[A92] Re: Forth-92
[A92] Re: Forth-92
>I've already completed an ANS Forth System for the TI-92/Fargo.
Groovy. Any help you can give is much appreciated.
Most of my effort right now is in porting pForth, the previously-mentioned
ANSI C Forth. However, it -is- a 32-bit Forth, and thus subject to the
issues you mentioned. It's basically just a first step. :-)
>If you're interested, I can send you the complete source. Btw, I solved
>the dictionary problem, by making just every Forth address 16bits and
>addressing via a Base pointer eg. 0(BP,d0.w) .
*nods* That's how I wound up doing it, before I found this -other- Forth. :-)
>The Forth system I wrote is unfortunately written using a selfmade
>CPP like preprocessor. This is quite pratical for writing high-level Forth
>code, especially control flow structures. But it's also ugly and not very
Hrm. Well, I'd love to take a look all the same. :-)
>A Forth cross compiler would be very, very much nicer. It could
>output an object file, that could be linked by `flink' for Fargo, or other
>tools (I don't know) for TI-92+/TI-89. Documentations about the object
>format (some Amiga thing) can be found at ticalc.org. I also have an 68k
>Forth-assembler, that I could send you.
We have a couple of linkers. The object format I'm using right now is a
little different (just because I've switched over to the GNU as assembler
instead of A68K, for use with GCC), but I actually prefer A68K's syntax. :-)
>A cross-compiler could also be used as a portable "high-level" programming
>language for TI-92(+)/89 programming, using a PC as host (who'd want to
>use ti-gcc, if a forth compiler is available? ;-)
That's my goal with this. Well, actually, my goal is for the PC not to
even need to be involved (have the full Forth interpreter/monitor on the
calculator). Since Forth (compiled or not) has a speed advantage over C, I
don't see any question. :-)
>If you're interested in writing your Forth system as a team-work, please
>let me know. I might not have much time, but since you own a TI-92+, that
>work might actually be useful.
Absolutely. I get the feeling there isn't a lot of interest in this on
this list (either that, or there's only about four people that still read
it), so you might want to contact me directly. The port of pForth is about
done--I just have to fake stdin and stdout because they're not implemented
in TIGCC--giving me time to work on bigger and better things. :-)