Trolltech Home | Qt-interest Home | Recent Threads | All Threads | Author | Date
All threads index page 5

Qt-interest Archive, March 2002
Alternatives to VC++ IDE?

Pages: Prev | 1 | 2 | 3 | 4 | Next

Message 1 in thread

Please pardon my frustration. I'm so sick of VC6!!! I want a modern C++
compiler!! I thought VC7 would be better but it doesn't even work with Qt.
I need a good IDE, a more standard-compliant compiler, and Qt compatability.

Does anyone know of a viable alternative to the MS VC6? I don't think C++
Builder builds Qt apps. Only VCL and CLX. What about CodeWarrior? Any
suggestions?

Craig


Message 2 in thread

On Tuesday 12 March 2002 12:29 pm, Craig Black wrote:
> Please pardon my frustration. I'm so sick of VC6!!! I want a modern C++
> compiler!! I thought VC7 would be better but it doesn't even work with Qt.
> I need a good IDE, a more standard-compliant compiler, and Qt
> compatability.
>
> Does anyone know of a viable alternative to the MS VC6? I don't think C++
> Builder builds Qt apps. Only VCL and CLX. What about CodeWarrior? Any
> suggestions?

You may want to check out Intel's C/C++ compiler.  That may be your best bet. 
 I hear that gcc 3.0.x is fairly C++-compliant but I don't know anyone who 
has actually used it yet.


Message 3 in thread

Correct me if I am wrong but gcc is only available
under the cygwin environment.  This means we cannot
use (I am assuming) compile Qt 3.0 for Windows with
gcc.

Salman


--- Christopher Thompson <christ@axionspatial.com> > 
I hear that gcc 3.0.x is fairly C++-compliant but I
> don't know anyone who 
> has actually used it yet.
> 
> --
> List archive and information: http://qt-interest.trolltech.com

__________________________________________________
Do You Yahoo!?
Try FREE Yahoo! Mail - the world's greatest free email!
http://mail.yahoo.com/


Message 4 in thread

> You may want to check out Intel's C/C++ compiler.  That may be
> your best bet.
>  I hear that gcc 3.0.x is fairly C++-compliant but I don't know
> anyone who
> has actually used it yet.

The latest gcc that ships with cygwin is very good from a standards
compliance point of view - it will compile lots of libraries (e.g. Boost,
Spirit, etc.) that need partial template specialisation without problems.
I've not personally tried to compile Qt/windows with it, but the (oldish)
version of tmake I have seems to mention it, so it is possibly worth
investigating.

Intel C++ definitely works for Qt, without any problems at all, and
generates somewhat better code too. It integrates perfectly with VC6, but
with the proviso that you MUST NOT have anything else installed that also
tries the same trick, e.g. Bounds Checker. (BC won't install if you have
Intel installed by the way - I know, I spent ages trying a few months ago).

I'm currently (in a .NET project, not Qt) using VC7 at work. It looks
pretty, the .NET integration is nice, the .NET class library is way better
than any of MS's previous attempts (its event handling looks oddly
reminiscent of Qt's signal/slot metaphor, though it's less powerful).
However, the VC7 C++ compiler is pretty sucky - I've been hit by a couple of
code generation bugs so far. It won't compile Boost (at all, as far as I can
tell). No partial template specialisation support. The debugger is flaky
under Win2k (can't speak for other platforms) - if you close a console app
by clicking its close box while it is halted at a breakpoint, the IDE hangs
maybe one time in two. The code it generates feels a bit slow - I can't
really put my finger on it, but I'd really like to do some back-to-back
comparisons. It may be the longer load time due to MSIL JIT compilation, I'm
not sure.

For Windows development under Qt, I usually just use the cygwin command line
and a copy of the gvim windows port. I always found that the MSVC IDE was
more trouble than it was worth for editing Qt source. If you hack the config
files, gvim does syntax colouring very nicely, and it is possible to add the
extra Qt-isms like slot and emit so that they get coloured too. Changing the
default (horrible) font to Lucida Console, and going for a white on black
scheme really worked for me (your mileage may vary).

