1 |
On Mon, 7 Nov 2016 22:47:24 +0200 |
2 |
Alon Bar-Lev <alonbl@g.o> wrote: |
3 |
|
4 |
> Just my 2 cents... |
5 |
> I kinda love the prefix nature of the expressions which is consistent |
6 |
> and easier to parse. |
7 |
> Using infix only for versions and leaving all the rest prefix will |
8 |
> create abnormality. |
9 |
|
10 |
You know what? Let me break all your dreams. |
11 |
|
12 |
So, the way I see it: someone back in the ol' days thought: let's do |
13 |
version syntax different than everyone else, so we look cool. And let's |
14 |
use some smart name for it, even though it makes no sense |
15 |
in the context. And as all bad things in Gentoo do, suddenly many |
16 |
developers were thinking the same mistaken idea. |
17 |
|
18 |
So let's keep it short: |
19 |
|
20 |
1. This is NOT *prefix notation*. This is some kind of sick weirdo |
21 |
prefix-suffix notation. |
22 |
|
23 |
2. The suggested solution is *suffix*, NOT infix. Like everything else |
24 |
in the package dependency spec. |
25 |
|
26 |
3. If everything else were 'prefix' like this, then package |
27 |
dependencies would look like >:[?app-foo/bar-1.2.3-1-gtk. |
28 |
|
29 |
4. We are *not* comparing package name against the version. We are |
30 |
applying various restrictions to a dependency. |
31 |
|
32 |
5. Dependency groups are actually using prefix notation. Still, I don't |
33 |
see any benefit from reversing package dependency specs and doing |
34 |
[gtk]3:>3.4.0-app-foo/bar. Though it's probably still better than what |
35 |
we're doing right now since we can at least sanely split version |
36 |
and category. |
37 |
|
38 |
So, sorry, the syntax does not make you look cool, it doesn't make |
39 |
sense and is inconsistent with everything else. |
40 |
|
41 |
-- |
42 |
Best regards, |
43 |
Michał Górny |
44 |
<http://dev.gentoo.org/~mgorny/> |