Information about BIF as of April 2019.

	Joel Matthew Rees, Amagasaki, Hyogo, Japan.
					https://ja.osdn.net/projects/bif-6809/
					joel.rees+knock@gmail.com
					http://reiisi.blogspot.com
					https://defining-computers.blogspot.com/
					https://ja.osdn.net/users/reiisi/
					https://sourceforge.net/u/reiisi/profile/
					etc.
	Copyright 2000, 2019 Joel Matthew Rees

-----

(The following can be found in its original form in the bif-c 
repositories. bif-c is buggy, by the way.)

-----

BIF was a dialect of the early fig-standard FORTH.  It was a student
project, not a commercial product, and no warranty has ever been made 
concerning it, whatsoever.  It was written on the Color Computer 2 in 
6809 assembler; the assembler used was disk EDTASM+.  It also ran on 
the Color Computer 3, with the expected limitations of requiring the 
32 column screen, etc. 

License information -- the rights to BIF and the materials with it, 
and restrictions on distribution -- are described in the file 
BIFDOC.TXT, which really should distributed with the source along with 
this README.TXT. 

The short version of the licensing information is that I am distributing 
BIF6809 under the essential terms of the Internet Systems Consortium 
ISC License:

=========
Permission to use, copy, modify, and/or distribute this software for 
any purpose with or without fee is hereby granted, provided that the 
accompanying copyright notices and this permission notice appear in 
all copies.

THE SOFTWARE IS PROVIDED “AS IS” AND ISC DISCLAIMS ALL WARRANTIES 
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF 
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL ISC BE LIABLE FOR ANY 
SPECIAL, DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES 
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN 
AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, 
ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS 
SOFTWARE.
=========

I add here the stipulation that I claim right to the word "BIF" as 
the name of a programming language.

If you are going to distribute or redistribute the obect or source of 
bif in any of its forms, it really makes no sense not to include the 
BIFDOC.TXT and this README.TXT. If you do something like that and 
you or anyone that gets the results has problems with it, and you come 
to me looking for help, expect to be teased mercilessly about it. And 
expect to be on the bottom of my priority list, not out of spite, 
out of self-protection.

-----

I have reconstructed something approximating my source disks (using the 
great emulator, xroar: https://www.6809.org.uk/xroar/ and the imgtool
distributed with MAME, along with the *nix tools). Some useful commands 
are noted in commands.txt.

The name of the disk image is (appropriately?) bifsource.dsk. It can be 
directly attached by xroar's disk emulator tools, and should be similarly 
useable with VCC, MAME, etc. al.

-----

The assembler source files are readable by the a variety of text editors,
that respond flexibly about line termination. But if you try editing them
with a normal editor and then assembling the result with EDTASM+, be very 
careful about line termination. Be prepared to use *nix command line tools 
like tr, cut, sed/awk, perl, etc. 

The source files retain the original line numbers as used by EDTASM+, 
which is why you will want the *nix tools. But I have included source to a 
short C program, stripln.c, which strips the line numbers. This program 
may also be useful for converting line endings, if that is necessary. Look 
for it in the junkbox directory, and be prepared to compile it. 

If you are under the burden of using Microsoft OSses, the Cygwin project 
should be of quite a bit of assistance, including the *nix command line 
tools and gcc and clang, etc.

-----

Macros are used in the source, but shouldn't cause too much confusion 
to a programmer with assembler experience. If anyone is brave enough 
to try to port it, key routines will be EMIT, KEY, ?TERMINAL, CR, R/W, 
and several routines in the EDITOR vocabulary where I used direct 
video I/O out of laziness.

-----

BIF might be useful for experimentation and for learning about FORTH, 
for someone who has access to a Color Computer or an emulator. The 
executable file may be run on the Color Computer by LOADMing it via 
Color Computer Disk BASIC. (More details in BIFDOC.TXT.) 

It is assembled to be EXECed at hexadecimal 1300 (&H1300). 

-----

One specific warning must be given:

DO NOT ATTEMPT TO ACCESS ORDINARY DISKS FORMATTED FOR USE BY OTHER
OPERATING SYSTEMS WHILE BIF IS RUNNING! Because of disk buffering, it 
will be difficult to avoid unintentional writes to the disk. 

If you do try looking at an OS-9 or Color Computer DOS (etc.) disk with 
BIF, don't blame me if you destroy the directory and/or other valuable 
data.

-----

I tried a re-write in C (bif-c), but it's not working well. Issues with 
file systems (no real file system in BIF) and the current standardization 
committees ideas about what is meaningful code, problems with my own 
attention span, etc. keep it full of bugs. And I used a number of neat 
tricks that get in the way in C. 

One thing I want to retain is the use of nested binary trees in the 
symbol table. 

Incidentally, said re-write has been in process for over thirty years.
Real life keeps getting in the way.