Sarah


Message 5 in thread


<<< text/html: EXCLUDED >>>


Message 6 in thread

On Wed, Mar 13, 2002 at 10:31:55AM +0100, Henrik Johnson wrote:
> Still they both have a pretty steep initial learning curve. I 
> recently had a debate on what was most user unfriendly CTRL-X+CTRL-C or 
> ESC+:+q+! for exiting the application. No matter what the answer neither 
> can really be called simple :).

no matter they're not called simple, they're still the most powerful ones you
can find on the planet now, once you get in, you can't get out

I'd suggest forgetting all this RAD IDE Visual mega super all-in-one bull**it, 
and taking a moment and looking at what the real hackers use...

don't forget that editor is still the most frequently used piece of software
coders need, and also don't forget that the best things are for free (love,
nature, etc., fill in your own), and that it *IS* worth spending some time
climbing the steep learning curve..., it'll return back later, how simple...

> /Mauritz
> Underscore AB

regards,
martin

-- 
 [ signature omitted ] 

Attachment: pgp00000.pgp
Description: PGP signature


Message 7 in thread


<<< text/html: EXCLUDED >>>


Message 8 in thread

On Wed, Mar 13, 2002 at 06:17:03PM +0100, Henrik Johnson wrote:
> You don't have to sell this to me :). I do all my work in Emacs except 
> for some documentation which I still need to resort to some kind of 
> office like tool.

sure I didn't want to tell you, rather to the list, your email just served as
a nice quote :-)

> 
> /Mauritz
> Underscore AB

rgds,
martin 

-- 
 [ signature omitted ] 

Attachment: pgp00001.pgp
Description: PGP signature


Message 9 in thread

Mauritz wrote:
 > You may want to check out Intel's C/C++ compiler.  That may be
>your best bet.
> I hear that gcc 3.0.x is fairly C++-compliant but I don't know
>anyone who
>has actually used it yet.

Definitely agree - Intel is solid and IMO an excellent compiler. I'd also
wholeheartedly recommend taking a look at Intel's VTune profiler, which is
quite simply the best profiler I've ever seen, and also works with Qt.

>I off course have to add the obligatory. Emacs is also available in a
native Windows version
>which is also a very nice IDE if you are used to it. Still they both have a
pretty steep
>initial learning curve. I recently had a debate on what was most user
unfriendly CTRL-X+CTRL-
>C or ESC+:+q+! for exiting the application. No matter what the answer
neither can really be
>called simple :).

Yes, indeed. Editors are such a personal thing. It is a shame, is it not,
that the two most popular varieties of editor (vi influenced and emacs
influenced) are both frankly quite insane pieces of design!

Sarah


Message 10 in thread

Hi
I am using a combination of VC 6 and Intel C++. Intel C++ plug into the 
VC++ IDE

At 13:29 12/03/2002 -0600, Craig Black wrote:
>Please pardon my frustration. I'm so sick of VC6!!! I want a modern C++
>compiler!! I thought VC7 would be better but it doesn't even work with Qt.
>I need a good IDE, a more standard-compliant compiler, and Qt compatability.
>
>Does anyone know of a viable alternative to the MS VC6? I don't think C++
>Builder builds Qt apps. Only VCL and CLX.

It can be used to build Qt/C++ apps

>What about CodeWarrior? Any
>suggestions?

I have tried CodeWarrior, but I could not get on with it since it does not 
appear to be able handle moc in the same way VC++ does.


>Craig
>
>--
>List archive and information: http://qt-interest.trolltech.com

Regards
John,

-----
john@theKompany.com
http://www.theKompany.com


Message 11 in thread

The Intel compiler may be much better than VC++.  However, I evaluated it a
few months ago with MSVC6 IDE. I found that it still uses the Microsoft
compiler to produce dependency information. This is a crippling problem,
because MSVC6 doesn't compile the kind of code that I aspire to write,
seeing it is not standards compiant. Thus, the intel compiler will not
compile anything that VC++ will not compile because it still uses it to
produce dependency information. Anyway, that's what I ran into when I
evaluated it.

-----Original Message-----
From: Christopher Thompson [mailto:christ@axionspatial.com]
Sent: Tuesday, March 12, 2002 1:59 PM
To: cblack@ara.com; qt-interest@trolltech.com
Subject: Re: Alternatives to VC++ IDE?


On Tuesday 12 March 2002 12:29 pm, Craig Black wrote:
> Please pardon my frustration. I'm so sick of VC6!!! I want a modern C++
> compiler!! I thought VC7 would be better but it doesn't even work with Qt.
> I need a good IDE, a more standard-compliant compiler, and Qt
> compatability.
>
> Does anyone know of a viable alternative to the MS VC6? I don't think C++
> Builder builds Qt apps. Only VCL and CLX. What about CodeWarrior? Any
> suggestions?

