KDevelop Roadmap 2001

[Roland Krause]

Ralf, 
thanks for the effort of putting this together. I would like to add a
few things and volunteer for certain tasks. 

--- Ralf Nolden <nolden at kde.org> wrote:
> 
> KDEVELOP_1_4
> ------------
> is our main branch for further improvements. This version will be
> shipped with KDE 2.2, the KDevelop version will be 2.0. 

This is very good, I was really hoping that development on 1.4 will
continue. I have talked to John about this on IRC for a few days now
and I have tried to convince him of this.

> 
> Changes planned/developers searched for:
> -KOffice template
> -documentation update to docbook
> -QextMDI by Falk Brettschneider
> -code completion (maybe) if the code of KDEStudio can fit into
> KDevelop.
> Volunteers please take a look at the code-completion of kdestudio,
> the
> source code is GPL and can be retrieved by anonymous CVS at
> theKompany.com, please look at http://www.theKompany.com. 

1. static library and shared library templates, if these arent there
   already. I haven't actually looked for a while. 
2  multiple projects
3  ctags support 
4  improved glimpse support 
5  general clean up of CKDevelop and CProject
6  support for Fortran

I have started to dig into 5 a little bit w/ a minor revision of the
BuildMake stuff. Once that is working correctly, I'd like to look into
how difficult it is to get a Workspace/Projects concept going. 

I have also begun to implement ctags support with the first, easy part,
being done and the second, hard part started. There was a discussion on
code completion earlier where ctags was suggested and I still think we
could pull this off for the exact reasons you have stated in your other
email to this list. Ctags is very fast (~1MByte/s) and it has most of
the information necessary. 

I read the earlier discussion, of course you can fool ctags, but that
is not the point.

Maybe, the person who suggested earlier on this mailing list that he
wants to work on this could please step forward once again. Maybe we
can join forces somehow. 

After that, I need to improve glimpse support as an enhancement of the
existing grep search. That requires the possibility to glimpse_index
source code and replace the grep command by glimpse, both short term
goals and not too hard to accomplish w/the existing code base. 

I also need to implement some Fortran support. Mostly for the editor
but maybe this can evolve in some additional compiler support. I will
look at what is in the HEAD branch and maybe I'd be able to backport
some of it. 

All these things I will do mostly for myself, because I need them to be
working, but I would like to contribute as much as possible back to the
main development branch. 

Code indexing and cross referencing is as much a concern for me as is
code completion. In fact, one will ultimatively lead to the other. 

We should also probably talk about "long term development" plans, since
these projects are all midterm. I know there is development in HEAD
going on, but as it was said before, this is a completely new
application and hence shares no code. 

Personally, if I wanted that, I might as well start on my own or join
one of the numerous other IDE projects. For me the advantage of
kdevelop is to have a working, functioning application w/ a user base
large enough to supply bug reports, ideas as well as fresh, additional
developers. 

Therefore, and again this is my personal opinion, I support the HEAD
effort and wish the best for it but I will concentrate my contributions
to kdevelop-1.4. 


> The reason
> I
> think we should feel free to share their code is that kDEStudio is
> based
> on KDevelop as well, so it really doesn't matter if we take the code
> from there as well this time for KDevelop :-) For this addition,
> please
> make some comments in the code where necessary for copyright issues
> if
> you copy 1:1 instead of rewriting the concept.

I agree.  I dont really like the way "theKompany" operates. They
constantly shove out half finished, buggy applications and then simply
never do anything on them anymore. If they were that serious why didnt
they join the kdevelop effort?

Nevertheless, if there is something to be harvested, by all means we
should do it. I have the KDEStudio code and kdestudio running here and,
"aehem", i dont think there is all that much to be won. 

> -the aboutpage is to be changed into konqueror's style so the
> contents
> goes into the po file as well.

> 
> HEAD
> -----
> The version in HEAD is going to be as follows: While Qt 3.0 will
> arrive
> with some major changes to the API and KDE 3.0 will be on the way, we
> will rename the HEAD version to KDevelop 3.0. The main work should be
> taking place here with all new developers approaching pointing to
> this
> tag and version except those who want to put improvements into 1.4
> resp.
> 2.0. 

Best regards
Roland



=====
Roland Krause



__________________________________________________
Do You Yahoo!?
Get email at your own domain with Yahoo! Mail. 
http://personal.mail.yahoo.com/

-
to unsubscribe from this list send an email to kdevelop-request at kdevelop.org with the following body:
unsubscribe »your-email-address«