-----

I am including source to a Q&D C program, 32col.c, which will re-format 
files extracted from BIF disks for normal text editors. Mac-isms and 
Codwarrior-isms can probably be discerned by comparing it with stripln.c, 
mentioned above. 

When I brought the BIF high-level source with me, I used some feature of 
Color Computer BASIC and BIF that I have forgotten to dump some of the 
Forth style screen listings. 

The program stripln can be used to strip line numbers from such screen 
listings, as well. 

The documentation is ASCII text, with CR/LF line termination, and should
be examined carefully by anyone considering a port.  Bear in mind that
it was written toward Color Computer users.

I apologize for not alphabetizing the FORTH words by name. I did it once 
with a C program, but got too ambitious and lost the results in the 
process of trying to split it up into modules. I haven't had enough time 
to finish the modularization, yet, either. (Getting an iBook so I could 
work on the train just gave me more things to do on the train.)

-----

There are three subdirectories at this point:

junkbox -- 

Various tools I've used. Most are one-time tools that might be useful 
again in a similar situation.

edtasm_v --

This is where I'm recreating what I used at college, to use as a baseline
in further projects.

cross_v --

Once I have the baseline re-established, and am able to compile the 
assembler and other tools, I'll switch to using cross-development tools
to reorganize and restructure the code to make it more generally useful.

-----

For the time being, until I get it more-or-less back to the state it 
was in when I used it in college, I'm focusing on the source in the
edtasm_v subdirectory.

===================

Files under edtasm_v --

6809 Assembly Language Source files under edtasm_v:

README.TXT
	this file.
BIFDOC.TXT
	general explanations, including descriptions of every word.
BIFU.I
	structure of the per-user variable page.
BIF.M
	macros, including the inner interpreter (basis of the virtual machine),
	the dictionary (symbol table) structure offsets,
	and invocations for the fundamental objects.
BIFDP.A
	things kept in the direct page, 
	including the behaviours for the fundamental objects (was not a good idea after all),
	and the index to the per user variable page. 
BIFST.A
	cold and warm boot routines and the initial value table for the per-user variable page.
BIF.ASM
	the main source file (includes other parts),
	basic expression evaluation, more of the inner interpreter, 
	basic vocabulary access, basic symbol parsing.
BIFB.A
	basic I/O, more of the inner interpreter, extended expression evaluation,
	the rest of the basic symbol table access.
BIF1.A
	data movers, common expression evaluation,
	stack pointer access, more of the inner interpreter,
	high-level compiler.
BIF1B.A
	common expression evaluation, extended expression evaluation,
	innards of the high-level compiler, more of the high-level compiler,
	compiler directive.
BIF2.A
	more common expression evaluation, common constants,
	I/O constants, character typing constants,
	symbol table globals, compiler globals, parser globals, I/O globals.
BIF2B.A
	compiler globals, more high-level compiler,
	more common expression evaluation, formatted output.
BIF3.A
	more basic symbol table, symbol table, more compiler, more formatted output,
	more data movers, more low-level parser (formatted input), more I/O,
	more extended expression evaluation, more expression evaluation, 
	more compiler directives, an extension to the inner interpreter.
BIF3B.A
	more formatted output, more innards of the high-level compiler,
	more high-level compiler.
BIF4.A
	more innards of the expression evaluator, more common expression evaluation,
	more I/O (buffer handling).
BIF4B.A
	more high-level compiler, more compiler directive.
BIF5.A
	more innards of the high-level compiler, more I/O (buffering),
	disk access, error handling, more formatted output.
BIF5B.A
	more error handling, screen-based sector (character) editor.
BIF6.A
	more parser (formatted input), I/O (terminal), compiler (input),
	symbol table (lookup).
BIF6B.A
	symbol table, compiler innards, null vector test, 
	more screen-based sector editor.
BIF7.A
	compiler, formatted output, compiler directives
BIF7B.A
	error handling, symbol tables, compiler directives.
stripln.c,
32col.c, 
	C language source and Macintosh executables for stripping line 
	numbers and reformatting 32 column source code "screens". The two 
	XXX.GXX.out files below are output of the 32col program.

-----

Hopefully, I will shortly have time to reconstruct useful things from the 
following files on the tools.dsk disk image and/or the cs431 disk image:

TOOLS.G00, TOOLS.G00.out
	FORTH source for disk listing, screen handling, definition dumping, 
	sector copying, forward referencing, buffer maintenance, 
	experimenting with hardware, double (32 bit) integer math, etc.,
	and a post-fix assembler.
PAIRS.G28, PAIRS.G28.out
	a "database" example from one of my FORTH books.
TOOLS_G00_ERRORS.text
	Contains the tools output readable in regular text editor format 
	and the error messages, with their corresponding number in 
	hexadecimal. I should make a separate file for the error messages
	(or something).

and

SCR33.ARR
	arrays for CS431.
SCR34.LOC
	some math for CS431.
SCR40.431
	test suite for CS431.

===================


