Contributing to KDevelop as a student

[David Nolden]

Am Mittwoch 06 Januar 2010 20:08:11 schrieb Ishan Jayawardena:
> Hi
> I'm a student and I am interested in participating in this year's GSoC
> for KDevelop.( I hope that KDE is going to participate as a mentoring
> organization this time as well.) And I want to prepare for it in
> advance and I'm trying to find some way that I can contribute to
> KDevelop. By now, I have compiled KDevelop and fixed a very little bug
> in the editor part. I'm really interested in the editor, auto
> completion, code navigation and re factoring issues.
> So I expect some guidelines from you to achieve my goal. What should I
> work on? Are there any issues with those modules? Who should I
> contact? what can I do to improve KDevelop? If I have a suggestion who
> should I tell? How to be selected for GSoC for KDevelop?

The whole language support part is pretty complex, at least it is not easy to 
get into. Last year's student working in this area had a lot of problems, and 
I think I won't mentor such a project again unless the Student proves 
beforehand that he's able to reach the goals.

I'm maintaining exactly the areas you mentioned. Except for refactoring they 
are pretty mature, but of course there is still bugs and glitches here and 
there (see bugs.kde.org). The first thing you could try to do is understanding 
how the whole thing works. Then you could try to fix some small code-
completion or duchain problems you find either by yourself, or on 
bugs.kde.org. You can take a look at the tests at 
kdevelop/languages/cpp/cppduchain/tests kdevelop/languages/cpp/tests to see 
how code is transformed into the duchain.

If you prove that you're able to get into the architecture, then the next 
thing to do would be defining the goals of a possible summer of code project. 
Such a project of course doesn't just consist of "fix a few bugs", but usually 
a bigger target should be set. You can use KDevelop for some time, and when 
you get interesting ideas on possible goals, you can discuss them here.

Btw. there is also other areas in KDevelop that are easier to get into, and 
would also well be worth a GSOC project, for example "rock solid git support", 
"debugger visualization", "extensive command-line integration", or things like 
that.

If you have questions you should just ask them on the mailing list.

Greetings, David