c4: a Forth system inspired by Tachyon and ColorForth

Tachyon's influence on C4

In C4, a program is a sequence of WORD-CODEs.
A WORD-CODE is a 32-bit unsigned number (a DWORD).
Primitives are assigned numbers sequentially from 0 to BYE.
If a WORD-CODE is less than or equal to BYE, it is a primitive.
If the top 3 bits are set, it is a 29-bit unsigned literal, 0-$1FFFFFFF.
If it is between BYE, and $E0000000, it is the code address of a word to execute.

ColorForth's influence on C4

C4 supports control characters in the whitespace that change the state.
C4 has 4 states: INTERPRET, COMPILE, DEFINE, AND COMMENT,
C4 also supports the standard state-change words.

Ascii	Word	State	Description
$01	]	1	Compile
$02	:	2	Define
$03	[	3	Interpret/execute/immediate
$04		4	Comment
	(	4	Comment, save current state
	)		End comment, restores saved state

NOTE: In the DEFINE state, C4 changes the state to COMPILE after adding the next word.
NOTE: Unlike ColorForth, ';' compiles EXIT and then changes the state to INTERPRET.

Building C4

Windows

• There is a .SLN file (either 32- or 64-bit)

Linux

• 32-bit: There is a makefile; use 'ARCH=32 make'
• 64-bit: There is a makefile; use 'make'

C4 memory usage

C4 provides a single memory area. See 'mem-sz' for its size.

• The total size of this memory is MEM_SZ (see c4.h)
• It is broken into 3 areas as follows [CODE][VARS][Dictionary].
• The CODE area is an aray of WORD-CODEs starting at the beginning of the memory.
• The VARS area is defined to begin at address &memory[CODE_SLOTS*WC_SZ].
• The Dictionary starts at the end and grows downward.
• The size of the CODE area is CODE_SLOTS (see c4.h). Modify this as desired.
• here is an offset into the CODE area, the next slot to be allocated.
• last is an also offset into the memory area.
• vhere is the absolute address of the first free byte the VARS area.
• Use ->memory to turn an offset into an address into the memory area.
• NOTE: CODE slots 0-25 (0 wc@ .. 25 wc@) are reserved for C4 system values.
• NOTE: CODE slots 26-75 (26 wc@ .. 75 wc@) are unused by C4.
• NOTE: These are free for the application to use as desired.
• NOTE: Use wc@ and wc! to get and set WORD-CODE values in the CODE area.

WORD	STACK	DESCRIPTION
memory	(--A)	A: starting address of the C4 memory
vars	(--A)	A: starting address of the VARS area
mem-sz	(--N)	N: size in BYTEs of the C4 memory
dstk-sz	(--N)	N: size in CELLs of the DATA and RETURN stacks
tstk-sz	(--N)	N: size in CELLs of the A and T stacks
wc-sz	(--N)	N: size in BYTEs of a WORD-CODE
de-sz	(--N)	N: size in BYTEs of a dictionary entry
(dsp)	(--N)	N: CODE slot for the data stack pointer
(rsp)	(--N)	N: CODE slot for the return stack pointer
(lsp)	(--N)	N: CODE slot for the loop stack pointer
(tsp)	(--N)	N: CODE slot for the T stack pointer
(asp)	(--N)	N: CODE slot for the A stack pointer
(here)	(--N)	N: CODE slot for the HERE variable
(vhere)	(--A)	A: address of the VHERE variable
(last)	(--N)	N: CODE slot for the LAST variable
base	(--N)	N: CODE slot for the BASE variable
state	(--N)	N: CODE slot for the STATE variable

C4 Strings

Strings in C4 are NULL-terminated with no count byte.

Format specifiers in ftype and ."

Similar to the printf() function in C, C4 supports formatted output using '%'.
For example : ascii dup dup dup ." char %c, decimal #%d, binary: %%%b, hex: $%x%n" ;.

