1 |
On Saturday 06 December 2003 02:09, Marius Mauch wrote: |
2 |
> Seeing this "language war" on -dev I think I should say again that the |
3 |
> component model should make us free from language restrictions. There is |
4 |
> no sense in saying "we should use language XXX for portage-ng" as the |
5 |
> goal should be that each component can be implemented in the best |
6 |
> fitting language. So it should be possible to have the dependency |
7 |
> resolver in prolog, the ebuild parser in perl, the frontend in python, |
8 |
> the storage backend in C and so on. |
9 |
|
10 |
I believe this is already part of the requirements. |
11 |
|
12 |
> Instead of arguing about the "best" |
13 |
> language for implementation we should discuss about the language for the |
14 |
> _interface_ for the component interaction. |
15 |
> Once we have decided on that we can start creating the global |
16 |
> architecture that describes which components interact with each other, |
17 |
> which components are mandatory or optional and so on. |
18 |
|
19 |
I think the point is that even designing the global architecture should come |
20 |
before deciding on a language for the component interaction interface. |
21 |
|
22 |
> Later in that |
23 |
> process we can specify the first function signatures and start |
24 |
> implementing the individual components. Then and not earlier we have to |
25 |
> choose the implementation language. |
26 |
|
27 |
This is correct. The implementation language for a particular component will |
28 |
be chosen once the framework is decided and it is time to create said |
29 |
particular component within that framework. |
30 |
|
31 |
Regards, |
32 |
Jason Stubbs |
33 |
|
34 |
-- |
35 |
gentoo-portage-dev@g.o mailing list |