Added README.md

git-svn-id: https://svn.code.sf.net/p/acme-crossass/code-0/trunk@295 4df02467-bbd4-4a76-a152-e7ce94205b78

ACME_Lib/6502/split.a

@@ -0,0 +1,45 @@
+;ACME 0.97
+!set split_cache = []	; start every pass with an empty cache
+!ifdef lib_6502_split_a !eof
+lib_6502_split_a = 1
+
+!macro split_lo @args {
+	+split_putbytes 0, @args
+	!set split_cache = split_cache + @args
+}
+!macro split_hi @args {
+	+split_putbytes 8, @args
+	!set split_cache = split_cache + @args
+}
+!macro split_lo {
+	+split_putbytes 0, split_cache
+	!set split_cache = []
+}
+!macro split_hi {
+	+split_putbytes 8, split_cache
+	!set split_cache = []
+}
+!macro split_putbytes @shift, @bytes {
+	!if len(@bytes) {
+		!for @idx, 0, len(@bytes) - 1 {
+			!by (@bytes[@idx] >> @shift) & $ff
+		}
+	}
+}
+!eof
+; these macros can be used to split address tables in two separate
+; parts for high- and low-bytes. example:
+
+table_hi
+	+split_hi [$0123, $4567, $89ab, $cdef]	; writes $01, $45, $89, $cd
+	+split_hi [$1122, $3344, $5566, $7788]	; writes $11, $33, $55, $77
+table_lo
+	+split_lo	; writes $23, $67, $ab, $ef, $22, $44, $66, $88 from cache
+
+; of course you can also put the low bytes first:
+
+table_lo
+	+split_lo [$0123, $4567, $89ab, $cdef]	; writes $23, $67, $ab, $ef
+	+split_lo [$1122, $3344, $5566, $7788]	; writes $22, $44, $66, $88
+table_hi
+	+split_hi	; writes $01, $45, $89, $cd, $11, $33, $55, $77 from cache

ACME_Lib/6502/std.a

@@ -1,4 +1,4 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_6502_std_a !eof
 lib_6502_std_a = 1
@@ -22,57 +22,57 @@ lib_6502_std_a = 1
 }
 
 ; increase 16-bit counter
-!macro inc16 .t {
-		inc .t
+!macro inc16 @t {
+		inc @t
 		bne +
-		inc .t + 1
+		inc @t + 1
 +
 }
 
 ; far branches
-!macro bcc .t {
+!macro bcc @t {
 		bcs +
-		jmp .t
+		jmp @t
 +
 }
-!macro bcs .t {
+!macro bcs @t {
 		bcc +
-		jmp .t
+		jmp @t
 +
 }
 
-!macro beq .t {
+!macro beq @t {
 		bne +
-		jmp .t
+		jmp @t
 +
 }
 
-!macro bne .t {
+!macro bne @t {
 		beq +
-		jmp .t
+		jmp @t
 +
 }
 
-!macro bmi .t {
+!macro bmi @t {
 		bpl +
-		jmp .t
+		jmp @t
 +
 }
 
-!macro bpl .t {
+!macro bpl @t {
 		bmi +
-		jmp .t
+		jmp @t
 +
 }
 
-!macro bvc .t {
+!macro bvc @t {
 		bvs +
-		jmp .t
+		jmp @t
 +
 }
 
-!macro bvs .t {
+!macro bvs @t {
 		bvc +
-		jmp .t
+		jmp @t
 +
 }

ACME_Lib/65816/std.a

@@ -1,7 +1,7 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_65816_std_a !eof
-lib_65816_std_a = 1
+lib_65816_std_a = 2
 
 ; Labels and macros for Western Digital's 65c816 processor
 
@@ -12,11 +12,12 @@ lib_65816_std_a = 1
 	cpu_e_reset = $fffc
 	cpu_e_irq   = $fffe
 
-	cpu_n_cop   = $fff4
-	cpu_n_brk   = $fff6
-	cpu_n_abort = $fff8
-	cpu_n_nmi   = $fffa
-	cpu_n_irq   = $fffe
+	cpu_n_cop   = $ffe4
+	cpu_n_brk   = $ffe6
+	cpu_n_abort = $ffe8
+	cpu_n_nmi   = $ffea
+	; no reset vector for native mode because reset always enters emulation mode
+	cpu_n_irq   = $ffee
 }
 
 !macro cpu_emu {; switch to emulation mode
@@ -71,11 +72,11 @@ lib_65816_std_a = 1
 	+i8
 }
 
-!macro inc24 .t {; increase 24-bit counter
-	inc .t
+!macro inc24 @t {; increase 24-bit counter
+	inc @t
 	bne +
-	inc .t + 1
+	inc @t + 1
 	bne +
-	inc .t + 2
+	inc @t + 2
 +
 }

ACME_Lib/cbm/264/basic.a

@@ -0,0 +1,6 @@
+;ACME 0.96.4
+
+!ifdef lib_cbm_264_basic_a !eof
+lib_cbm_264_basic_a = 1
+
+!source <cbm/basic3.5.a>

ACME_Lib/cbm/264/petscii.a

@@ -0,0 +1,46 @@
+;ACME 0.96.4
+
+!ifdef lib_cbm_264_petscii_a !eof
+lib_cbm_264_petscii_a = 1
+
+!source <cbm/petscii.a>
+
+; the sixteen predefined color/luma combinations available via petscii:
+petscii_BLACK		= 144	; $00	here's the actual bits given to TED
+petscii_WHITE		= 5	; $71	(first digit is luma bits,
+petscii_RED		= 28	; $32	second digit is color nibble)
+petscii_CYAN		= 159	; $63
+petscii_PURPLE		= 156	; $44
+petscii_GREEN		= 30	; $35
+petscii_BLUE		= 31	; $46
+petscii_YELLOW		= 158	; $77
+petscii_ORANGE		= 129	; $48
+petscii_BROWN		= 149	; $29
+petscii_YELLOWGREEN	= 150	; $5a
+petscii_PINK		= 151	; $6b
+petscii_BLUEGREEN	= 152	; $5c
+petscii_LIGHTBLUE	= 153	; $6d
+petscii_DARKBLUE	= 154	; $2e
+petscii_LIGHTGREEN	= 155	; $5f
+
+; switching character set (same as C64)
+petscii_LOCK		=   8	; forbid CBM-shift (C128 uses 11 instead)
+petscii_UNLOCK		=   9	; allow CBM-shift (C128 uses 12 instead)
+petscii_LOWERCASE	=  14	; switch to lowercase/uppercase character set
+petscii_UPPERCASE	= 142	; switch to uppercase/graphics character set
+
+; 264-specific stuff
+petscii_FLASHON		= 130	; (C128 uses 15 instead)
+petscii_FLASHOFF	= 131	; (C128 uses 143 instead)
+
+; function keys
+petscii_F1		= 133
+petscii_F3		= 134
+petscii_F5		= 135
+petscii_F7		= 136
+petscii_F2		= 137
+petscii_F4		= 138
+petscii_F6		= 139
+petscii_HELP		= 140	; (C128 uses 132 instead)
+
+petscii_ESCAPE		=  27