Format	Stack	Description
%b	(N--)	Print TOS in base 2.
%c	(N--)	EMIT TOS.
%d	(N--)	Print TOS in base 10.
%e	(--)	EMIT escape (#27).
%i	(N--)	Print TOS in the current base.
%n	(--)	Print CR/LF (13/10).
%q	(--)	EMIT " (#34).
%s	(A--)	Print TOS as a string (formatted).
%S	(A--)	Print TOS as a string (unformatted).
%x	(N--)	Print TOS in base 16.
%[x]	(--)	EMIT [x].

The A stack

C4 includes an A stack.
This is somewhat similar to MachineForth's operations for 'a', but in C4, it is a stack.
The size of the A stack is configurable (see tstk-sz).

WORD	STACK	DESCRIPTION
>a	(N--)	Push N onto the A stack.
a!	(N--)	Set A-TOS to N.
a@	(--N)	N: copy of A-TOS.
a@+	(--N)	N: copy of A-TOS, then increment A-TOS.
a@-	(--N)	N: copy of A-TOS, then decrement A-TOS.
a>	(--N)	Pop N from the A stack.
adrop	(--)	Drop A-TOS

The T Stack

C4 includes a T stack, with same ops as the A stack.
Note that there are also additional words for the return stack.

WORD	STACK	DESCRIPTION
>t	(N--)	Push N onto the T stack.
t!	(N--)	Set T-TOS to N.
t@	(--N)	N: copy of T-TOS.
t@+	(--N)	N: copy of T-TOS, then increment T-TOS.
t@-	(--N)	N: copy of T-TOS, then decrement T-TOS.
t>	(--N)	Pop N from the T stack.
tdrop	(--)	Drop T-TOS

Inline words

In c4, an "INLINE" word is similar to a macro. When compiling a word that is INLINE, c4 copies the contents of the word (up to, but not including the first EXIT) to the target, as opposed to compiling a CALL to the word. This improves performance, but uses extra space.

Note that if a word might have an embedded 7 (EXIT) in its implementation (eg - a byte in an address for example), then it should not be marked as INLINE.

C4 WORD-CODE primitives

Stack effect notation conventions:

TERM	DESCRIPTION
SZ/NM/MD	String, uncounted, NULL terminated
A	Address
C	Number, 8-bits
WC	WORD-CODE, 32-bits
N/X/Y	Number, CELL sized
F	Flag: 0 means false, <>0 means true
FH	File handle: 0 means no file
I	For loop index counter

The primitives:

