Hello Mike!

Replying to a message of Mike Luther to Darin McBride:

 ML> My whole world world (Well not quite!) revolves around multiple
 ML> dimension variable length string arrays as well as automatic garbage
 ML> collection.  For research purposes I've managed to create a

std::vector > > should do pretty
much all of the above.

 ML> cross-source translator from PowerBASIC 3.5 for DOS to Watcom C/C++
 ML> for about 135 of the 225 total functions I use in PowerBASIC.  So
 ML> done, I have a great deal of respect for Thales and some other folks
 ML> who purport to offer very expensive translation services .. and the
 ML> specificity of the particular code style for which they will do that
 ML> ... chuckle.

 ML> I've not worked with this for a good while, partly because of the
 ML> crush of things real I have to do with my projects, partly because of
 ML> my failure to be able to handle multiple dimension char arrays by and
 ML> between CALLS to SUBS that I need in Watcom C/C++ Version 11.   I
 ML> have the latest release, but haven't worked with it since I upgraded
 ML> all the stuff to the free version.

I don't write anything for OS/2 anymore.  My primary compiler seems to be
perl v5.8.0 ;-)  For C/C++, I use gcc on Linux, followed very distantly by
IBM C++ for AIX.  While I know my code also compiles on Sun, HP, and
others, I've not really done anything for that.

 ML> From a thinking standpoint, I believe I would like to return to the
 ML> issue of conversion of the following from BASIC to C/C++ with
 ML> automatic allocation and deallocation of memory plus the ability to
 ML> work with in structs as opposed to UDT's, calls to subs and functions
 ML> ...

The following constructs are foreign to me.  Care to explain what the
intention of this code is?  This isn't a BASIC echo, and I don't frequent
BASIC echos on purpose - that purpose being I don't care to know that much
about BASIC in general ;-)

 ML>        DIM X$(10, 25)  ' Where X$ is a variable length string

 ML>        DIM X$(-RDIM% to +RDIM%) ' same comment

 ML>        DIM X$(-RDIM% to +RDIM%, -SDIM% to +SDIM%) ' same comment.

 ML> Yes, I *DO* use this all the time in PowerBASIC.  Direct memory writes
 ML> to move whole professional templates between massive action switches
 ML> on the desktop is horribly fast and efficient for me.

This paragraph went way over my head.

 ML> No, this is *NOT* a bully-bait, my dog is bigger than your dog issue.

 ML> Whether you know it or not, IBM has given up on Visual Age for C++ as
 ML> a compiler for Mozilla for OS/2 and is moving to GCC+ for all future
 ML> browser work .. platform wide, I think.

Doesn't surprise me.  IBM has given up on OS/2 in general.  (The IBMers who
compile Mozilla for OS/2 probably don't do it as part of their job duties,
but more likely in their spare time on IBM hardware.  Saying "IBM has
given up" is probably misleading in that corporate IBM probably never
had anything to do with this.)

 ML> GCC whatever is now being used to port MAX to LINUX now, as I'm sure
 ML> everyone who has the MUFFIN echo knows.

I have it, I don't know.  Your assumption that having it and reading it are
the same thing is a little faulty ;-)

 ML> Yes, PowerBASIC has been promised for LINUX and now .. officially, in
 ML> writing, by Bob as for, "other platforms", as well!

Heck, there's a BASIC available for Parrot - a new virtual machine being
developed for Perl 6.

 ML> But any thoughts here and a return to this cesspool would be
 ML> appreciated.

Cesspool?

Darin

---