ACME_Lib/cbm/264/ted.a

@@ -0,0 +1,215 @@
+;ACME 0.96.4
+
+!ifdef lib_cbm_264_ted_a !eof
+lib_cbm_264_ted_a = 1
+
+; Text Display (TED) 7360/8360
+
+; Definitions:
+; HiRes:	High resolution (320x200 pixels, two colors per 8x8 block)
+; MCM:		Multi-color mode (160x200 pixels, four colors per 4x8 block)
+; Text mode:	Pixel data is taken from charset, index/color/luminance is
+;		taken from video ram.
+; Bitmap mode:	Pixel data is taken from 8 KiB bitmap, color/luminance is taken
+;		from video ram.
+; ECM:		Extended color mode. Like HiRes text mode, but instead of 256
+;		different characters on a shared background color, 64 different
+;		characters on four different background colors are displayed.
+; MC text mode is actually a mixed mode: each character position has its own
+; bit to decide whether it is displayed in HiRes or MC.
+
+; info taken from
+;	http://mclauchlan.site.net.au/scott/C=Hacking/C-Hacking12/gfx.html	(Document Revision B)
+;	http://www.zimmers.net/anonftp/pub/cbm/maps/C16.MemoryMap
+
+; Video RAM is $0800 bytes
+; in text mode:
+;	first $0400 bytes is color/luminance info:
+;		%7.......	flash (ignored in MCM and ECM)
+;		%.654....	luminance
+;		%....3...	MCM: multicolor flag
+;		%.....210	MCM: color
+;		%....3210	HiRes and ECM: color
+;	second $0400 bytes hold screen codes:
+;		%76543210	HiRes and MCM: screen code (in HiRes, bit 7 can
+;					be configured to be "hardware reverse",
+;					see register $ff07)
+;		%76......	ECM: background color index (see registers
+;					$ff15..$ff18)
+;		%..543210	ECM: screen code
+; in bitmap mode:
+;	CAUTION!
+;	LOW nibble of luma corresponds to HIGH nibble of color and vice versa!
+;	first $0400 bytes is luminance info:
+;		%7.......	unused
+;		%.654....	luminance of %0 HiRes pixels or %01 MC pixels
+;		%....3...	unused
+;		%.....210	luminance of %1 HiRes pixels or %10 MC pixels
+;	second $0400 bytes is color info:
+;		%7654....	color of %1 pixels or %10 MC pixels
+;		%....3210	color of %0 pixels or %01 MC pixels
+
+; colors:
+tedcolor_BLACK		= $0	; ignores luminance, black is always black!
+tedcolor_WHITE		= $1	; actually gray, only max luminance gives white
+tedcolor_RED		= $2
+tedcolor_CYAN		= $3
+tedcolor_PURPLE		= $4
+tedcolor_GREEN		= $5
+tedcolor_BLUE		= $6
+tedcolor_YELLOW		= $7	; needs high luminance to look yellow...
+; these cannot be used as main color in MCM:
+tedcolor_ORANGE		= $8
+tedcolor_BROWN		= $9
+tedcolor_YELLOWGREEN	= $a
+tedcolor_PINK		= $b
+tedcolor_BLUEGREEN	= $c
+tedcolor_LIGHTBLUE	= $d
+tedcolor_DARKBLUE	= $e
+tedcolor_LIGHTGREEN	= $f
+; the color names are somewhat useless, because they refer not to the sixteen
+; base colors, but to the sixteen pre-defined luminance/color combinations
+; available by pressing the ctrl or cbm key with a digit key (see the ROM table
+; at $e143 for these combined values).
+; for a "nice" color display, use this order: 2,8,9,7,a,f,5,c,3,d,6,e,4,b,
+; that will give a red/yellow/green/blue/purple spectrum.
+; to display the whole palette, try this program:
+;	10 for i = 0 to 15:color 1, 1
+;	20 read c$:print " " c$ " ";
+;	30 poke 194, 128:for lu = 0 to 7
+;	40 color 1, dec(c$) + 1, lu
+;	50 print "   ";
+;	60 next:print:next:color 1, 1
+;	70 data 0,1,2,8,9,7,a,f
+;	80 data 5,c,3,d,6,e,4,b
+
+; TED registers
+!addr	ted_base	= $ff00
+
+; timers
+ted_timer1_lo	= ted_base + $00	; decrements from the last value written into it
+ted_timer1_hi	= ted_base + $01
+ted_timer2_lo	= ted_base + $02	; runs freely from $ffff
+ted_timer2_hi	= ted_base + $03
+ted_timer3_lo	= ted_base + $04	; runs freely from $ffff
+ted_timer3_hi	= ted_base + $05
+
+; mode bits
+ted_vertctrl	= ted_base + $06	; mostly the same as VIC's $d011
+	tedvert_TEST	= %#.......	; internal test, it should be 0
+	tedvert_EBCM	= %.#......	; extended background color mode
+	tedvert_BITMAP	= %..#.....	; bitmap mode
+	tedvert_ENABLE	= %...#....	; display enable (0: blank screen)
+	tedvert_25ROWS	= %....#...	; 0: 24 rows, 1: 25 rows
+	tedvert_SMOOTH	= %.....###	; vertical smooth scroll (std: %011, higher values move text down)
+
+ted_horzctrl	= ted_base + $07	; most similar VIC-reg is $d016
+	tedhorz_256CHRS	= %#.......	; 0: 128 characters and hardware reverse, 1: 256 characters
+	tedhorz_NTSC	= %.#......	; 0: PAL, 1: NTSC
+	tedhorz_FREEZE	= %..#.....	; if set, TED stops its counters and screen-generating, only single clock and refresh cycles remain.
+	tedhorz_MULTI	= %...#....	; multicolor mode
+	tedhorz_40COLS	= %....#...	; 0: 38 columns, 1: 40 columns
+	tedhorz_SMOOTH	= %.....###	; horizontal smooth scroll (std: %000, higher values move text right)
+
+; i/o
+ted_keyboard	= ted_base + $08	; keyboard input latch (write to trigger sampling, then read input byte)
+	; to scan keyboard, write output byte to $fd30, then write $ff here (to disable joysticks) and read input byte.
+	; to scan joysticks, write $ff to $fd30 (to disable keyboard), then write SELECT value here and read input byte.
+	tedjoy1_SELECT	= %#####.##	; d2 low selects joy #1
+	tedjoy2_SELECT	= %######.#	; d1 low selects joy #2
+	; CAUTION, the joystick bits are low-active, so invert input byte before using these:
+	tedjoy2_FIRE	= %#.......
+	tedjoy1_FIRE	= %.#......
+	tedjoy_RIGHT	= %....#...
+	tedjoy_LEFT	= %.....#..
+	tedjoy_DOWN	= %......#.
+	tedjoy_UP	= %.......#
+
+; interrupts
+ted_irq		= ted_base + $09	; interrupt request register
+	; if bit is set, interrupt has occurred.
+	; to acknowledge interrupts, write back the value read from this register
+	; (setting bits clears them, i.e. writing $82 clears ACTIVE and RASTER
+	; but does not change the COUNTER bits)
+	tedirq_ACTIVE	= %#.......	; 1: TED is signalling interrupt to cpu
+	tedirq_COUNTER3	= %.#......	; 1: counter #3 underflow
+	;		  %..#.....	; always 1 (unused)
+	tedirq_COUNTER2	= %...#....	; 1: counter #2 underflow
+	tedirq_COUNTER1	= %....#...	; 1: counter #1 underflow
+	;		  %.....#..	; always 1 (light pen, not implemented)
+	tedirq_RASTER	= %......#.	; 1: raster line reached
+	;		  %.......#	; always 1 (unused)
+
+ted_mask	= ted_base + $0a	; interrupt mask register (enable/disable)
+	; if bit is set, interrupt is enabled.
+	; use the tedirq_* constants from above.
+	;		  %#.......	; always 1 (unused)
+	;COUNTER3	= %.#......	; 1: enable counter #3 underflow interrupt
+	;		= %..#.....	; always 1 (unused)
+	;COUNTER2	= %...#....	; 1: enable counter #2 underflow interrupt
+	;COUNTER1	= %....#...	; 1: enable counter #1 underflow interrupt
+	;		= %.....#..	; always 1 (light pen, not implemented)
+	;RASTER		= %......#.	; 1: enable raster line interrupt
+	tedmask_LINEBIT8= %.......#	; bit 8 of raster line register $ff0b (see below)
+
+ted_line	= ted_base + $0b	; raster line to trigger interrupt. bit 8 is in bit 0 of $ff0a (see above)
+
+; hardware cursor, given as screen offset (0...999). use 1000 or greater to hide.
+ted_hwcursor_hi	= ted_base + $0c	; %765432.. are unused, %......10 are MSBs of hw cursor offset
+ted_hwcursor_lo	= ted_base + $0d	; hardware cursor offset (low byte)
+
+; sound
+; to calculate the register values:
+; NTSC: reg = 1024 - (111860.781 / frequency_in_Hz)
+; PAL:  reg = 1024 - (111840.45  / frequency_in_Hz)
+ted_sound1_low	= ted_base + $0e	; low 8 bits of voice #1 frequency
+ted_sound2_low	= ted_base + $0f	; low 8 bits of voice #2 frequency
+ted_sound2_hi	= ted_base + $10	; bits 7..2 are unused, bits 1..0 are the highmost 2 bits of voice #2 frequency
+ted_sound_ctrl	= ted_base + $11	; sound control register
+	tedsound_DA	= %#.......	; D/A mode (sound reload)
+	tedsound_NOISE2	= %.#......	; voice #2 noise on/off
+	tedsound_SQUARE2= %..#.....	; voice #2 square wave on/off (if you set both, the square will sound)
+	tedsound_ONOFF1	= %...#....	; voice #1 on/off
+	tedsound_VOLUME	= %....####	; volume, maximum value is 8
+
+ted_misc	= ted_base + $12	; some memory and some sound bits:
+	;		= %##......	; unused
+	tedmisc_BITMAP	= %..###...	; a15/a14/a13 of bitmap in memory
+	tedmisc_ROMCHAR	= %.....#..	; 0: charset is in RAM, 1: charset is in ROM
+	tedmisc_SOUNDMSB= %......##	; highmost bits of voice #1 frequency
+
+ted_mem		= ted_base + $13	;
+	tedmem_CHARGEN	= %######..	; high address bits of character generator
+	tedmem_SCLOCK	= %......#.	; 1: force single clock mode
+	tedmem_ROM	= %.......#	; read-only: ROM is active over $8000
+
+ted_videoram	= ted_base + $14	; bits 7..3 are the high address bits of video ram
+
+; color registers
+; %7.......	unused
+; %.654....	luminance
+; %....3210	color
+ted_background	= ted_base + $15	; color for %00 pixels (or %0 in HiRes)
+ted_color1	= ted_base + $16	; color for %01 pixels (MCM and ECM)
+ted_color2	= ted_base + $17	; color for %10 pixels (MCM and ECM)
+ted_color3	= ted_base + $18	; color for %11 pixels (MCM graphics and ECM)
+ted_border	= ted_base + $19
+
+; current character position
+ted_charpos_hi	= ted_base + $1a	; bits 1..0 are highmost bits
+ted_charpos_lo	= ted_base + $1b	; low bits of actual character position
+
+; current raster position (writable!)
+ted_vertical_hi	= ted_base + $1c	; bit 0 is msb of raster line
+ted_vertical_lo	= ted_base + $1d	; low 8 bits of raster line
+ted_horizontal	= ted_base + $1e	; horizontal position (given in double pixels)
+
+ted_count	= ted_base + $1f
+	;		= %#.......	; always 1 (unused)
+	tedcount_FLASH	= %.####...	; flashing counter
+	tedcount_CHAR	= %.....###	; raster line in character row (r/w!)
+
+; pseudo registers
+; these are not really registers inside TED, but here they are, anyway:
+!addr	ted_enable_roms	= $ff3e	; any write here enables ROMs above $8000
+!addr	ted_enable_rams	= $ff3f	; any write here enables RAM above $8000

ACME_Lib/cbm/basic10.a

@@ -1,4 +1,4 @@
-;ACME 0.94.4
+;ACME 0.96.4
 
 !ifdef lib_cbm_basic10_a !eof
 lib_cbm_basic10_a = 1
@@ -18,12 +18,20 @@ token_DIR	= $ee	; was called DIRECTORY in basic v7
 !macro b_DIR	{!by token_DIR}		; aka DIRECTORY
 
 ; STASH/FETCH/SWAP are all decoded to DMA:
-!macro b_DMA	{!by $fe, $1f}
+!macro b_DMA	{!by $fe, $1f}		; command, length, source, target, subcommand
 ; extended token $fe $20 isn't used ($20 is ' ')
 ;!macro b_DMA	{!by $fe, $21}
 ; extended token $fe $22 isn't used ($20 is '"')
 ;!macro b_DMA	{!by $fe, $23}
 
+; renamed function:
+!macro b_LPEN		{!by $ce, $04:!pet '('}	; was called PEN in basic7
+; basic7 defines functions up to $ce, $0a (POINTER)
+; new functions:
+;!macro b_		{!by $ce, $0b:!pet '('}	; ?
+!macro b_PIXEL		{!by $ce, $0c:!pet '('}	; return pixel color
+!macro b_RPALETTE	{!by $ce, $0d:!pet '('}	; return palette color
+
 ; new instructions:
 !macro b_TYPE		{!by $fe, $27}	; display sequential disk file
 !macro b_BVERIFY	{!by $fe, $28}
@@ -51,3 +59,8 @@ token_DIR	= $ee	; was called DIRECTORY in basic v7
 !macro b_MOUSE		{!by $fe, $3e}	; set mouse parameters
 !macro b_RMOUSE		{!by $fe, $3f}	; read mouse position
 !macro b_DISK		{!by $fe, $40}	; send disc command
+!macro b_CURSOR		{!by $fe, $41}	; place text cursor?
+;!macro b_		{!by $fe, $42}	; ?
+!macro b_LOADIFF	{!by $fe, $43}	; load IFF graphics
+!macro b_SAVEIFF	{!by $fe, $44}	; save IFF graphics
+!macro b_EDIT		{!by $fe, $45}	; switch between program and text editing

ACME_Lib/cbm/c128/kernal.a

@@ -9,26 +9,28 @@ lib_cbm_c128_kernal_a = 1
 !address {
 	k_spin_spout	= $ff47
 	k_close_all	= $ff4a
-	k_c64mode	= $ff4d
+	k_c64mode	= $ff4d	; enter c64 mode (so does not return)
 	k_dma_call	= $ff50
 	k_boot_call	= $ff53
 	k_phoenix	= $ff56
-	k_lkupla	= $ff59
-	k_lkupsa	= $ff5c
+	k_lkupla	= $ff59	; find channel with file number A
+	k_lkupsa	= $ff5c	; find channel with secondary address Y
 	k_swapper	= $ff5f
 	k_dlchr		= $ff62
 	k_pfkey		= $ff65
-	k_setbnk	= $ff68
+	k_setbnk	= $ff68	; set banks for file name and load/save calls
 	k_getcfg	= $ff6b
 	k_jsrfar	= $ff6e
 	k_jmpfar	= $ff71
 	k_indfet	= $ff74
 	k_indsta	= $ff77
 	k_indcmp	= $ff7a
-	k_primm		= $ff7d
+	k_primm		= $ff7d	; "print immediate" - output string without having to setup a pointer:
+		; string must follow JSR $ff7d in memory, code execution will resume after terminating zero.
+		; A/X/Y are preserved
 	k_release_number	= $ff80
 }
-!source <cbm/kernal.a>	; $ff81-$fff5 is backward compatible to older machines
+!source <cbm/kernal.a>	; $ff81-$fff5 is mostly compatible to older machines
 ; $fff6/$fff7 are unused (ff ff)
 !address {
 	k_indirect128mode	= $fff8	; indirect vector, without JMP command!

ACME_Lib/cbm/c128/mmu.a

@@ -12,10 +12,10 @@ lib_cbm_c128_mmu_a = 2
 	; configuration register
 	mmu_cr_d500	= $d500	; same as "mmu_cr" at $ff00. Use "mmu_cr" instead, as that is *always* available.
 	; preconfiguration registers (internal format just like mmu_cr)
-	mmu_pcr_a	= $d501	; c128 kernel default is $3f (BANK 0)
-	mmu_pcr_b	= $d502	; c128 kernel default is $7f (BANK 1)
-	mmu_pcr_c	= $d503	; c128 kernel default is $01 (BANK 14)
-	mmu_pcr_d	= $d504	; c128 kernel default is $41 (all system roms, with ram 1)
+	mmu_pcr_a	= $d501	; c128 kernal default is $3f (BANK 0)
+	mmu_pcr_b	= $d502	; c128 kernal default is $7f (BANK 1)
+	mmu_pcr_c	= $d503	; c128 kernal default is $01 (BANK 14)
+	mmu_pcr_d	= $d504	; c128 kernal default is $41 (all system roms, with ram 1)
 }
 ; contents of cr and all four pcr:
 mmu_CR_RAMBANK_MASK	= %##......	; this controls which RAM bank is used in areas where RAM is enabled
@@ -25,7 +25,7 @@ mmu_CR_RAMBANK_2	= %#.......	; on an unmodified c128, there is no ram bank 2 (0
 mmu_CR_RAMBANK_3	= %##......	; on an unmodified c128, there is no ram bank 3 (1 will be used instead)
 
