Re: [jamboost] Adding QT support to Boost Build V2

From: Jürgen Hunold (hunold+lists.Boost_at_[hidden])
Date: 2003-01-17 16:03:52

Hi !
Am Freitag 17 Januar 2003 15:43 schrieb Felix E. Klee:

> > Well. As stated above, each "special" extension causes much work
> > with existing projects.
>
> I have the impression that you misread my answer. 

Well, at first I thought you still insisted on a special extension.
Sorry if I overreacted.

> I simply
> recommended to replace ".h" with ".hpp" (not ".qpp") which is the
> boost standard extension for header files. Letting the user specify
> the extension (or extensions) for files to be moc'ed would even be
> more flexible. Then you could continue using ".h" extensions for your
> header files and I can continue using ".hpp" extensions.

I would like .h as default extension. 
Rationale:
I've do a short find | grep | wc for Q_OBJECT in .h files on my local 
sandboxes. I found 1528 header files which needs moc'ing. 
All headers CVS-Controlled. (Some .ui generated, but still...)
And believe, it is no fun to "simply" rename files in CVS. Ask the 
boost-CVS maintainers or the list archive for examples. The worst I 
remember was renaming is_POD.hpp to is_pod.hpp. 
Please consider that many possible users have a running setup (qmake, 
old tmake, even handcraftet Makefiles, MSVC-Projects using the 
Qt-Visual Studio binding, ...) All these depend on .ui -> .h -> 
moc_.cpp Depencies.

> > For me, this is just inacceptable.

Just to be sure: The .hpp approach is acceptable if you start fresh. I 
my (and many other cases) you have so much legacy code lying around 
which simply need the prefix .h . 
Naturally, the extension should be made to be customised.
But Qt uses .h everywhere, so it seems best, not to change this...

> > And we're using doxygen to generate docs out of _every_ header.

> Did you again misread my original text or are there indeed features
> in Jam that support Doxygen? If so, where can I find more
> documentation?

Oh sorry, this is a clear misunderstanding. I was focused on the moc 
issue. We're just using doxygen with the standard doxygen configuration 
file. I would be glad if someone could improve this, because generating 
docs by "grab all and run...." is not the best thing to do...

BTW, sourceforge seems to up again for me too. Strange thing, that.

Yours,

Jürgen

--
* Dipl.-Math. Jürgen Hunold ! Institut für Verkehrswesen, Eisenbahnbau
* voice: ++49 511 762-2529 ! und -betrieb, Universität Hannover
* fax : ++49 511 762-3001 ! Appelstrasse 9a, D-30167 Hannover
* hunold_at_[hidden] ! www.ive.uni-hannover.de