> teasing out the details is proving more time consumeing than actually |
34 |
> implementing new code. Perhaps in /etc/make.conf there could be |
35 |
> config lines like, "EXECUTE_AFTER_OK_EMERGE" and we could fill in a |
36 |
> path to our own cleanup scripts... Or maybe |
37 |
> "EXECUTE_AFTER_BAD_EMERGE" in the same case as above, but after a bad |
38 |
> emerge run, etc.... |
39 |
> |
40 |
> |
41 |
> So, what do you guys/gals think? |
42 |
> |
43 |
HTH.Rumen |