 mmu_CR_HIGH_MASK	= %..##....	; this controls the "high area" (c000..ffff), but i/o (d000..dfff) is separate
-;mmu_CR_HIGH_SYSROM	= %........	; editor, charset (or i/o, see below), kernel
+;mmu_CR_HIGH_SYSROM	= %........	; editor, charset (or i/o, see below), kernal
 mmu_CR_HIGH_INTFUNCROM	= %...#....
 mmu_CR_HIGH_EXTFUNCROM	= %..#.....
 mmu_CR_HIGH_RAM		= %..##....
@@ -57,10 +57,10 @@ mmu_CR_BANK8	= $2a	; 32 KiB bank 0; 32 KiB EFROM with i/o overlay
 mmu_CR_BANK9	= $6a	; 32 KiB bank 1; 32 KiB EFROM with i/o overlay
 mmu_CR_BANK10	= $aa	; 32 KiB bank 2; 32 KiB EFROM with i/o overlay
 mmu_CR_BANK11	= $ea	; 32 KiB bank 3; 32 KiB EFROM with i/o overlay
-mmu_CR_BANK12	= $06	; 32 KiB bank 0; 16 KiB IFROM; 16 KiB kernel with i/o overlay
-mmu_CR_BANK13	= $0a	; 32 KiB bank 0; 16 KiB EFROM; 16 KiB kernel with i/o overlay
-mmu_CR_BANK14	= $01	; 16 KiB bank 0; 32 KiB basic; 16 KiB kernel with font overlay
-mmu_CR_BANK15	= $00	; 16 KiB bank 0; 32 KiB basic; 16 KiB kernel with i/o overlay
+mmu_CR_BANK12	= $06	; 32 KiB bank 0; 16 KiB IFROM; 16 KiB kernal with i/o overlay
+mmu_CR_BANK13	= $0a	; 32 KiB bank 0; 16 KiB EFROM; 16 KiB kernal with i/o overlay
+mmu_CR_BANK14	= $01	; 16 KiB bank 0; 32 KiB basic; 16 KiB kernal with font overlay
+mmu_CR_BANK15	= $00	; 16 KiB bank 0; 32 KiB basic; 16 KiB kernal with i/o overlay
 ; An unmodified C128 does not have a "ram bank 2" or "ram bank 3".
 ; Whenever one of these is activated, ram banks 0 and 1 will be used instead.
 ; IFROM means internal function ROM (socket U36)
@@ -71,13 +71,13 @@ mmu_CR_BANK15	= $00	; 16 KiB bank 0; 32 KiB basic; 16 KiB kernel with i/o overla
 	mmu_mcr		= $d505
 }
 ; contents:
-mmu_MCR_40COLUMNS	= %#.......	; 40/80 key: 0 means pressed, 1 means released (writable! if cleared, will always read as 0!)
-mmu_MCR_C64MODE		= %.#......	; setting this bit makes the MMU disappear from the memory map :)
-mmu_MCR_EXROM		= %..#.....	; if zero on boot, system will enter c64 mode (writable!)
-mmu_MCR_GAME		= %...#....	; if zero on boot, system will enter c64 mode (writable!)
-mmu_MCR_FSDIR_OUTPUT	= %....#...	; direction of fast serial bus
+mmu_MCR_40COLUMNS	= %#.......	; (pin 48) 40/80 key: 0 means pressed, 1 means released (writable! if cleared, will always read as 0!)
+mmu_MCR_C64MODE		= %.#......	; (pin 47) setting this bit makes the MMU disappear from the memory map :)
+mmu_MCR_EXROM		= %..#.....	; (pin 46) if zero on boot, system will enter c64 mode (writable!)
+mmu_MCR_GAME		= %...#....	; (pin 45) if zero on boot, system will enter c64 mode (writable!)
+mmu_MCR_FSDIR_OUTPUT	= %....#...	; (pin 44) direction of fast serial bus
 mmu_MCR_UNUSED		= %.....##.	; always set
-mmu_MCR_8502MODE	= %.......#	; setting this to zero switches to Z80 cpu
+mmu_MCR_8502MODE	= %.......#	; (pin 43 inverted) setting this to zero switches to Z80 cpu
 
 !address {
 	; ram configuration register
@@ -148,10 +148,10 @@ mmu_VR_VERSION_MASK	= %....####	; mmu version 0
 	; load configuration registers:
 	; a read access will return the value of the corresponding preconfiguration register
 	; any write access will copy the value of the corresponding preconfiguration register to mmu_cr
-	mmu_lcr_a	= $ff01	; c128 kernel default is $3f (BANK 0)
-	mmu_lcr_b	= $ff02	; c128 kernel default is $7f (BANK 1)
-	mmu_lcr_c	= $ff03	; c128 kernel default is $01 (BANK 14)
-	mmu_lcr_d	= $ff04	; c128 kernel default is $41 (all system roms, with ram 1)
+	mmu_lcr_a	= $ff01	; c128 kernal default is $3f (BANK 0)
+	mmu_lcr_b	= $ff02	; c128 kernal default is $7f (BANK 1)
+	mmu_lcr_c	= $ff03	; c128 kernal default is $01 (BANK 14)
+	mmu_lcr_d	= $ff04	; c128 kernal default is $41 (all system roms, with ram 1)
 
 	; the c128 ROMs contain a look-up table to convert bank numbers to their
 	; corresponding configuration register values:

ACME_Lib/cbm/c128/petscii.a

@@ -1,4 +1,4 @@
-;ACME 0.94.4
+;ACME 0.96.4
 
 !ifdef lib_cbm_c128_petscii_a !eof
 lib_cbm_c128_petscii_a = 1
@@ -25,8 +25,8 @@ petscii_LBLUE		= 154
 petscii_GRAY3		= 155:	petscii_DWHITE		= 155:	petscii_LGRAY		= 155
 
 ; switching character set
-petscii_LOCK		=  11	; forbid CBM-shift (C64 uses 8)
-petscii_UNLOCK		=  12	; allow CBM-shift (C64 uses 9)
+petscii_LOCK		=  11	; forbid CBM-shift (C64 uses 8 instead)
+petscii_UNLOCK		=  12	; allow CBM-shift (C64 uses 9 instead)
 petscii_LOWERCASE	=  14	; switch to lowercase/uppercase character set
 petscii_UPPERCASE	= 142	; switch to uppercase/graphics character set
 
@@ -39,12 +39,12 @@ petscii_F7		= 136:	petscii_F8		= 140
 ; C128-specific stuff
 petscii_UNDERLINEON	=   2
 petscii_UNDERLINEOFF	= 130
-petscii_FLASHON		=  15
-petscii_FLASHOFF	= 143
+petscii_FLASHON		=  15	; (264 series machines use 130 instead)
+petscii_FLASHOFF	= 143	; (264 series machines use 131 instead)
 petscii_BELL		=   7
 petscii_TAB		=   9
 petscii_SETTAB		=  24
 petscii_LINEFEED	=  10
 petscii_ESCAPE		=  27
 petscii_SHIFTSTOP	= 131
-petscii_HELP		= 132
+petscii_HELP		= 132	; (264 series machines use 140 instead)

ACME_Lib/cbm/c128/vic.a

@@ -1,19 +1,22 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_cbm_c128_vic_a !eof
 lib_cbm_c128_vic_a = 1
 
 !source <cbm/c64/vic.a>	; registers 0..2e
 
-!address {
 	; registers only present in the C128 variant of this chip:
-	vic_keyboard	= $d02f
-	vic_clock	= $d030
+	vic_keyboard	= vic_base + $2f
+	; %76543...	always set
+	; %.....210	output pins for extended keyboard layout
+	vic_clock	= vic_base + $30
+	; %765432.. always set
+	; %......1. "test" bit: 0 = normal, 1 = system clock changes raster line
+	; %.......0 system clock: 0 = 1 MHz, 1 = 2 MHz (VIC bus cycles will be given to CPU instead)
 
 	; the c128 ROMs contain two copies of a look-up table to convert vic color
 	; values to their corresponding petscii color codes:
 	; rom4_* needs "low rom area" enabled ($4000..$7fff)
 	; romc_* needs "high rom area" enabled ($c000..$ffff)
-	rom4_vic_to_petscii_color_table	= $76b5 ; 90 05 1c 9f 9c 1e 1f 9e 81 95 96 97 98 99 9a 9b
-	romc_vic_to_petscii_color_table	= $ce4c ; 90 05 1c 9f 9c 1e 1f 9e 81 95 96 97 98 99 9a 9b
-}
+!addr	rom4_vic_to_petscii_color_table	= $76b5 ; 90 05 1c 9f 9c 1e 1f 9e 81 95 96 97 98 99 9a 9b
+!addr	romc_vic_to_petscii_color_table	= $ce4c ; 90 05 1c 9f 9c 1e 1f 9e 81 95 96 97 98 99 9a 9b

ACME_Lib/cbm/c64/cia1.a

@@ -1,23 +1,27 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_cbm_c64_cia1_a !eof
 lib_cbm_c64_cia1_a = 1
 
-!address {
-	cia1_pra	= $dc00
-	cia1_prb	= $dc01
-	cia1_ddra	= $dc02
-	cia1_ddrb	= $dc03
-	cia1_ta_lo	= $dc04
-	cia1_ta_hi	= $dc05
-	cia1_tb_lo	= $dc06
-	cia1_tb_hi	= $dc07
-	cia1_tod10ths	= $dc08
-	cia1_todsec	= $dc09
-	cia1_todmin	= $dc0a
-	cia1_todhr	= $dc0b
-	cia1_sdr	= $dc0c
-	cia1_icr	= $dc0d
-	cia1_cra	= $dc0e
-	cia1_crb	= $dc0f
-}
+!source <cbm/cia.a>	; chip stuff (same for both cias)
+; stuff for cia 1 only:
+!addr	cia1_base	= $dc00
+	cia1_pra	= cia1_base + cia_port_a	; PA0..PA4 are joy port 2	PA6+PA7 select paddle port(s)
+	cia1_prb	= cia1_base + cia_port_b	; PB0..PB4 are joy port 1
+	; both ports are used for keyboard matrix
+	cia1_ddra	= cia1_base + cia_data_direction_a
+	cia1_ddrb	= cia1_base + cia_data_direction_b
+	cia1_ta_lo	= cia1_base + cia_timer_a_low
+	cia1_ta_hi	= cia1_base + cia_timer_a_high
+	cia1_tb_lo	= cia1_base + cia_timer_b_low
+	cia1_tb_hi	= cia1_base + cia_timer_b_high
+	cia1_tod10ths	= cia1_base + cia_timeofday_10ths
+	cia1_todsec	= cia1_base + cia_timeofday_seconds
+	cia1_todmin	= cia1_base + cia_timeofday_minutes
+	cia1_todhr	= cia1_base + cia_timeofday_hours
+	cia1_sdr	= cia1_base + cia_serial_data
+	cia1_icr	= cia1_base + cia_interrupt_control
+	cia1_cra	= cia1_base + cia_control_a
+	cia1_crb	= cia1_base + cia_control_b
+; the interrupt output is connected to CPU's /IRQ input
+; in the C128, the shift register is used for the fast serial port (burst mode)

ACME_Lib/cbm/c64/cia2.a

@@ -1,23 +1,32 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_cbm_c64_cia2_a !eof
 lib_cbm_c64_cia2_a = 1
 
-!address {
-	cia2_pra	= $dd00
-	cia2_prb	= $dd01
-	cia2_ddra	= $dd02
-	cia2_ddrb	= $dd03
-	cia2_ta_lo	= $dd04
-	cia2_ta_hi	= $dd05
-	cia2_tb_lo	= $dd06
-	cia2_tb_hi	= $dd07
-	cia2_tod10ths	= $dd08
-	cia2_todsec	= $dd09
-	cia2_todmin	= $dd0a
-	cia2_todhr	= $dd0b
-	cia2_sdr	= $dd0c
-	cia2_icr	= $dd0d
-	cia2_cra	= $dd0e
-	cia2_crb	= $dd0f
-}
+!source <cbm/cia.a>	; chip stuff (same for both cias)
+; stuff for cia 2 only:
+!addr	cia2_base	= $dd00
+	cia2_pra	= cia2_base + cia_port_a
+	; %7.......	DATA in (0 means GND)
+	; %.6......	CLK in (0 means GND)
+	; %..5.....	DATA out (gets inverted, so 1 means GND)
+	; %...4....	CLK out	(gets inverted, so 1 means GND)
+	; %....3...	ATN out (gets inverted, so 1 means GND)
+	; %.....2..	PA2 (pin M at user port, 0 means GND)
+	; %......10	VIC bank (gets inverted, so value %11 means address $0000)
+	cia2_prb	= cia2_base + cia_port_b
+	cia2_ddra	= cia2_base + cia_data_direction_a
+	cia2_ddrb	= cia2_base + cia_data_direction_b
+	cia2_ta_lo	= cia2_base + cia_timer_a_low
+	cia2_ta_hi	= cia2_base + cia_timer_a_high
+	cia2_tb_lo	= cia2_base + cia_timer_b_low
+	cia2_tb_hi	= cia2_base + cia_timer_b_high
+	cia2_tod10ths	= cia2_base + cia_timeofday_10ths
+	cia2_todsec	= cia2_base + cia_timeofday_seconds
+	cia2_todmin	= cia2_base + cia_timeofday_minutes
+	cia2_todhr	= cia2_base + cia_timeofday_hours
+	cia2_sdr	= cia2_base + cia_serial_data
+	cia2_icr	= cia2_base + cia_interrupt_control
+	cia2_cra	= cia2_base + cia_control_a
+	cia2_crb	= cia2_base + cia_control_b
+; the interrupt output is connected to CPU's /NMI input

