1 |
On Sunday 13 August 2006 12:20, Duncan wrote: |
2 |
|
3 |
> |
4 |
> > Oh, and for local-flags, there are several descriptions, have a look at |
5 |
> > ufed. For the global ones 'pulls in X' or 'needed for mp3/wmv/avi |
6 |
> > support' is really enough to know.It does not matter, that the single |
7 |
> > package does. I want them to have wmv/mp3/X support, how they are do it, |
8 |
> > is the ebuild's problem, not mine. I set a flag, the ebuild maintainer |
9 |
> > has to figure out how to react to it. |
10 |
> |
11 |
> What I'm suggesting is that use.desc stay more or less as it is, with a |
12 |
> general description for global USE flags. However, instead of |
13 |
> use.local.desc only having non-global USE flags, have it list all flags |
14 |
> (or split it into two or more files if it gets unmanageably huge) for all |
15 |
> packages, with what they do for that package. |
16 |
> |
17 |
> For a quick idea of what the USE flag does in general, then, if it's a |
18 |
> global USE flag, one would check the entry in use.desc which would be |
19 |
> much as it is now. For a better idea of what it does in a particular |
20 |
> package, check the corresponding entry in use.local.desc, which would |
21 |
> describe the effects of the flag on that particular package. That's what |
22 |
> I'm proposing. Users could just check the general description if that's |
23 |
> all they wanted/needed, and have exactly the same level of info they have |
24 |
> now, with a possible tweak to a description here or there. If they wanted |
25 |
> to know for example what the gnome flag did in the pan package, however, |
26 |
> they'd look in use.local.desc and see something to the effect of "Builds |
27 |
> against libgnome to let pan use the configured gnome browser." |
28 |
|
29 |
but we already have that! |
30 |
|
31 |
Start ufed. Read some of the flag descriptions. For a lot of them, there are |
32 |
several ones. |
33 |
|
34 |
avahi has since descriptions - for six different packages, or atm, two |
35 |
descriptions, audacious, three... for each package a different one. |
36 |
|
37 |
|
38 |
> |
39 |
> See, the problem is that a flag, while it generally adds support for |
40 |
> <flagfeature>, can mean very different things in different ebuilds. An |
41 |
> example is the perl flag. In some ebuilds, it means build perl bindings. |
42 |
> In others, it means install documentation for use with perl. In still |
43 |
> others, it controls building optional package documentation that requires |
44 |
> perl to build -- documentation for the package, not for using it with |
45 |
> perl, but requiring perl to build that documentation. Those are three |
46 |
> VERY different meanings, applying to different packages, with USE=perl |
47 |
> used to control them. Having a per-package entry would allow the user to |
48 |
> see precisely which of these the perl flag was used for in a particular |
49 |
> package, or if it was used for something else entirely. There's simply no |
50 |
> way to convey that with a global description, unless you effectively |
51 |
> include the per-package descriptions right in the global description, of |
52 |
> course making it long enough to do so, which would then leave us without a |
53 |
> way to get a short and concise general description whet that's all that's |
54 |
> needed. |
55 |
> |
56 |
> Still think it's insane, or did I actually convey the idea in a way that |
57 |
> makes a bit more sense, now (whether you agree with it or not)? =8^) |
58 |
|
59 |
for local flags it is already done - and global flags... is such an amount of |
60 |
information really needed? |
61 |
|
62 |
If I have perl installed, why should I not want perl bindings, perl |
63 |
documentation and perl support in a package? Or pan - if I have gnome |
64 |
installed, why should I deactivate gnome support? 'It has gnome support, |
65 |
fine' why should I need more information? And if I really need to know, what |
66 |
gnome support means, I can always look into the ebuild. |
67 |
|
68 |
Lots of information is a nice thing, but too much of it is not good either. |
69 |
Struck dead by the amount of information... (Er wurde von der Last des |
70 |
Wissens erschlagen.) it can happen, and it does happen. ufeds informations |
71 |
are already on the verge of getting to much - removing some here and there |
72 |
would be helpfull (like three of the 6 avahi comments), because you won't get |
73 |
to the end, if you have to read all of it. |
74 |
-- |
75 |
gentoo-amd64@g.o mailing list |