c4 supports control characters in the whitespace that change the state.
c4 has 'A', 'B' and 'T' stacks, inspired by ColorForth's 'a' register.
c4 has 4 states: INTERPRET, COMPILE, DEFINE, AND COMMENT,
c4 also supports the standard state-change words.

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.
NOTE: The ( word simply skips to the next ) word, no nesting.

In c4, a program is a sequence of OPCODEs.
An OPCODE is an unsigned CELL, 32 or 64 bits.
Primitives are assigned numbers sequentially from 0 to BYE.
If the OPCODE is less than or equal to BYE, it is a primitive.
Else if the top 3 bits are set, it is an unsigned literal.
Else, it is the offset in the CODE area (XT) of a word to execute.

c4 has a built-in editor. See Editor.md for details.
The editor can be excluded from c4 by undefining EDITOR in c4.h.
It is built-in so that the editor is available when running c4 from any folder.
In c4, the size if a block is 3072 bytes (3x1024).
The editor is 32 lines, 96 columns, and has a VI like feel.

• There is a Visual Studio solution file, c4.sln (either 32- or 64-bit)

• There is a makefile.
• The default architecture is 64-bits. That is a little faster on my systems.
• 64-bit: use 'make'
• 32-bit: use 'ARCH=32 make'

c4 provides a single memory area with size 'mem-sz' (see c4.h, MEM_SZ).

• It is divided into 3 areas as follows [CODE][VARS][Dictionary].
• The CODE area is an array of OPCODEs 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-sz' (see c4.h, CODE_SLOTS).
• here is an offset into the CODE area, the next slot to be allocated.
• last is the address of the most recently created word.
• vhere is the address of the first free byte the VARS area.
• Use ->memory to turn an offset into an address into the memory area.
• Use ->code to turn an offset into an address of an OPCODE.
• NOTE: CODE slots 0-25 (0 wc@ .. 25 wc@) are reserved for c4 system values.
• NOTE: CODE slots 26-BYE (26 wc@ .. <bye> 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 OPCODE values in the CODE area.

Strings in c4 are NULL-terminated with no count byte.

Similar to the printf() function in C, c4 supports formatted output using '%'.
For example : ascii ( n-- ) dup dup dup ." char %c, decimal #%d, binary: %%%b, hex: $%x%n" ;.

c4 includes A, B and T stacks.
These are similar to ColorForth's operations for 'a', but in c4, they are stacks.
The size of the stacks is 'tstk-sz' (see c4.h, TSTK_SZ).
Note that there are also additional words for the return stack.

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 (the opcode value for EXIT) in its implementation, then it should not be marked as INLINE.

c4 provides 10 temporary words, 't0' thru 't9'.

• Defining a temporary word does not add an entry to the dictionary.
• Temporary words are intended to be helpful in factoring code.
• Their names are case-sensitive (t0-t9, not T0-T9).
• A temporary word can be redefined as often as desired.
• When redefined, code references to the previous definition are unchanged.
• t0 thru t5 are 'normal', t6 thru t9 are 'inline'.

The primitives:

If SYS_LOAD is defined in file c4.h, call function sys_load().
Else, load file block 0 (file block-000.fth).
For details, or to add or change the default words, modify that function.