ACME_Lib/cbm/c64/float.a

@@ -1,7 +1,7 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_cbm_c64_float_a !eof
-lib_cbm_c64_float_a = 1
+lib_cbm_c64_float_a = 2
 
 ; Here are some definitions to help you call the floating-point functions of the
 ; C64's BASIC ROM. They work on "float registers", which are actually just
@@ -18,33 +18,33 @@ lib_cbm_c64_float_a = 1
 
 ; convenience macros:
 
-; some float functions need a memory address in A (low) and Y (high)
-!macro movAY .adr {
-	lda #<.adr
-	ldy #>.adr
+; some float functions need a memory address in YYAA
+!macro movYYAA @addr {
+	lda #<@addr
+	ldy #>@addr
 }
-; ...or in X (low) and Y (high)
-!macro movXY .adr {
-	ldx #<.adr
-	ldy #>.adr
+; ...or in YYXX
+!macro movYYXX @addr {
+	ldx #<@addr
+	ldy #>@addr
 }
-; other float functions expect or output a value in Y (low) and A (high)
-!macro movYA .val {
-	ldy #<.val
-	lda #>.val
+; other float functions expect or output a value in AAYY
+!macro movAAYY @val {
+	ldy #<@val
+	lda #>@val
 }
-!macro ldYA .adr {
-	ldy .adr
-	lda .adr + 1
+!macro ldAAYY @addr {
+	ldy @addr
+	lda @addr + 1
 }
-!macro stYA .adr {
-	sty .adr
-	sta .adr + 1
+!macro stAAYY @addr {
+	sty @addr
+	sta @addr + 1
 }
-; ...or in X (low) and A (high)
-!macro ldXA .adr {
-	ldx .adr
-	lda .adr + 1
+; ...or in AAXX
+!macro ldAAXX @addr {
+	ldx @addr
+	lda @addr + 1
 }
 
 !address {
@@ -69,50 +69,71 @@ lib_cbm_c64_float_a = 1
 
 	; functions - a few points to note:
 	; fac1/2 might get clobbered even if not mentioned in the function's name,
-	;	because stuff like fac1_times_memAY will load the value from memory
+	;	because stuff like fac1_times_memYYAA will load the value from memory
 	;	into fac2 first.
 	; for subtraction and division, the left operand is in fac2, the right operand in fac1.
 	fac1_print			= $aabc	; print string representation of contents of fac1
-	fac1_to_signedYA		= $b1aa	; might throw ILLEGAL QUANTITY
+	fac1_to_signedAAYY		= $b1aa	; might throw ILLEGAL QUANTITY
 	fac1_to_signed16		= $b1bf	; might throw ILLEGAL QUANTITY
-	fac1_read_signedYA		= $b391	; convert 16 bit signed int to float
+	fac1_read_signedAAYY		= $b391	; convert 16 bit signed int to float
 	fac1_read_unsignedY		= $b3a2	; convert 8 bit unsigned int to float
 	fac1_read_string		= $b7b5	; $22/23 must point to string, A must be string length
-	fac1_to_unsignedYA		= $b7f7	; might throw ILLEGAL QUANTITY (result is also in $14/15)
+	fac1_to_unsignedAAYY		= $b7f7	; might throw ILLEGAL QUANTITY (result is also in $14/15)
 	fac1_add_point5			= $b849	; for rounding, call this before fac1_int
-	fac1_memAY_minus_fac1		= $b850	; subtract fac1 from mflpt value
+	fac1_memYYAA_minus_fac1		= $b850	; subtract fac1 from mflpt value
 	fac1_fac2_minus_fac1		= $b853
-	fac1_add_memAY			= $b867	; add mflpt value
+	fac1_add_memYYAA		= $b867	; add mflpt value
 	fac1_add_fac2			= $b86a
 	fac1_log			= $b9ea	; LOG()
-	fac1_times_memAY		= $ba28	; multiply by mflpt value
-	fac2_read_memAY			= $ba8c	; load mflpt value from memory into fac2
+	fac1_times_memYYAA		= $ba28	; multiply by mflpt value
+	fac2_read_memYYAA		= $ba8c	; load mflpt value from memory into fac2
 	fac2_read_mem_via0x22ptr	= $ba90	; load mflpt value from memory into fac2
 	fac1_times_10			= $bae2
 	fac1_divide_by_10		= $bafe	; CAUTION: result is always positive!
-	fac1_divide_memAY_by_fac1	= $bb0f	; divide mflpt value by fac1 value
-	fac1_read_memAY			= $bba2	; load mflpt value from memory into fac1
+	fac1_divide_memYYAA_by_fac1	= $bb0f	; divide mflpt value by fac1 value
+	fac1_read_memYYAA		= $bba2	; load mflpt value from memory into fac1
 	fac1_read_mem_via0x22ptr	= $bba6	; load mflpt value from memory into fac1
-	fac1_to_memXY			= $bbd4	; store fac1 to memory as mflpt
+	fac1_to_memYYXX			= $bbd4	; store fac1 to memory as mflpt
 	fac1_read_fac2			= $bbfc	; copy fac2 to fac1
 	fac2_read_fac1			= $bc0c	; copy fac1 to fac2
 	fac1_sign_to_A			= $bc2b	; $ff, $0, $1 for negative, zero, positive
 	fac1_sgn			= $bc39 ; SGN()
 	fac1_abs			= $bc58 ; ABS()
-	fac1_compare_to_memAY		= $bc5b	; compare to mflpt value in memory
+	fac1_compare_to_memYYAA		= $bc5b	; compare to mflpt value in memory
 	fac1_to_signed32		= $bc9b
 	fac1_int			= $bccc ; INT()
 	fac1_read_string0		= $bcf3	; use b7b5 instead; this only works after calling CHRGET
-	fac1_print_unsignedXA		= $bdcd
-	fac1_to_string			= $bddd	; string is stored at $0100 (address returned in AY)
+	fac1_print_unsignedAAXX		= $bdcd
+	fac1_to_string			= $bddd	; string is stored at $0100 (address returned in YYAA)
 	fac1_sqr			= $bf71	; SQR()
-	fac1_fac2_to_the_power_of_memAY	= $bf78
+	fac1_fac2_to_the_power_of_memYYAA	= $bf78
 	fac1_negate			= $bfb4
 	fac1_exp			= $bfed	; EXP()
-	; end of basic rom jumps to start of kernel rom!
+	; end of basic rom jumps to start of kernal rom!
 	fac1_rnd			= $e097	; RND()
 	fac1_cos			= $e264 ; COS()
 	fac1_sin			= $e26b	; SIN()
 	fac1_tan			= $e2b4	; TAN()
 	fac1_atn			= $e30e	; ATN()
 }
+
+; wrappers for names from older version of this file:
+!macro movAY @addr {	+movYYAA @addr	}
+!macro movXY @addr {	+movYYXX @addr	}
+!macro movYA @val {	+movAAYY @val	}
+!macro ldYA @addr {	+ldAAYY @addr	}
+!macro stYA @addr {	+stAAYY @addr	}
+!macro ldXA @addr {	+ldAAXX @addr	}
+	fac1_to_signedYA		= fac1_to_signedAAYY
+	fac1_read_signedYA		= fac1_read_signedAAYY
+	fac1_to_unsignedYA		= fac1_to_unsignedAAYY
+	fac1_memAY_minus_fac1		= fac1_memYYAA_minus_fac1
+	fac1_add_memAY			= fac1_add_memYYAA
+	fac1_times_memAY		= fac1_times_memYYAA
+	fac2_read_memAY			= fac2_read_memYYAA
+	fac1_divide_memAY_by_fac1	= fac1_divide_memYYAA_by_fac1
+	fac1_read_memAY			= fac1_read_memYYAA
+	fac1_to_memXY			= fac1_to_memYYXX
+	fac1_compare_to_memAY		= fac1_compare_to_memYYAA
+	fac1_print_unsignedXA		= fac1_print_unsignedAAXX
+	fac1_fac2_to_the_power_of_memAY	= fac1_fac2_to_the_power_of_memYYAA

ACME_Lib/cbm/c64/memcopy.a

@@ -1,4 +1,4 @@
-;ACME 0.95.7
+;ACME 0.96.4
 
 !ifdef lib_cbm_c64_memcopy_a !eof
 lib_cbm_c64_memcopy_a = 1
@@ -11,24 +11,23 @@ lib_cbm_c64_memcopy_a = 1
 ; higher addresses are copied first, so:
 ;	- moving data to higher addresses works even if areas overlap
 ;	- moving data to lower addresses only works if areas do not overlap
-!macro basic_memcopy .src_start, .src_end, .target_start {
-	!address {
-		.z_target_end	= $58
-		.z_src_end	= $5a
-		.z_src_start	= $5f
-		.fn		= $a3bf
-	}
-		lda #<.src_start
-		ldx #>.src_start
-		sta .z_src_start
-		stx .z_src_start + 1
-		lda #<.src_end
-		ldx #>.src_end
-		sta .z_src_end
-		stx .z_src_end + 1
-		lda #<(.target_start + .src_end - .src_start)
-		ldx #>(.target_start + .src_end - .src_start)
-		sta .z_target_end
-		stx .z_target_end + 1
-		jsr .fn
+!macro basic_memcopy @src_start, @src_end, @target_start {
+	!addr	@z_target_end	= $58
+	!addr	@z_src_end	= $5a
+	!addr	@z_src_start	= $5f
+	!addr	@fn		= $a3bf
+
+		lda #<@src_start
+		ldx #>@src_start
+		sta @z_src_start
+		stx @z_src_start + 1
+		lda #<@src_end
+		ldx #>@src_end
+		sta @z_src_end
+		stx @z_src_end + 1
+		lda #<(@target_start + @src_end - @src_start)
+		ldx #>(@target_start + @src_end - @src_start)
+		sta @z_target_end
+		stx @z_target_end + 1
+		jsr @fn
 }

ACME_Lib/cbm/c64/reu.a

@@ -47,8 +47,8 @@ rec_COMMAND_FETCH	= %#.#....#	;	starting and then reload values.
 	; Upgraded units and clones may have more, but the REC chip will always
 	; "wrap around" after eight banks if crossing bank borders!
 	; amount of bytes to process
-	rec_amount_low	= $df07
-	rec_amount_high	= $df08
+	rec_amount_low	= $df07	; using $0000 results in
+	rec_amount_high	= $df08	;	64 KiB being transferred
 	; when to request interrupts
 	rec_irqctrl	= $df09
 }

ACME_Lib/cbm/c64/vic.a

@@ -1,8 +1,25 @@
-;ACME 0.95
+;ACME 0.96.4
 
 !ifdef lib_cbm_c64_vic_a !eof
 lib_cbm_c64_vic_a = 1
 
+; This is from Christian Bauer's VIC text: {
+;	6566: designed for static RAM, never used in C64/128
+;	6567/8562: C64, NTSC
+;	6569/8565: C64, PAL
+;	6572: Drean C64, PAL-N
+;	8564: C128, NTSC
+;	8566: C128, PAL
+;	PAL chips generate 312 lines with 63 cycles per line.
+;	PAL-N chips generate 312 lines with 65 cycles per line.
+;	NTSC chips generate 263 lines with 65 cycles per line.
+;	Very early NTSC chips (6567R56A) have an off-by-one
+;	error, they generate 262 lines with 64 cycles per line.
+; }
+
+; The c128 version is officially called "VIC-IIe".
+; It has two additional registers and more pins.
+
 ; color codes
 viccolor_BLACK	= $0
 viccolor_WHITE	= $1
@@ -21,58 +38,85 @@ viccolor_LGREEN	= $d
 viccolor_LBLUE	= $e
 viccolor_GRAY3	= $f
 
