1 |
err, DO_NOT_COMPILE="kmail" in /etc/make.conf does just that, and |
2 |
supports every package or even service KDE has to offer. Feel free to |
3 |
not compile ksmserver or kinit as well -- this will break things, but |
4 |
the fact is, gentooists always had the possibility to do this! |
5 |
|
6 |
What exactly do you propose here? -- do you actually propose gentoo |
7 |
developers should split the metapackages into close to 100 ebuilds? -- |
8 |
what gain over DO_NOT_COMPILE does this give? |
9 |
|
10 |
How about putting this in the ebuilds: |
11 |
kmail? || DO_NOT_COMPILE="${DO_NOT_COMPILE} kmail" |
12 |
konqueror? || DO_NOT_COMPILE="$DO_NOT_COMPILE} konqueror" |
13 |
kdm? || DO_NOT_COMPILE="$DO_NOT_COMPILE kdm" |
14 |
... |
15 |
|
16 |
I'm not in favor of this as this will make many extra use flags, but |
17 |
it still seems a far better solution than splitting the meta packages. |
18 |
|
19 |
If only you checked out the forum ;) -- its a popular question as you |
20 |
rightly said. |
21 |
|
22 |
Roman |
23 |
|
24 |
|
25 |
On Tue, 19 Oct 2004 19:26:13 +0200, Dan Armak <danarmak@g.o> wrote: |
26 |
> Hi all, |
27 |
> |
28 |
> Perhaps the oldest request of Gentoo KDE is to have individual ebuilds for all |
29 |
> KDE apps - konqueror, kmail etc. Most people will probably never use over |
30 |
> half the stuff 'emerge kde' installs, and it's hardly Gentooish to force it |
31 |
> on them. |
32 |
> |
33 |
> I've spent the last few years pointing out to various people why this couldn't |
34 |
> be done (well). Eventually I got really tired, so we did it. |
35 |
> |
36 |
> So without further ado, I point you to http://kde-metaebuilds.berlios.de/ and |
37 |
> the more detailed explanations there. Brought to you by Simone Gotti |
38 |
> <motaboy@g.o> and yours truly. |
39 |
> |
40 |
> *** Current status: early beta. Works For Us (tm). DO NOT send bugreports to |
41 |
> *.gentoo.org - join the mailing list at berlios.de. *** |
42 |
> |
43 |
> --- |
44 |
> |
45 |
> For everyone who's still reading... |
46 |
> |
47 |
> You probably remember me saying this wouldn't be done because an 'emerge |
48 |
> kde-meta' would take a few hours longer than a classic 'emerge kde'. This is |
49 |
> true. So I refer you to http://dev.gentoo.org/~danarmak/ where you'll find |
50 |
> the experimental confcache patch to portage, which virtually eliminates the |
51 |
> overhead of running configure for every tiny ebuild. |
52 |
> |
53 |
> However, while the ebuilds are in early beta, confcache is in late alpha. A |
54 |
> few known (non-dangerous) problems exist. So if you use it, be sure of what |
55 |
> you're doing. I'd also like to ask anyone and everyone to help make confcache |
56 |
> better... |
57 |
> |
58 |
> I release when ready. And what this's ready for is a lot of testing, rather |
59 |
> than a lot of production use :-) But there are no known issues (with the |
60 |
> ebuilds), so please go ahead and use them. |
61 |
> |
62 |
> Eventually, esp. if and when confcache goes stable and is added to portage, |
63 |
> these ebuilds can make their way into portage also. |
64 |
> |
65 |
> -- |
66 |
> Dan Armak |
67 |
> Gentoo Linux developer (KDE) |
68 |
> Matan, Israel |
69 |
> Public GPG key: http://dev.gentoo.org/~danarmak/danarmak-gpg-public.key |
70 |
> Fingerprint: DD70 DBF9 E3D4 6CB9 2FDD 0069 508D 9143 8D5F 8951 |
71 |
> |
72 |
> |
73 |
> |
74 |
|
75 |
-- |
76 |
gentoo-dev@g.o mailing list |