BIF-C is a near fig-FORTH implemented in C.

The first release version is mostly a direct transliteration of BIF-6809, 
a near fig-FORTH I wrote for school many years ago. 

This collection of files (step 06) is from a backup taken at the point 
where I finally had stable post-fix expression evaluation and functional 
colon definitions. It has a lot of debugging code and macros to turn it 
on and off (see bifu_i.h). 

It compiles and runs on PPC Mac OS X 10.3, and on Fedora 11 PPC. 
Compiles and can find words on Fedora 12 x86, but doesn't parse numbers.

Still doesn't handle files, still needs work on the error messages, 
all sorts of things need to be shimmed or cleaned up a bit.

This is the first release version. This is still, however, primarily for 
my personal experimentation and such, as well as to let other interested 
people take a look at the fun I'm having.

For those wanting a standard FORTH, other FORTHs, such as gforth, are 
recommended. 

See LICENSE.TXT for terms of use.

Plans:

I plan to set up a proper file interface, although I'm not sure what 
form it will take yet. I also plan to add a bit of regular expression 
parsing stuff.

I plan to strip the static initial symbol table and build it at load 
time, to make it possible to build on older C compilers/environments, 
and to make it somewhat easier to restructure the symbol table. 

I should then be able to get it running on the classic Mac, which 
would be a sweet, if somewhat meaningless reward to myself. 

If I'm lucky, I may get the time and opportunity to get it running on 
the old 16 bit OS-9 6809 on my old Color Computer hardware stored on 
the other side of the ocean. (Would take some careful planning and 
optimization.) But I don't really expect to ever get the chance; 
other things will probably take precedence.

I also plan to modularize the code significantly, hiding utility 
words in their own vocabularies and such. During this step, I hope to 
define a kernel set of definitions and convert them all to C code. In 
the process, I also hope to develop a set of regression tools in 
FORTH, to fully exercise and test the language.

After modularization of this BIF-C, I will probably consider BIF-C 
version one complete. I'm not sure what I'll call the next step, but 
it will probably include some interesting things like arbitrary 
precision math and a new character encoding scheme I presently have 
on a back burner somewhere. Maybe I'll win the lottery (although I 
don't play it) and get a chance to do compiling to native code on 
some CPU architectures. 

Heh. Counting my chickens before they hatch. Right now I have to say 
I'll be lucky to get the file system and parsing based on something 
other than space working in time to use it in my present job.

Reiisi Kenkyuu is, at the point of this writing, a business operating 
name of Joel Matthew Rees, originally of Odessa, Texas, currently of 
Amagasaki, Hyougo, Japan. Someday I plan to incorporate under the name. 