-!address {
-	; register addresses
-	vic_xs0		= $d000
-	vic_ys0		= $d001
-	vic_xs1		= $d002
-	vic_ys1		= $d003
-	vic_xs2		= $d004
-	vic_ys2		= $d005
-	vic_xs3		= $d006
-	vic_ys3		= $d007
-	vic_xs4		= $d008
-	vic_ys4		= $d009
-	vic_xs5		= $d00a
-	vic_ys5		= $d00b
-	vic_xs6		= $d00c
-	vic_ys6		= $d00d
-	vic_xs7		= $d00e
-	vic_ys7		= $d00f
-	vic_msb_xs	= $d010
-	vic_controlv	= $d011	; vertical control (and much other stuff)
-	vic_line	= $d012	; raster line
-	vic_xlp		= $d013	; light pen coordinates
-	vic_ylp		= $d014
-	vic_sactive	= $d015	; sprites: active
-	vic_controlh	= $d016	; horizontal control (and much other stuff)
-	vic_sdy		= $d017	; sprites: double height
-	vic_ram		= $d018	; RAM pointer
-	vic_irq		= $d019
-	vic_irqmask	= $d01a
-	vic_sback	= $d01b	; sprites: background mode
-	vic_smc		= $d01c	; sprites: multi color mode
-	vic_sdx		= $d01d	; sprites: double width
-	vic_ss_collided	= $d01e	; sprite-sprite collision detect
-	vic_sd_collided	= $d01f	; sprite-data collision detect
-	; color registers
-	vic_cborder	= $d020	; border color
-	vic_cbg		= $d021	; general background color
-	vic_cbg0	= $d021
-	vic_cbg1	= $d022	; background color 1 (for EBC and MC text mode)
-	vic_cbg2	= $d023	; background color 2 (for EBC and MC text mode)
-	vic_cbg3	= $d024	; background color 3 (for EBC mode)
-	vic_sc01	= $d025	; sprite color for MC-bitpattern %01
-	vic_sc11	= $d026	; sprite color for MC-bitpattern %11
-	vic_cs0		= $d027	; sprite colors
-	vic_cs1		= $d028
-	vic_cs2		= $d029
-	vic_cs3		= $d02a
-	vic_cs4		= $d02b
-	vic_cs5		= $d02c
-	vic_cs6		= $d02d
-	vic_cs7		= $d02e
-}
+!addr	vic_base	= $d000
+	; sprite coordinates:
+	vic_xs0		= vic_base + $00
+	vic_ys0		= vic_base + $01
+	vic_xs1		= vic_base + $02
+	vic_ys1		= vic_base + $03
+	vic_xs2		= vic_base + $04
+	vic_ys2		= vic_base + $05
+	vic_xs3		= vic_base + $06
+	vic_ys3		= vic_base + $07
+	vic_xs4		= vic_base + $08
+	vic_ys4		= vic_base + $09
+	vic_xs5		= vic_base + $0a
+	vic_ys5		= vic_base + $0b
+	vic_xs6		= vic_base + $0c
+	vic_ys6		= vic_base + $0d
+	vic_xs7		= vic_base + $0e
+	vic_ys7		= vic_base + $0f
+	vic_msb_xs	= vic_base + $10	; bit 8 of x position	(one bit per sprite)
+	vic_controlv	= vic_base + $11	; vertical control and other stuff:
+	; %7....... "bit 8" of $d012
+	; %.6...... extended background color mode (1=on)
+	; %..5..... 0 = text mode, 1 = bitmap mode
+	; %...4.... display enable: 0 = disable (border only), 1 = enable	VIC only checks this once per frame!
+	; %....3... lines: 0 = 24, 1 = 25	; if set, upper and lower border gain 4 pixels each
+	; %.....210 vertical smooth scroll (default %011), screen contents are moved down by this amount
+	vic_line	= vic_base + $12	; raster line (also see bit 7 of $d011)	reading returns current value, writing sets interrupt value
+	vic_xlp		= vic_base + $13	; light pen coordinates, x (only half the resolution)
+	vic_ylp		= vic_base + $14	; light pen coordinates, y
+	vic_sactive	= vic_base + $15	; sprite enable	(one bit per sprite)
+	vic_controlh	= vic_base + $16	; horizontal control and other stuff:
+	; %76...... always set
+	; %..5..... "test", should be 0
+	; %...4.... 0 = hires, 1 = multicolor
+	; %....3... columns: 0 = 38, 1 = 40	; if set, left border gains 7 pixels and right border gains 9 pixels
+	; %.....210 horizontal smooth scroll (default %000), screen contents are moved to the right by this amount
+	vic_sdy		= vic_base + $17	; sprites, double height	(one bit per sprite)
+	vic_ram		= vic_base + $18	; RAM pointer
+	; %7654.... which K of VIC bank is video ram (default %0001)
+	; %....321. text: which 2K of VIC bank is character set (default %010)
+	; %.......0 text: unused
+	; %....3... bitmap: which 8K of VIC bank is bitmap
+	; %.....210 bitmap: unused
+	vic_irq		= vic_base + $19	; writing back acknowledges!
+	; %7....... 1: VIC requested interrupt
+	; %.654.... always set
+	; %....3... 1: light pen active
+	; %.....2.. 1: sprite-sprite collision
+	; %......1. 1: sprite-data collision
+	; %.......0 1: raster interrupt
+	vic_irqmask	= vic_base + $1a
+	; %7654.... always set
+	; %....3... 1: enable lightpen interrupt
+	; %.....2.. 1: enable sprite-sprite collision interrupt
+	; %......1. 1: enable sprite-data collision interrupt
+	; %.......0 1: enable raster interrupt
+	vic_sback	= vic_base + $1b	; sprites, background	(one bit per sprite)
+	vic_smc		= vic_base + $1c	; sprites, multicolor	(one bit per sprite)
+	vic_sdx		= vic_base + $1d	; sprites, double width	(one bit per sprite)
+	vic_ss_collided	= vic_base + $1e	; sprites, sprite collision	(one bit per sprite)	reading clears register!
+	vic_sd_collided	= vic_base + $1f	; sprites, data collision	(one bit per sprite)	reading clears register!
+	; color registers (high nibbles are always %1111):
+	vic_cborder	= vic_base + $20	; border color
+	vic_cbg		= vic_base + $21	; general background color
+	vic_cbg0	= vic_base + $21
+	vic_cbg1	= vic_base + $22	; background color 1 (for EBC and MC text mode)
+	vic_cbg2	= vic_base + $23	; background color 2 (for EBC and MC text mode)
+	vic_cbg3	= vic_base + $24	; background color 3 (for EBC mode)
+	vic_sc01	= vic_base + $25	; sprite color for MC-bitpattern %01
+	vic_sc11	= vic_base + $26	; sprite color for MC-bitpattern %11
+	; individual sprite colors (in MC mode, these are used for bit pattern %10):
+	vic_cs0		= vic_base + $27	; sprite 0 color
+	vic_cs1		= vic_base + $28	; sprite 1 color
+	vic_cs2		= vic_base + $29	; sprite 2 color
+	vic_cs3		= vic_base + $2a	; sprite 3 color
+	vic_cs4		= vic_base + $2b	; sprite 4 color
+	vic_cs5		= vic_base + $2c	; sprite 5 color
+	vic_cs6		= vic_base + $2d	; sprite 6 color
+	vic_cs7		= vic_base + $2e	; sprite 7 color
 ; See <cbm/c128/vic.a> for the C128's two additional registers at $d02f/$d030.
 ; They are accessible even in C64 mode and $d030 can garble the video output,
 ; so be careful not to write to it accidentally in a C64 program!

ACME_Lib/cbm/cia.a

@@ -0,0 +1,109 @@
+;ACME 0.96.4
+
+!ifdef lib_cbm_cia_a !eof
+lib_cbm_cia_a = 1
+
+; CBM's "complex interface adapter" chip, known as 6526.
+; A newer version of this chip was initially called 8521, but later the name was
+; changed back to 6526 again.
+
+; There are two of these in a C64/128 computer, and one in 1570/1571 drives.
+
+; pinout:
+;               ____ ____
+;              |    V    |
+;          GND |  1   40 | cnt  clock for shift register
+;        / pa0 |  2   39 | sp   data for shift register
+;       |  pa1 |  3   38 | a0 \
+;       |  pa2 |  4   37 | a1  |_  address
+; Port _|  pa3 |  5   36 | a2  |   bus
+;   A   |  pa4 |  6   35 | a3 /
+;       |  pa5 |  7   34 | /reset
+;       |  pa6 |  8   33 | d0 \
+;        \ pa7 |  9   32 | d1  |
+;        / pb0 | 10   31 | d2  |
+;       |  pb1 | 11   30 | d3  |_  data
+;       |  pb2 | 12   29 | d4  |   bus
+; Port _|  pb3 | 13   28 | d5  |
+;   B   |  pb4 | 14   27 | d6  |
+;       |  pb5 | 15   26 | d7 /
+;       |  pb6 | 16   25 | 0in  clock
+;        \ pb7 | 17   24 | /flag
+;          /pc | 18   23 | /cs  chip select
+;      tod_clk | 19   22 | r/w
+;          +5V | 20   21 | /irq
+;              |_________|
+
+
+; register offsets:
+	; two 8-bit ports:
+	cia_port_a		= $0
+	cia_port_b		= $1
+	cia_data_direction_a	= $2	; clear means input,
+	cia_data_direction_b	= $3	; set means output
+
+	; two 16-bit timers, can be combined to form a single 32-bit timer:
+	cia_timer_a_low		= $4
+	cia_timer_a_high	= $5
+	cia_timer_b_low		= $6
+	cia_timer_b_high	= $7
+	; reading returns current counter value,
+	; writing sets start value.
+	; in 32-bit mode, timer A is "low word" and timer B is "high word".
+
+	; TOD (time of day) clock, clocked with 50 or 60 Hz:
+	; (CAUTION, registers use binary coded decimal format)
+	cia_timeofday_10ths	= $8	; %....3210	0..9, counts 10ths of seconds
+	cia_timeofday_seconds	= $9	; %.6543210	0..59, as two BCD digits:
+		; %.654....	0..5	"tens"
+		; %....3210	0..9	"ones"
+	cia_timeofday_minutes	= $a	; %.6543210	0..59, as two BCD digits:
+		; %.654....	0..5	"tens"
+		; %....3210	0..9	"ones"
+	cia_timeofday_hours	= $b	; %7..43210	AM/PM and 1..12 as two BCD digits:
+		; %7.......	0 means AM, 1 means PM
+		; %...4....	0..1	"tens"
+		; %....3210	0..9	"ones"
+	; when reading or writing time, start with hours and end with 10ths:
+	; accessing hours uncouples registers from clock, accessing 10ths re-couples them.
+	; if your read access does not start with the hours register,
+	; you have a race condition!
+
+	; shift register:
+	; msb is sent/received first. send clock is half of timer A's underflow rate.
+	cia_serial_data		= $c
+
+	; control registers:
+	cia_interrupt_control	= $d
+	; %7.......	read: 1 = interrupt requested (reading clears register)
+	;		write:	0 = disable interrupts indicated by set bits
+	;			1 = enable interrupts indicated by set bits
+	; %.65.....	unused
+	; %...4....	negative edge on /flag detected
+	; %....3...	shift register has finished sending or receiving
+	; %.....2..	time of day alarm
+	; %......1.	timer B underflow
+	; %.......0	timer A underflow
+
+	cia_control_a		= $e
+	; %7.......	TOD clock: 0 means 60 Hz, 1 means 50 Hz
+	; %.6......	shift register direction: 0 = input, 1 = output
+	; %..5.....	timer A clock: 0 = system clock, 1 = CNT
+	; %...4....	1 = force load timer start value
+	; %....3...	timer mode: 0 = continuous, 1 = one-shot (needs restart via bit 0)
+	; %.....2..	0 = pulse on PB6, 1 = invert PB6
+	; %......1.	1 = timer underflow shows on PB6 (this forces PB6 to output)
+	; %.......0	0 stops timer, 1 starts timer
+
+	cia_control_b		= $f
+	; %7.......	TOD write mode: 0 = actual time, 1 = alarm time (it is not possible to _read_ the alarm time!)
+	; %.65.....	timer B clock:
+	;		%00 = system clock
+	;		%01 = CNT
+	;		%10 = underflow of timer A
+	;		%11 = combination of %01 and %10
+	; %...4....	1 = force load timer start value
+	; %....3...	timer mode: 0 = continuous, 1 = one-shot (needs restart via bit 0)
+	; %.....2..	0 = pulse on PB7, 1 = invert PB7
+	; %......1.	1 = timer underflow shows on PB7 (this forces PB7 to output)
+	; %.......0	0 stops timer, 1 starts timer

ACME_Lib/cbm/flpt.a

@@ -1,4 +1,4 @@
-;ACME 0.94.5
+;ACME 0.96.4
 
 !ifdef lib_cbm_flpt_a !eof
 lib_cbm_flpt_a = 1
@@ -46,48 +46,49 @@ lib_cbm_flpt_a = 1
 