WORD	STACK	DESCRIPTION
(lit)	(--WC)	WC: WORD-CODE for LIT primitive
(jmp)	(--WC)	WC: WORD-CODE for JMP primitive
(jmpz)	(--WC)	WC: WORD-CODE for JMPZ primitive
(jmpnz)	(--WC)	WC: WORD-CODE for JMPNZ primitive
(njmpz)	(--WC)	WC: WORD-CODE for NJMPZ primitive
(njmpnz)	(--WC)	WC: WORD-CODE for NJMPNZ primitive
(exit)	(--WC)	WC: WORD-CODE for EXIT primitive
exit	(--)	EXIT word
dup	(X--X X)	Duplicate TOS (Top-Of-Stack)
swap	(X Y--Y X)	Swap TOS and NOS (Next-On-Stack)
drop	(N--)	Drop TOS
over	(N X--N X N)	Push NOS
@	(A--N)	N: the CELL at absolute address A
c@	(A--C)	C: the CHAR at absolute address A
d@	(A--D)	D: the DWORD at absolute address A
wc@	(N--WC)	WC: the WORD-CODE in CODE slot N
!	(N A--)	Store CELL N to absolute address A
c!	(C A--)	Store CHAR C to absolute address A
d!	(D A--)	Store DWORD D to absolute address A
wc!	(WC N--)	Store WORD-CODE WC to CODE slot N
+	(X Y--N)	N: X + Y
-	(X Y--N)	N: X - Y
*	(X Y--N)	N: X * Y
/	(X Y--N)	N: X / Y (integer division)
/mod	(X Y--M Q)	M: X modulo Y, Q: quotient of X / Y
1+	(X--Y)	Increment TOS
1-	(X--Y)	Decrement TOS
<	(X Y--F)	F: 1 if (X < Y), else 0
=	(X Y--F)	F: 1 if (X = Y), else 0
>	(X Y--F)	F: 1 if (X > Y), else 0
0=	(N--F)	F: 1 if (N=0), else 0
and	(X Y--N)	N: X AND Y
or	(X Y--N)	N: X OR Y
xor	(X Y--N)	N: X XOR Y
com	(X--Y)	Y: X with all bits flipped (complement)
for	(N--)	Begin FOR loop with bounds 0 and N-1.
i	(--I)	N: Current FOR loop index.
next	(--)	Increment I. If I >= N, exit, else start loop again.
>r	(N--)	Push N onto the return stack
r!	(N--)	Set R-TOS to N
r@	(--N)	N: copy of R-TOS
r@+	(--N)	N: copy of R-TOS, then increment it
r@-	(--N)	N: copy of R-TOS, then decrement it
r>	(--N)	Pop N from the return stack
rdrop	(--)	Drop R-TOS
>t	(N--)	Push N onto the T stack
t!	(N--)	Set T-TOS to N
t@	(--N)	N: copy of T-TOS
t@+	(--N)	N: copy of T-TOS, then increment T-TOS
t@-	(--N)	N: copy of T-TOS, then decrement T-TOS
t>	(--N)	Pop N from the T stack
tdrop	(--)	Drop T-TOS
>a	(N--)	Push N onto the A stack
a!	(N--)	Set A-TOS to N
a@	(--N)	N: copy of A-TOS
a@+	(--N)	N: copy of A-TOS, then increment A-TOS
a@-	(--N)	N: copy of A-TOS, then decrement A-TOS
a>	(--N)	Pop N from the A stack
adrop	(--)	Drop A-TOS
emit	(C--)	Output char C
;	(--)	Compile EXIT, set STATE=INTERPRET
lit,	(N--)	Compile a push of number N
next-wd	(--A L)	L: length of the next word (A) from the input stream
		- If L=0, then A is an empty string (end of input)
immediate	(--)	Mark the last created word as IMMEDIATE
inline	(--)	Mark the last created word as INLINE
outer	(S--)	Send string S to the C4 outer interpreter
addword	(--)	Add the next word to the dictionary
timer	(--N)	N: Current time
see X	(--)	Output the definition of word X
ztype	(S--)	Print string at S (unformatted)
ftype	(S--)	Print string at S (formatted)
s-cpy	(D S--D)	Copy string S to D
s-eq	(D S--F)	F: 1 if string S is equal to D (case sensitive)
s-eqi	(D S--F)	F: 1 if string S is equal to D (NOT case sensitive)
s-len	(S--N)	N: Length of string S
z" str"	(--)	- COMPILE: generate code to push address of str (vhere)
	(--A)	- RUN: A=address of str
	(--A)	- INTERPRET: A=address of str (only 1 string supported)
." msg"	(--)	- COMPILE: execute z", compile ftype
	(--)	- RUN: perform ftype on msg
	(--)	- INTERPRET: output msg using ftype
find	(--XT A)	XT: Execution Token, A: Dict Entry address (0 0 if not found)
loaded?	(XT A--)	Stops current load if A <> 0 (see find)
fopen	(NM MD--FH)	NM: File Name, MD: Mode, FH: File Handle (0 if error/not found)
fclose	(FH--)	FH: File Handle to close
fdelete	(NM--)	NM: File Name to delete
fread	(A N FH--X)	A: Buffer, N: Size, FH: File Handle, X: num chars read
fwrite	(A N FH--X)	A: Buffer, N: Size, FH: File Handle, X: num chars written
fgets	(A N FH--X)	A: Buffer, N: Size, X: num chars read (0 if EOF/Error)
include X	(--)	Load file named X (X: next word)
load	(N--)	N: Block number to load (file named "block-NNN.fth")
load-next	(N--)	Close the current block and load block N next
system	(S--)	PC ONLY: S: String to send to system()
bye	(--)	PC ONLY: Exit C4

C4 default words

Default words are defined in function sys_load() in file sys-load.c.
For details, or to add or change the default words, modify that function.