how to turn off -fno-exceptions in kdevelop

[Juergen Suessmaier]

Hi Ralph,

> Ok, can't resist to answer :-))
Great, as I'm always interested to learn about internals -regardless whether
concerning softare techniques or just environmen- I appreciate your answer
very much.

> Well, with KDE *1*. With KDE 2 a lot of things have changed so we had to
> deal with two things:
> a) the KDE 2 architecture with XML stuff, changing API until almost end
> of last year, integrating that admin directory and it's problems and
> keeping kdevelop still as compatible as possible to KDE 1 until KDevelop
> 1.3. Now, kdevelop 1.4 is a port and although the kde 1 templates are
> removed that didn't mean there are still places where work is necessary.
Yep, and despite the user inconveniences regarding the building process
(which I guess isn't the fault of the KDevelop team anyway) you did an
absolute excellent job!

> But generally, using the admin dir for KDE development is the way to go,
Ok, but based on what decisions?

> as well for Qt development. It is constantly updated by the KDE guys
> like Stephan and you have to mind that there are a lot of easy things in
> the Makefile.am's that can be used if you know how to write the
> Makefile.am's. This is what's going on in the main KDE CVS' modules,
Agree, but probably due to a lack of information I still don't see a
reason for diabling exceptions or 64 bit int arithmetic ny default. I
assume that you guys don't fiddle with compiler switches without having good
reasons to do so. So where's the actual problem with enabled exceptions or
'long long' stuff? Reentrancy? I personally use the GNU compiler among other
stuff as a cross development system for 68K embedded targets and I switched
off exceptions as they make (along with rtti) the code a lot larger. That's
an issue for systems with a very limited amount of ROM space but I don't
think it's an issue for a standard "bloated" workstation with tons of
RAM and harddisk space...

> let's say with Icons. I myself have to look up these rules there
> sometimes. What is needed is to find out the rules automatically and
> dynamically because we used the admin dir of the checks it makes and
> that it should make things easier to update with each KDE / Qt version.
> The pro's are more than the con's here definetly, although it may seem
> like it's causing you some trouble right now.
Is there any pointer to some discussions or documentation concerning the
pros and cons? I'd be really interested in getting more information about
current development plans and the issues you are aiming at.

> needed by Stephan for KDE stuff. If you're not happy, take Stephan's
> solution to avoid using them and write a patch to the project managment
> for 2.0 and take part on that stuff in the active development. As much
>From my personal point of view I'd really like äto help a bit in active
development. I've been using Linux and open source stuff for about 8 years
now and I'm definitely convinced of its quality and power, so it would
be time to do something for the community too (except testing and debugging
as much as possible). The problem is time. I'm working full time as a hired SW
developer, I'm driving my own little software company in my spare time. I'm
married and have 2 little kids and I'm a student of computer science at
"Fernuniversitaet Hagen". Honestly, that's almost more than I can handle...
However, as this posting may become some sort of OT please feel free to
contact me at juergen at suessmaier.de. Maybe we can find some small subprojects
that I can help to work on.

> as I appreciate comments and hints, I would like to see the developers
> using kdevelop taking more part actively in the development, even if it
> would be in small areas of the code. The codebase is so big that
Yep, I really do understand this kind of problems!

> everyone can look for a part to work on and everyone is invited to do
> so. Take Roland as a good example. He ranted a bit first and is now
> trying to do what he can do in his spare time, like we all do. I
> wouldn't be that front up normally but there is, in opposition to many
> other free projects, *nobody* paid for KDevelop development - although
> some of us could really need it because we invest more than 10 hours a
> day into our project. 
Yes, I can imagine that. A project like KDevelop isn't anything simple and
I personally really admire the effort as well as the results of your
team's work.

> I have it working here. You need to run make messages and merge first to
> rebuild the pot file and add a po file of course. The differences will
> be merged into the po file then.
Thanks a lot for the hint. I'll check this out tomorrow - too late for
experiments right now :-(

Regards,
Juergen
-- 
-------------------------------------------------------------------------
J. Süßmaier Systementwicklungen
Jürgen Süßmaier
juergen at suessmaier.de                       Realtime Software Development
Katharina Geisler Str. 14                       Embedded Applications
D-85356 Freising                                     Automation
Germany
-------------------------------------------------------------------------
                      http://www.suessmaier.de
-------------------------------------------------------------------------
      The day Microsoft makes something that doesn't suck is
        probably the day they start making vacuum cleaners
-------------------------------------------------------------------------

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