 ; this is ugly, but it gets the job done
 ; (if it's stupid, but it works, then it's not stupid)
-!macro flpt .value {
-	!set .float = float(.value)	; make sure to do passes until value is defined
-	!ifndef .float {
-		!by $ff, $ff, $ff, $ff, $ff, $ff	; six place holder bytes
+!macro flpt @value {
+	!set @float = float(@value)
+	!ifndef @float {
+		!by <@float, $ff, $ff, $ff, $ff, $ff	; six place holder bytes
+		; (first one depends on @float just to make sure more passes are done until value is defined)
 	} else {
 		; value is defined, so split up into sign and non-negative value
-		!if .float < 0 {
-			!set .sign = $80
-			!set .float = -.float
+		!if @float < 0 {
+			!set @sign = $80
+			!set @float = -@float
 		} else {
-			!set .sign = $00
+			!set @sign = $00
 		}
-		!if .float = 0 {
+		!if @float = 0 {
 			!by 0, 0, 0, 0, 0, 0	; six zeroes (zero is represented by all bits zero)
 		} else {
 			; split up into exponent and mantissa
-			!set .exponent = 128 + 32	; 128 is cbm's bias, 32 is this algo's bias
+			!set @exponent = 128 + 32	; 128 is cbm's bias, 32 is this algo's bias
 			; if mantissa is too large, shift right and adjust exponent
-			!do while .float >= (2.0 ^ 32.0) {
-				!set .float = .float >> 1
-				!set .exponent = .exponent + 1
+			!do while @float >= (2.0 ^ 32.0) {
+				!set @float = @float >> 1
+				!set @exponent = @exponent + 1
 			}
 			; if mantissa is too small, shift left and adjust exponent
-			!do while .float < (2.0 ^ 31.0) {
-				!set .float = .float << 1
-				!set .exponent = .exponent - 1
+			!do while @float < (2.0 ^ 31.0) {
+				!set @float = @float << 1
+				!set @exponent = @exponent - 1
 			}
-			!if .exponent < 1 {
+			!if @exponent < 1 {
 				!warn "FLPT underflow, using zero instead"
-				!set .float = 0
-				!set .exponent = 0
-				!set .sign = 0
+				!set @float = 0
+				!set @exponent = 0
+				!set @sign = 0
 			}
-			!if .exponent > 255 {
+			!if @exponent > 255 {
 				!error "FLPT overflow"
 			}
-			!by .exponent
-			!by 255 & int(.float >> 24)
-			!by 255 & int(.float >> 16)
-			!by 255 & int(.float >> 8)
-			!by 255 & int(.float)
-			!by .sign
+			!by @exponent
+			!by 255 & int(@float >> 24)
+			!by 255 & int(@float >> 16)
+			!by 255 & int(@float >> 8)
+			!by 255 & int(@float)
+			!by @sign
 		}
 	}
 }

ACME_Lib/cbm/ioerror.a

@@ -3,7 +3,7 @@
 !ifdef lib_cbm_ioerror_a !eof
 lib_cbm_ioerror_a = 1
 
-; if kernel i/o routine exits with carry set, A holds one of these:
+; if kernal i/o routine exits with carry set, A holds one of these:
 ioerror_BREAK			= 0
 ioerror_TOO_MANY_FILES		= 1
 ioerror_FILE_OPEN		= 2

ACME_Lib/cbm/kernal.a

@@ -8,8 +8,10 @@ lib_cbm_kernal_a = 1
 ; There are alternative names for some calls.
 
 !address {
+	; for additional c128 calls, see <cbm/c128/kernal.a>
 	k_cint		= $ff81
 	k_ioinit	= $ff84
+	; cbm-ii rom starts here:
 	k_ramtas	= $ff87
 	k_restor	= $ff8a
 	k_vector	= $ff8d
@@ -27,24 +29,58 @@ lib_cbm_kernal_a = 1
 	k_listen	= $ffb1:k_listn		= $ffb1
 	k_talk		= $ffb4
 	k_readss	= $ffb7
-	k_setlfs	= $ffba
-	k_setnam	= $ffbd	; A is length, X is ptr-low, Y is ptr-high
-	k_open		= $ffc0
-	k_close		= $ffc3:k_close_A	= $ffc3
-	k_chkin		= $ffc6:k_chkin_X	= $ffc6
-	k_chkout	= $ffc9:k_chkout_X	= $ffc9:k_ckout		= $ffc9
-	k_clrchn	= $ffcc:k_clrch		= $ffcc
-	k_chrin		= $ffcf:k_basin		= $ffcf
-	k_chrout	= $ffd2:k_basout	= $ffd2:k_bsout		= $ffd2
-	k_load		= $ffd5:k_load_AXY	= $ffd5	; A means verify, YYXX is desired load address (if channel == 0), returns end+1 in YYXX
-	k_save		= $ffd8:k_save_AXY	= $ffd8	; A is zp address of start ptr(!), YYXX is end address (+1)
-	k_settim	= $ffdb
-	k_rdtim		= $ffde
-	k_stop		= $ffe1
-	k_getin		= $ffe4:k_get		= $ffe4
+	k_setlfs	= $ffba	; set file parameters (A = logical file number, X = device, Y = secondary address)
+	k_setnam	= $ffbd	; set file name (A = length, YYXX = pointer)
+	; pet rom starts here:
+	; i/o calls: these may set C on error. in that case, A holds error code, see <cbm/ioerror.a> for the actual values.
+	k_open		= $ffc0	; open channel/file (call setlfs/setnam before!)
+	k_close		= $ffc3:k_close_A	= $ffc3	; close channel/file (A = logical file number)
+	k_chkin		= $ffc6:k_chkin_X	= $ffc6	; set input channel (X = logical file number)
+	k_chkout	= $ffc9:k_chkout_X	= $ffc9:k_ckout		= $ffc9	; set output channel (X = logical file number)
+	k_clrchn	= $ffcc:k_clrch		= $ffcc	; restore default input/output channels
+	k_chrin		= $ffcf:k_basin		= $ffcf	; read byte from current input channel (not the same as $ffe4, see note* below)
+		; A is result byte
+		; X is preserved
+		; Y gets clobbered by tape access (preserved by disk access)
+	k_chrout	= $ffd2:k_basout	= $ffd2:k_bsout		= $ffd2	; send byte to current output channel
+		; A/X/Y are preserved
+	k_load		= $ffd5:k_load_AXY	= $ffd5	; load file to memory, or verify (call setlfs/setnam before!)
+		; A: zero means LOAD, nonzero means VERIFY
+		; YYXX is desired load address (only used if secondary address == 0), returns end address plus 1
+	k_save		= $ffd8:k_save_AXY	= $ffd8	; save memory to file (call setlfs/setnam before!)
+		; A is zp address of start ptr(!)
+		; YYXX is end address plus 1
+	k_settim	= $ffdb	; set time
+	k_rdtim		= $ffde	; read time
+	k_stop		= $ffe1 ; check STOP key
+	k_getin		= $ffe4:k_get		= $ffe4	; get input byte (not the same as $ffcf, see note* below)
+		; A is result byte
+		; X is preserved
+		; Y gets clobbered by tape access (preserved by disk access)
 	k_clall		= $ffe7
 	k_udtim		= $ffea
-	k_scrorg	= $ffed
-	k_plot		= $fff0	; get/set cursor (to set, clear carry)
-	k_iobase	= $fff3
+	; pet rom stops here!?
+	k_scrorg	= $ffed	; returns screen size (X = number of columns, Y = number of lines)
+		; CAUTION: the c128 uses a new format:
+		; c128: X/Y now return maximum values in current window (so 0..39/0..24 instead of 40/25).
+		; c128: A returns max column on current screen (39 or 79)
+	k_plot		= $fff0:k_plot_CXY	= $fff0	; get/set cursor (X is line, Y is column)
+		; C = 0: set cursor position.
+		; C = 1: read cursor position.
+	k_iobase	= $fff3	; returns first i/o address (i.e. memory limit) in YYXX
+		; cbm-ii:	$dc00
+		; vic20:	$9110
+		; c64:		$d000
+		; 264:		$fd00
+		; c128:		$d000
 }
+
+;note*
+; the difference between CHRIN and GETIN depends on the current input device:
+; input device 0 (keyboard): CHRIN reads from input buffer, GETIN reads from keyboard buffer
+;	(the same difference as between INPUT and GET in basic)
+; input device 2 (rs232): CHRIN does some error handling, GETIN may just return zero on error.
+; roughly speaking, CHRIN returns a "processed" byte while GETIN returns a "raw" byte.
+; for devices on the IEC bus there should be no difference between the two calls.
+; when reading from the console (keyboard/screen), a zero byte means "no data".
+; do not expect a valid Z flag in this case! some devices may clobber the Z flag.

ACME_Lib/cbm/mflpt.a

@@ -41,47 +41,48 @@ lib_cbm_mflpt_a = 1
 
 ; this is ugly, but it gets the job done
 ; (if it's stupid, but it works, then it's not stupid)
-!macro mflpt .value {
-	!set .float = float(.value)	; make sure to do passes until value is defined
-	!ifndef .float {
-		!by $ff, $ff, $ff, $ff, $ff	; five place holder bytes
+!macro mflpt @value {
+	!set @float = float(@value)
+	!ifndef @float {
+		!by <@float, $ff, $ff, $ff, $ff	; five place holder bytes
+		; (first one depends on @float just to make sure more passes are done until value is defined)
 	} else {
 		; value is defined, so split up into sign and non-negative value
-		!if .float < 0 {
-			!set .sign = $80
-			!set .float = -.float
+		!if @float < 0 {
+			!set @sign = $80
+			!set @float = -@float
 		} else {
-			!set .sign = $00
+			!set @sign = $00
 		}
-		!if .float = 0 {
+		!if @float = 0 {
 			!by 0, 0, 0, 0, 0	; five zeroes (zero is represented by all bits zero)
 		} else {
 			; split up into exponent and mantissa
-			!set .exponent = 128 + 32	; 128 is cbm's bias, 32 is this algo's bias
+			!set @exponent = 128 + 32	; 128 is cbm's bias, 32 is this algo's bias
 			; if mantissa is too large, shift right and adjust exponent
-			!do while .float >= (2.0 ^ 32.0) {
-				!set .float = .float >> 1
-				!set .exponent = .exponent + 1
+			!do while @float >= (2.0 ^ 32.0) {
+				!set @float = @float >> 1
+				!set @exponent = @exponent + 1
 			}
 			; if mantissa is too small, shift left and adjust exponent
-			!do while .float < (2.0 ^ 31.0) {
-				!set .float = .float << 1
-				!set .exponent = .exponent - 1
+			!do while @float < (2.0 ^ 31.0) {
+				!set @float = @float << 1
+				!set @exponent = @exponent - 1
 			}
-			!if .exponent < 1 {
+			!if @exponent < 1 {
 				!warn "MFLPT underflow, using zero instead"
-				!set .float = 0
-				!set .exponent = 0
-				!set .sign = 0
+				!set @float = 0
+				!set @exponent = 0
+				!set @sign = 0
 			}
-			!if .exponent > 255 {
+			!if @exponent > 255 {
 				!error "MFLPT overflow"
 			}
-			!by .exponent
-			!by (127 & int(.float >> 24)) | .sign
-			!by 255 & int(.float >> 16)
-			!by 255 & int(.float >> 8)
-			!by 255 & int(.float)
+			!by @exponent
+			!by (127 & int(@float >> 24)) | @sign
+			!by 255 & int(@float >> 16)
+			!by 255 & int(@float >> 8)
+			!by 255 & int(@float)
 		}
 	}
 }

ACME_Lib/cbm/msbstring.a

@@ -0,0 +1,20 @@
+;ACME 0.97
+
+; macro to store a petscii string with msb set in last byte
+!macro msbstring @s {
+	!ct pet {
+		@l = len(@s)
+		!if @l < 1 {
+			!error "String is empty!"
+		}
+		!for @i, 0, @l - 1 {
+			!if $80 & @s[@i] {
+				!error "String already contains character(s) with MSB set!"
+			}
+		}
+		!for @i, 0, @l - 2 {
+			!byte @s[@i]
+		}
+		!byte $80 | @s[-1]
+	}
+}

