In article <3aa508f0@news.povray.org> , "Scott Hill" 
<sco### [at] ncgraphicsnet> wrote:

I apologize in advance for my very clear words.  They are not personal.
Please don't be offended.

>     So, why not use a cross-platform STL implementation ?
>
>     See : http://www.stlport.org

You are missing the point.  The point is not having to _deal_ with the
hassle of incompatible STLs.  Limiting to a specific implementation that
works on a subset of compilers is by no means a solution to the problem.
The goal is not writing a program using STL but writing a ray-tracer,
thus we need tools that work _flawlessly_ not not those one first has to
fight with for days to know their limitations.

>        - SUN Workshop C++ Compilers 4.0.x-5.0, 6.0EA, for Solaris
>         - GNU GCC 2.7.2-2.95 Compilers, for all GCC platforms
>         - IBM xlC 3.1.4.0, for AIX
>         - IBM OS/390 C/C++ 1.x - 2.x Compiler, for OS/390 (STLport is the
> only available and official ANSI library for this platform)
>         - IBM VisualAge/ILE C++ for AS/400 Version 4 Release 4, for AS/400
>         - IBM Visual Age C++ 3.x, for OS/2 and Win32
>         - HP ANSI C++ Compiler, for HP-UX
>         - SGI MIPSpro C++ 7.x, for IRIX
>         - SGI MIPSpro C++ 4.0, for IRIX
>         - DEC C++ V5.5-004, V6.x for OSF 3.2/4.0, for Digital Unix.
>         - Tandem NCC, for Tandem OS
>         - SCO UDK C++, for UnixWare
>         - Apogee C++ 4.0, for SUN Solaris
>         - KAI C++ 3.1c, for all KAI platforms
>         - Portland Group C++ compiler ( pgCC), for all pgCC platforms
>         - Microsoft Visual C++ 4.x - 6.x, for Win32
>         - Borland C++ 5.02-5.5, for Win16/32
>         - Metrowerks CodeWarrior Pro 1.8-5.3, for MacOS, Win32
>         - Intel (R) C++ 4.0 Reference Compiler, for Win32
>         - Watcom C++ 10.x-11.0, for DOS, Win16, Win32
>         - Powersoft's Power++ 2.0, for Win32
>         - Symantec C++ 7x/8.x, for MacOS and Win32
>         - Apple MPW and MRC compilers, for MacOS

Half of the compilers on this list are not even _able_ to support STL.
Why?  Some of them don't support templates (like gcc 2.7.2 or MrC 3.x or
VC 4.x or Symantec C++, which has not been updated for more than four
years at least on the Mac) in any way even close to any pre-standard.
The very fact that this STL had to be "ported" shows that it is a single
big hack, no a usable piece of software.  Being able to compile code
does not imply that it is fit for any purpose...

> basically the message was that, due to the inconsistency
> in the various compiler providers STL implementations, one would not be able
> to use the STL within the POV 4.0 source code.

No, you misunderstood.


Last but not least, do not worry about this issue.  The team will find a
working compromise, but as pointed out in previous posts about 4.0:
There is no point discussing 4.0 features because nothing has been
decided yet.  Do don't spend your time on things that are not needed.
The world is full of interesting and more rewarding problems seeking a
solution.  Or, if you have a lot of time, just enjoy ray-tracing :-)

     Thorsten


NOTE: Opinions expressed are my own and do not necessarily reflect those
of the POV-Team.

____________________________________________________
Thorsten Froehlich, Duisburg, Germany
e-mail: tho### [at] trfde

Visit POV-Ray on the web: http://mac.povray.org

Post a reply to this message