1 |
Alec Warner wrote: |
2 |
> Vaeth <vaeth@××××××××××××××××××××××××.de> wrote: |
3 |
> |
4 |
> > (Moreover, in most cases, |
5 |
> > this would not even save some characters, because the variable |
6 |
> > names would have to be much longer...) |
7 |
> |
8 |
> I don't think the variable names are really in scope (we can chose |
9 |
> them arbitrarily). |
10 |
|
11 |
But you cannot choose them much smaller unless you want to decrease |
12 |
readability even more. |
13 |
|
14 |
> I disagree with less readable; it is less intuitive |
15 |
|
16 |
The point is that in contrast to shell code you need additional |
17 |
pre-knowledge to read or write it. |
18 |
|
19 |
> the syntax looks fine and the syntax is in fact still bash. |
20 |
|
21 |
I do not want to start a discussion now whether this is |
22 |
implicit semantic or sort of an extended syntax - it depends on the point |
23 |
of view. But in any case it involves new (and actually redundant) |
24 |
"keywords" in the ebuild. |
25 |
|
26 |
> However, ebuilds are already |
27 |
> sufficiently complicated by eclass inheritance that reading |
28 |
> many ebuilds is already difficult |
29 |
|
30 |
This is the point. For certain very specialized eclasses like kde or qt |
31 |
this might be inavoidable, but one should strive to minimize these things |
32 |
as much as possible. So an idea how to go into the *opposite* direction |
33 |
is what is actually needed. |
34 |
|
35 |
> and I think this extension does not make that significantly worse. |
36 |
|
37 |
"because we made some mistakes anyway let us make even more..." |
38 |
|
39 |
> Arguing about intentions is not really a compelling argument. Spec |
40 |
|
41 |
It is, if something does not satisfy (anymore) what is supposed to do. |
42 |
|
43 |
> files have more than magic variables too; many have |
44 |
> similar constructs to ebuilds (postinst and prerm phases, file |
45 |
> manifests, etc.) Specfiles and ebuilds are more alike than different. |
46 |
|
47 |
Yes, meanwhile specfiles and ebuilds are very similar which is really |
48 |
a problem: It is not accidental that many users of other distributions |
49 |
prefer to install into /usr/local instead of writing rpms - most users |
50 |
do not want to learn a specification. It is fatal if ebuilds go the |
51 |
same road. The knowledge needed to write or read ebuilds should be kept |
52 |
as small as possible. |