ACME_Lib/cbm/multicolor.a

@@ -0,0 +1,94 @@
+;ACME 0.97
+
+!ifdef lib_cbm_multicolor_a !eof
+lib_cbm_multicolor_a = 1
+
+; this file contains macros to convert strings into bit patterns.
+; the idea is to use four different characters to indicate the four
+; different bit patterns of multicolor graphics, so mc sprites can be
+; "drawn" in the source code even though ACME does not support any
+; four-based number system. see the end of this file for an example.
+
+; macro to set "digit" characters
+; example:
+; +mc_set " .o#"
+!macro mc_set .s {
+	!if is_number(.s) or is_list(.s) {
+		!error "Argument to +mc_set must be a string."
+	} else if len(.s) != 4 {
+		!error "Argument to +mc_set must be four characters."
+	} else {
+		!set multicolor_alphabet = .s
+	}
+}
+
+; macro to convert string to number
+!macro mc_value ~.result, .in, .len {
+	!ifndef multicolor_alphabet {
+		!error "Called +mc_value before calling +mc_set."
+	} else if is_number(.in) or is_list(.in) {
+		!error "Argument to +mc_value must be a string."
+	} else if len(.in) != .len {
+		!error "Argument to +mc_value must have ", .len, " characters."
+	} else {
+		!set .result = 0
+		!for .idx, 0, (.len / 2) - 1 {
+			!set .char = .in[2 * .idx]	; get first of pair
+			!if .char != .in[2 * .idx + 1] {	; compare to second
+				!error "Characters in argument to +mc_value must be given in pairs."
+			} else {
+				!if .char = multicolor_alphabet[0] {
+					!set .result = (.result << 2)
+				} else if .char = multicolor_alphabet[1] {
+					!set .result = (.result << 2) + 1
+				} else if .char = multicolor_alphabet[2] {
+					!set .result = (.result << 2) + 2
+				} else if .char = multicolor_alphabet[3] {
+					!set .result = (.result << 2) + 3
+				} else {
+					!error "Characters in argument to +mc_value must be from alphabet set via +mc_set."
+				}
+			}
+		}
+	}
+}
+
+; macro for a multicolor byte (for charsets)
+!macro mc_8 .in {
+	+mc_value ~.result, .in, 8
+	!by .result
+}
+
+; macro for a multicolor sprite line
+!macro mc_be24 .in {
+	+mc_value ~.result, .in, 24
+	!be24 .result
+}
+
+!eof
+	; Here's an example on how to use this:
+	!to "mc-sprites.prg", cbm
+	*=$e00
+	+mc_set " .o#"	; set four characters
+	; and now use those four characters to "paint" the sprite:
+	+mc_be24 "                        "
+	+mc_be24 "..                    .."
+	+mc_be24 "....                ...."
+	+mc_be24 "......  ##....##  ......"
+	+mc_be24 "  ....................  "
+	+mc_be24 "  ....................  "
+	+mc_be24 "    ................    "
+	+mc_be24 "  ######........######  "
+	+mc_be24 "  ..oo####....####oo..  "
+	+mc_be24 "##..oo  ########  oo..##"
+	+mc_be24 "....oo  oo....oo  oo...."
+	+mc_be24 "......oooo....oooo......"
+	+mc_be24 "........................"
+	+mc_be24 "......oooo....oooo......"
+	+mc_be24 "....oooooooooooooooo...."
+	+mc_be24 "....  oooooooooooo  ...."
+	+mc_be24 "..  ####oooooooo####  .."
+	+mc_be24 "..  ######oooo######  .."
+	+mc_be24 "    ######    ####      "
+	+mc_be24 "    ######    ##        "
+	+mc_be24 "                        "

ACME_Lib/m65/std.a

@@ -0,0 +1,13 @@
+;ACME 0.96.5
+
+!ifdef lib_m65_std_a !eof
+lib_m65_std_a = 1
+
+	; macro to load immediate constant:
+	!macro movq @v {
+		; going from lsb to msb, so at least the N flag is correct:
+		lda #<@v
+		ldx #>@v
+		ldy #^@v
+		ldz #@v >>> 24
+	}

README.md

@@ -0,0 +1,12 @@
+# ACME
+ Multi-platform cross-assembler for MOS 6502/65C02/6510/65816 CPUs 
+
+ACME is a free cross assembler released under the GNU GPL.
+It can produce code for the following processors: 6502, 6510 (including illegal opcodes), 65c02 and 65816.
+ACME supports the standard assembler stuff like global/local/anonymous labels, offset assembly, conditional assembly and looping assembly. It can include other source files as well as binaries while assembling.
+Calculations can be done in integer or float mode.
+Oh, and it is fast.
+
+Imported from SourceForge SVN repository: https://sourceforge.net/projects/acme-crossass/
+
+Release tags added - based on SVN commit messages

contrib/joe_syntax/INSTALL

@@ -8,7 +8,7 @@ If you destroy your system, don't come whining to me.
 --------------------
 
 1) Copy the syntax file to the correct directory by typing:
-cp acme.jsf /etc/joe/syntax/
+cp acme.jsf /usr/share/joe/syntax/
 
 2) Add the following lines to the "SECOND SECTION" of "/etc/joe/joerc":
 

contrib/joe_syntax/acme.jsf

@@ -4,6 +4,11 @@
 # new in version 5: changed mnemo colors
 # new in version 6: added !ifndef, !addr
 # new in version 7: added !symbollist
+# new in version 8: adjusted for ACME 0.97
+#	added backslash escaping,
+#	added "//" comments,
+#	added new mnemonics, keywords and pseudo opcodes,
+#	reduced colors for different instruction sets
 
 # define colors
 #
@@ -18,16 +23,13 @@
 =Call		bold
 =Comment 	green
 =Constant 	cyan
+=Escape		bold cyan
 =Keyword 	bold
 =Pseudo		bold
 =Mnemo6502	bold yellow
 =PCMnemo6502	bold red
-=Mnemo6510	bg_red bold yellow
-=PCMnemo6510	bg_red bold red
-=Mnemo65c02	bg_cyan bold yellow
-=PCMnemo65c02	bg_cyan bold red
-=Mnemo65816	bg_blue bold yellow
-=PCMnemo65816	bg_blue bold red
+=MnemoExt	bg_blue bold yellow
+=PCMnemoExt	bg_blue bold red
 
 :reset Idle
 	*		idle		noeat
@@ -36,6 +38,7 @@
 :idle Idle
 	*		idle
 	";"		line_comment	recolor=-1
+	"//"		line_comment	recolor=-1
 	":{\n"		reset
 	"!.a-zA-Z_€-"	checkstring	recolor=-1 buffer
 	"+"		anonf_or_macro	recolor=-1
@@ -101,10 +104,18 @@
 :string	Constant
 	*		string
 	"\""		idle
+	"\\"		string_escape	recolor=-1
+
+:string_escape Escape
+	*		string
 
 :char Constant
 	*		char
 	"'"		idle
+	"\\"		char_escape	recolor=-1
+
+:char_escape Escape
+	*		char
 
 :ident Idle
 	*		idle		noeat
@@ -117,10 +128,16 @@
 	"!by"		pseudo
 	"!byte"		pseudo
 	"!16"		pseudo
+	"!le16"		pseudo
+	"!be16"		pseudo
 	"!wo"		pseudo
 	"!word"		pseudo
 	"!24"		pseudo
+	"!le24"		pseudo
+	"!be24"		pseudo
 	"!32"		pseudo
+	"!le32"		pseudo
+	"!be32"		pseudo
 	"!tx"		pseudo
 	"!text"		pseudo
 	"!raw"		pseudo
@@ -148,10 +165,17 @@
 	"!set"		pseudo
 	"!macro"	pseudo
 	"!if"		pseudo
-	"!do"		pseudo
-	"!for"		pseudo
 	"!ifdef"	pseudo
 	"!ifndef"	pseudo
+	"else"		keyword
+	"if"		keyword
+	"ifdef"		keyword
+	"ifndef"	keyword
+	"!for"		pseudo
+	"!while"	pseudo
+	"!do"		pseudo
+	"until"		keyword
+	"while"		keyword
 	"!al"		pseudo
 	"!as"		pseudo
 	"!rl"		pseudo
@@ -162,6 +186,10 @@
 	"!serious"	pseudo
 	"!addr"		pseudo
 	"!address"	pseudo
+	"!h"		pseudo
+	"!hex"		pseudo
+	"!xor"		pseudo
+	"!skip"		pseudo
 	"ora"	mnemo6502
 	"asl"	mnemo6502
 	"and"	mnemo6502
@@ -205,9 +233,9 @@
 	"inx"	mnemo6502
 	"nop"	mnemo6502
 	"sed"	mnemo6502
+	"jsr"	mnemo6502
 	"brk"	pcmnemo6502
 	"jmp"	pcmnemo6502
-	"jsr"	pcmnemo6502
 	"bpl"	pcmnemo6502
 	"bmi"	pcmnemo6502
 	"bvc"	pcmnemo6502
@@ -218,61 +246,149 @@
 	"beq"	pcmnemo6502
 	"rti"	pcmnemo6502
 	"rts"	pcmnemo6502
-	"phy"	mnemo65c02
-	"ply"	mnemo65c02
-	"phx"	mnemo65c02
-	"plx"	mnemo65c02
-	"tsb"	mnemo65c02
-	"trb"	mnemo65c02
-	"stz"	mnemo65c02
-	"bra"	pcmnemo65c02
-	"wai"	mnemo65816
-	"pei"	mnemo65816
-	"per"	mnemo65816
-	"mvp"	mnemo65816
-	"mvn"	mnemo65816
-	"rep"	mnemo65816
-	"sep"	mnemo65816
-	"pea"	mnemo65816
-	"phd"	mnemo65816
-	"tcs"	mnemo65816
-	"pld"	mnemo65816
-	"tsc"	mnemo65816
-	"wdm"	mnemo65816
-	"phk"	mnemo65816
-	"tcd"	mnemo65816
-	"tdc"	mnemo65816
-	"phb"	mnemo65816
-	"txy"	mnemo65816
-	"plb"	mnemo65816
-	"tyx"	mnemo65816
-	"xba"	mnemo65816
-	"xce"	mnemo65816
-	"brl"	pcmnemo65816
-	"cop"	pcmnemo65816
-	"jml"	pcmnemo65816
-	"jsl"	pcmnemo65816
-	"rtl"	pcmnemo65816
-	"stp"	pcmnemo65816
-	"slo"	mnemo6510
-	"rla"	mnemo6510
-	"sre"	mnemo6510
-	"rra"	mnemo6510
-	"sax"	mnemo6510
-	"lax"	mnemo6510
-	"dcp"	mnemo6510
-	"isc"	mnemo6510
-	"anc"	mnemo6510
-	"asr"	mnemo6510
-	"arr"	mnemo6510
-	"sbx"	mnemo6510
-	"dop"	mnemo6510
-	"top"	mnemo6510
-	"lxa"	mnemo6510
-	"jam"	pcmnemo6510
-	"else"	keyword
-	"until"	keyword
-	"while"	keyword
+	"phy"	mnemoExt
+	"ply"	mnemoExt
+	"phx"	mnemoExt
+	"plx"	mnemoExt
+	"tsb"	mnemoExt
+	"trb"	mnemoExt
+	"stz"	mnemoExt
+	"bra"	pcmnemoExt
+	"rmb0"	mnemoExt
+	"bbr0"	mnemoExt
+	"smb0"	mnemoExt
+	"bbs0"	mnemoExt
+	"rmb1"	mnemoExt
+	"bbr1"	mnemoExt
+	"smb1"	mnemoExt
+	"bbs1"	mnemoExt
+	"rmb2"	mnemoExt
+	"bbr2"	mnemoExt
+	"smb2"	mnemoExt
+	"bbs2"	mnemoExt
+	"rmb3"	mnemoExt
+	"bbr3"	mnemoExt
+	"smb3"	mnemoExt
+	"bbs3"	mnemoExt
+	"rmb4"	mnemoExt
+	"bbr4"	mnemoExt
+	"smb4"	mnemoExt
+	"bbs4"	mnemoExt
+	"rmb5"	mnemoExt
+	"bbr5"	mnemoExt
+	"smb5"	mnemoExt
+	"bbs5"	mnemoExt
+	"rmb6"	mnemoExt
+	"bbr6"	mnemoExt
+	"smb6"	mnemoExt
+	"bbs6"	mnemoExt
+	"rmb7"	mnemoExt
+	"bbr7"	mnemoExt
+	"smb7"	mnemoExt
+	"bbs7"	mnemoExt
+	"wai"	mnemoExt
+	"pei"	mnemoExt
+	"per"	mnemoExt
+	"mvp"	mnemoExt
+	"mvn"	mnemoExt
+	"rep"	mnemoExt
+	"sep"	mnemoExt
+	"pea"	mnemoExt
+	"phd"	mnemoExt
+	"tcs"	mnemoExt
+	"pld"	mnemoExt
+	"tsc"	mnemoExt
+	"wdm"	mnemoExt
+	"phk"	mnemoExt
+	"tcd"	mnemoExt
+	"tdc"	mnemoExt
+	"phb"	mnemoExt
+	"txy"	mnemoExt
+	"plb"	mnemoExt
+	"tyx"	mnemoExt
+	"xba"	mnemoExt
+	"xce"	mnemoExt
+	"brl"	pcmnemoExt
+	"cop"	mnemoExt
+	"jml"	pcmnemoExt
+	"jsl"	mnemoExt
+	"rtl"	pcmnemoExt
+	"stp"	pcmnemoExt
+	"slo"	mnemoExt
+	"rla"	mnemoExt
+	"sre"	mnemoExt
+	"rra"	mnemoExt
+	"sax"	mnemoExt
+	"lax"	mnemoExt
+	"dcp"	mnemoExt
+	"isc"	mnemoExt
+	"anc"	mnemoExt
+	"ane"	mnemoExt
+	"asr"	mnemoExt
+	"arr"	mnemoExt
+	"alr"	mnemoExt
+	"sbx"	mnemoExt
+	"sha"	mnemoExt
+	"shx"	mnemoExt
+	"shy"	mnemoExt
+	"las"	mnemoExt
+	"tas"	mnemoExt
+	"dop"	mnemoExt
+	"top"	mnemoExt
+	"lxa"	mnemoExt
+	"jam"	pcmnemoExt
+	"map"	mnemoExt
+	"eom"	mnemoExt
+	"aug"	mnemoExt
+	"sac"	mnemoExt
+	"sir"	mnemoExt
+	"orq"	mnemoExt
+	"aslq"	mnemoExt
+	"inq"	mnemoExt
+	"bitq"	mnemoExt
+	"andq"	mnemoExt
+	"rolq"	mnemoExt
+	"deq"	mnemoExt
+	"asrq"	mnemoExt
+	"eorq"	mnemoExt
+	"lsrq"	mnemoExt
+	"adcq"	mnemoExt
+	"rorq"	mnemoExt
+	"stq"	mnemoExt
+	"ldq"	mnemoExt
+	"cpq"	mnemoExt
+	"sbcq"	mnemoExt
+	"cle"	mnemoExt
+	"see"	mnemoExt
+	"tsy"	mnemoExt
+	"inz"	mnemoExt
+	"tys"	mnemoExt
+	"dez"	mnemoExt
+	"neg"	mnemoExt
+	"taz"	mnemoExt
+	"tab"	mnemoExt
+	"bsr"	mnemoExt
+	"tza"	mnemoExt
+	"tba"	mnemoExt
+	"ldz"	mnemoExt
+	"cpz"	mnemoExt
+	"dew"	mnemoExt
+	"asw"	mnemoExt
+	"phz"	mnemoExt
+	"inw"	mnemoExt
+	"row"	mnemoExt
+	"phw"	mnemoExt
+	"plz"	mnemoExt
+	"lbpl"	pcmnemoExt
+	"lbmi"	pcmnemoExt
+	"lbvc"	pcmnemoExt
+	"lbvs"	pcmnemoExt
+	"lbra"	pcmnemoExt
+	"lbcc"	pcmnemoExt
+	"lbcs"	pcmnemoExt
+	"lbne"	pcmnemoExt
+	"lbeq"	pcmnemoExt
+	"rtn"	pcmnemoExt
 done
 	"!a-zA-Z0-9"	checkstring
 #	" \t"		idle		noeat
@@ -283,17 +399,9 @@ done
 	*		idle		noeat
 :pcmnemo6502	PCMnemo6502
 	*		idle		noeat
-:mnemo65c02	Mnemo65c02
+:mnemoExt	MnemoExt
 	*		idle		noeat
-:pcmnemo65c02	PCMnemo65c02
-	*		idle		noeat
-:mnemo65816	Mnemo65816
-	*		idle		noeat
-:pcmnemo65816	PCMnemo65816
-	*		idle		noeat
-:mnemo6510	Mnemo6510
-	*		idle		noeat
-:pcmnemo6510	PCMnemo6510
+:pcmnemoExt	PCMnemoExt
 	*		idle		noeat
 :keyword	Keyword
 	*		idle		noeat

