[MLton] Status of MLton releases and merging MinGW target?

Anoq of the Sun anoq@HardcoreProcessing.com
Sun, 09 Nov 2003 13:06:17 +0200


Stephen Weeks wrote:
> > compilation with the SDL-library and stuff like that?
> 
> It seems to me like it might help.  Using whatever development
> environment most other people use should be easiest, and that
> certainly rules out cross-compilation.

Actually - with SDL it is only recently that it has become
possible to use MinGW natively on Windows.
Until then there was only cross-compilation :)
(As far as I know...)

The last time I tried to even just install Cygwin on Win NT,
that failed :( And I certainly don't want to trade MLton
with Visual C++ or something like that! :)

I actually work with Visual C++ (6.0) using MFC and
Rational Rose (98) at a freelance project now - and it's
not very impressive... - Rose's code generation is
nice but it's also a bit of a Hazzle sometimes.
Only the fact that everything is automatically
documented as UML diagrams is a true advantage.
And _someone_ sure didn't study algorithms and time
complexity enough with implementing the
array-classes of MFC ;)

Of course all the very Windows specific things are a breeze
with Visual C++ and MFC (e.g. when the code has to run on
both Windows and a Windows CE embedded device where all
strings in Win CE all of a sudden become unicode and the
standard C functions like sprintf etc. disappear...).

But all in all, cross-compilation with MLton is a very
viable option if things has to run on e.g. both Linux
and Windows (and not on e.g. a Windows CE device without
also allocating time to port MLton ;) - but writing in ML
AND porting MLton taken together might be faster in the end
than writing a large app. in C / C++ anyway...)

Sorry that I got carried away ;)


Cheers
-- 
http://www.HardcoreProcessing.com