You may want to check out Intel's C/C++ compiler.  That may be your best
bet.
 I hear that gcc 3.0.x is fairly C++-compliant but I don't know anyone who
has actually used it yet.


Message 12 in thread

On Tuesday 12 March 2002 1:22 pm, Craig Black wrote:
> The Intel compiler may be much better than VC++.  However, I evaluated it a
> few months ago with MSVC6 IDE. I found that it still uses the Microsoft
> compiler to produce dependency information. This is a crippling problem,
> because MSVC6 doesn't compile the kind of code that I aspire to write,
> seeing it is not standards compiant. Thus, the intel compiler will not

Hhhmm.  Well, remember that not one single C++ compiler is fully compliant 
with the latest C++ standards (or, afaik, any official C++ standard).  It is 
a very complex standard.

> compile anything that VC++ will not compile because it still uses it to
> produce dependency information. Anyway, that's what I ran into when I
> evaluated it.

Did you try it using makefiles?


Message 13 in thread

> > compile anything that VC++ will not compile because it still uses it to
> > produce dependency information. Anyway, that's what I ran into when I
> > evaluated it.
>
The only part of the compiler that is being used to produce dependency info is the pre-compiler, yes?  And the pre-compiler doesn't
look at any (or at least doesn't process any) C/C++ syntax - it just looks for it's own directives.

So, I would think you would be able to write the C++ to the back-end compiler, even though you may have to write to the MSVC
pre-compiler.

My 2c
rickb


Message 14 in thread

>The only part of the compiler that is being used to produce
>dependency info is the pre-compiler, yes?  And the pre->
>compiler doesn't
>look at any (or at least doesn't process any) C/C++ syntax -
>it just looks for it's own directives.

Not when I evaluated Intel C++. It was fully installed. I tried to compile a
program and got an error from the MSVC++ compiler as it parsed through my
code to generate dependency information. Since MSVC failed, the compilation
could not continue. The Intel C++ didn't even get a chance to start. If you
don't believe me evaluate it for yourself for free. And try something like
partial specialization on a template.

-----Original Message-----
From: owner-qt-interest@trolltech.com
[mailto:owner-qt-interest@trolltech.com]On Behalf Of Rick Berger
Sent: Tuesday, March 12, 2002 4:08 PM
To: qt-interest@trolltech.com
Subject: RE: Alternatives to VC++ IDE?


> > compile anything that VC++ will not compile because it still uses it to
> > produce dependency information. Anyway, that's what I ran into when I
> > evaluated it.
>
So, I would think you would be able to write the C++ to the back-end
compiler, even though you may have to write to the MSVC
pre-compiler.

My 2c
rickb

--
 [ signature omitted ] 

Message 15 in thread

At 18:12 12/03/2002 -0600, Craig Black wrote:
> >The only part of the compiler that is being used to produce
> >dependency info is the pre-compiler, yes?  And the pre->
> >compiler doesn't
> >look at any (or at least doesn't process any) C/C++ syntax -
> >it just looks for it's own directives.
>
>Not when I evaluated Intel C++. It was fully installed. I tried to compile a
>program and got an error from the MSVC++ compiler as it parsed through my
>code to generate dependency information. Since MSVC failed, the compilation
>could not continue. The Intel C++ didn't even get a chance to start. If you
>don't believe me evaluate it for yourself for free. And try something like
>partial specialization on a template.

All I can say is that using Intel C++ in place of MSVC overcame the 
problems I was having.


>-----Original Message-----
>From: owner-qt-interest@trolltech.com
>[mailto:owner-qt-interest@trolltech.com]On Behalf Of Rick Berger
>Sent: Tuesday, March 12, 2002 4:08 PM
>To: qt-interest@trolltech.com
>Subject: RE: Alternatives to VC++ IDE?
>
>
> > > compile anything that VC++ will not compile because it still uses it to
> > > produce dependency information. Anyway, that's what I ran into when I
> > > evaluated it.
> >
>So, I would think you would be able to write the C++ to the back-end
>compiler, even though you may have to write to the MSVC
>pre-compiler.
>
>My 2c
>rickb
>
>--
>List archive and information: http://qt-interest.trolltech.com
>
>--
>List archive and information: http://qt-interest.trolltech.com

Regards
John,

-----
john@theKompany.com
http://www.theKompany.com


Pages: Prev | 1 | 2 | 3 | 4 | Next