contrib/joe_syntax/color.a

@@ -1,27 +1,22 @@
-;ACME 0.91	; comments are green
+;ACME 0.97	; comments are green
 !serious "This file is not meant to be assembled."
 
-binary1=%00001000	; label names are grey, constants are cyan
-binary2=%....#...
-octal=&0123456789	; bad constants are bold red
-decimal=63
-hex1=0xcd
-hex2=$ef
+	binary1	= %00001000	; label names are grey, constants are cyan
+	binary2	= %....#...
+	octal	= &0123456789	; bad constants are bold red
+	decimal	= 63
+	hex1	= 0xcd
+	hex2	= $ef
 !sl "labeldump.l"	; strings are cyan
-	*=$1300
+	* = $1300
 	+dings		; macro calls are bold
 	else		; keyword: bold
 	!eof		; pseudo: bold
 --			; anonymous labels should be bold (white)
-	; 6502 mnemonics
-		nop	; normal ones are yellow
-		rts	; PC-changing ones are red
-	; illegals
-		dop	; most of them are yellow on red
-		jam	; this single one's red on red. Guess why.
-	; 65c02 extensions
-		stz	; normal ones are yellow on cyan
-		bra	; PC-changing ones (just "BRA") are red
-	; 65816 extensions
-		xce	; yellow on blue
-		cop	; PC-changing ones are red
+	; base 6502 mnemonics:
+		inx	; normal ones are yellow,
+		beq --	; all that break sequential flow are red
+		rts
+	; all extended instruction sets:
+		stz	; normal ones are yellow on blue
+		bra	; flow-breaking ones are red

contrib/toacme/src/Makefile.mingw

@@ -0,0 +1,59 @@
+CFLAGS		= -O3 -Wall -Wstrict-prototypes
+#LIBS		= -lm
+CC		= gcc
+RM		= rm
+
+#SRC		=
+
+PROGS		= toacme.exe
+
+all: $(PROGS)
+
+vis.o: config.h acme.h io.h mnemo.h scr2iso.h vis.c
+
+ab3.o: config.h ab.h acme.h io.h mnemo.h scr2iso.h ab3.c
+
+ab.o: config.h ab.h acme.h io.h scr2iso.h ab.c
+
+f8ab.o: config.h ab.h acme.h io.h mnemo.h scr2iso.h f8ab.c
+
+giga.o: config.h acme.h gighyp.h io.h mnemo.h pet2iso.h giga.c
+
+gighyp.o: config.h acme.h io.h pet2iso.h gighyp.h gighyp.c
+
+hypra.o: config.h acme.h gighyp.h io.h pet2iso.h hypra.c
+
+obj.o: config.h acme.h io.h mnemo.h obj.c
+
+acme.o: config.h acme.h acme.c
+
+main.o: config.h version.h main.c
+
+mnemo.o: config.h mnemo.c
+
+pet2iso.o: config.h pet2iso.h pet2iso.c
+
+platform.o: config.h platform.h platform.c
+
+prof.o: config.h prof.c
+
+scr2iso.o: config.h scr2iso.h scr2iso.c
+
+version.o: config.h version.c
+
+toacme.exe: vis.o ab.o ab3.o acme.o f8ab.o giga.o gighyp.o hypra.o io.o main.o mnemo.o obj.o pet2iso.o platform.o prof.o scr2iso.o version.o resource.res
+	$(CC) $(LIBS) $(CFLAGS) -o toacme vis.o ab.o ab3.o acme.o f8ab.o giga.o gighyp.o hypra.o io.o main.o mnemo.o obj.o pet2iso.o platform.o prof.o scr2iso.o version.o resource.res
+	strip toacme.exe
+
+win/resource.rc: main.c
+	cd win; sh setRelease.sh
+
+resource.res: win/resource.rc win/logo.ico
+	cd win; windres resource.rc -O coff -o ../resource.res
+	cp -f win/logo.ico .
+
+clean:
+	-$(RM) -f *.o $(PROGS) *~ core resource.res logo.ico win/resource.rc
+
+
+# DO NOT DELETE

contrib/toacme/src/mnemo.c

@@ -78,6 +78,7 @@ const char	MnemonicSLO[]	= " SLO",
 		MnemonicSBX[]	= " SBX",
 		MnemonicDOP[]	= " DOP",
 		MnemonicTOP[]	= " TOP",
+		MnemonicSHX[]	= " SHX",
 		MnemonicJAM[]	= " JAM";
 
 // mnemonics of 65c02 instructions

contrib/toacme/src/mnemo.h

@@ -38,6 +38,7 @@ extern const char	MnemonicDCP[], MnemonicDOP[], MnemonicISC[];
 extern const char	MnemonicJAM[], MnemonicLAX[], MnemonicRLA[];
 extern const char	MnemonicRRA[], MnemonicSAX[], MnemonicSBX[];
 extern const char	MnemonicSLO[], MnemonicSRE[], MnemonicTOP[];
+extern const char	MnemonicSHX[];
 
 // mnemonics of 65c02 instructions
 extern const char	MnemonicBRA[];

contrib/toacme/src/obj.c

@@ -54,7 +54,7 @@ const char	*mnemo_of_code[]	= {
 	MnemonicBCC,	MnemonicSTA,	" JAM;0x92",	NULL,		// $90-$93
 	MnemonicSTY,	MnemonicSTA,	MnemonicSTX,	MnemonicSAX,	// $94-$97
 	MnemonicTYA,	MnemonicSTA,	MnemonicTXS,	NULL,		// $98-$9b
-	NULL,		MnemonicSTA,	NULL,		NULL,		// $9c-$9f
+	NULL,		MnemonicSTA,	MnemonicSHX,	NULL,		// $9c-$9f
 	MnemonicLDY,	MnemonicLDA,	MnemonicLDX,	MnemonicLAX,	// $a0-$a3
 	MnemonicLDY,	MnemonicLDA,	MnemonicLDX,	MnemonicLAX,	// $a4-$a7
 	MnemonicTAY,	MnemonicLDA,	MnemonicTAX,	NULL,		// $a8-$ab
@@ -217,7 +217,7 @@ int (*addressing_mode_of_code[])(void)	= {
 	am_relative,	am_indirect_y,	am_implied,	am_implied,	// $90-$93
 	am_abs_x8,	am_abs_x8,	am_abs_y8,	am_abs_y8,	// $94-$97
 	am_implied,	am_abs_y16,	am_implied,	am_implied,	// $98-$9b
-	am_implied,	am_abs_x16,	am_implied,	am_implied,	// $9c-$9f
+	am_implied,	am_abs_x16,	am_abs_y16,	am_implied,	// $9c-$9f
 	am_immediate,	am_indirect_x,	am_immediate,	am_indirect_x,	// $a0-$a3
 	am_absolute8,	am_absolute8,	am_absolute8,	am_absolute8,	// $a4-$a7
 	am_implied,	am_immediate,	am_implied,	am_implied,	// $a8-$ab

contrib/toacme/src/prof.c

@@ -1,5 +1,5 @@
 // ToACME - converts other source codes to ACME format.
-// Copyright (C) 1999-2016 Marco Baye
+// Copyright (C) 1999-2019 Marco Baye
 // Have a look at "main.c" for further info
 //
 // "Professional Ass by Oliver Stiller" stuff (NOT "Profi-Ass" by Data Becker!)
@@ -40,6 +40,12 @@
 #define REV_START	";-=#"
 #define REV_END		"#=-"
 
+enum linestate {
+	LINESTATE_REMOVEPLUS,		// before label def
+	LINESTATE_DOUBLENEXTSPACE,	// entered when removing '+'
+	LINESTATE_NOTHINGSPECIAL	// afterward
+};
+
 // main
 void prof_main(void)
 {
@@ -49,6 +55,7 @@ void prof_main(void)
 		byte,
 		hibit,
 		length2;
+	enum linestate	linestate;
 
 	IO_set_input_padding(0);
 	IO_process_load_address();
@@ -57,7 +64,8 @@ void prof_main(void)
 		if (length1 == EOF)
 			break;
 		if (length1 < 3) {
-			fprintf(stderr, "Error: Short line (%d bytes)\n", length1);
+			fprintf(stderr, "Error: Short line (%d bytes), stopping.\n", length1);
+			return;
 		}
 		// read amount of indentation and output tabs/spaces
 		indent = IO_get_byte();
@@ -71,13 +79,34 @@ void prof_main(void)
 		}
 		// now convert line
 		hibit = 0;
+		linestate = LINESTATE_REMOVEPLUS;
 		for (ii = 0; ii < length1 - 3; ii++) {
 			byte = IO_get_byte();
 			if ((byte & 128) != hibit) {
 				hibit = byte & 128;
 				IO_put_string(hibit ? REV_START : REV_END);
 			}
-			IO_put_byte(SCR2ISO(byte & 127));
+			byte = SCR2ISO(byte & 127);
+			// outside of comments, remove leading '+'
+			if (hibit == 0) {
+				if (byte == '+') {
+					if (linestate == LINESTATE_REMOVEPLUS) {
+						linestate = LINESTATE_DOUBLENEXTSPACE;
+						continue;	// eat '+'
+					}
+				} else if (byte == ' ') {
+					if (linestate == LINESTATE_DOUBLENEXTSPACE) {
+						linestate = LINESTATE_NOTHINGSPECIAL;
+						IO_put_byte(' ');	// add space to compensate for eaten '+'
+					}
+				} else {
+					// any other char -> do not remove any '+'
+					if (linestate == LINESTATE_REMOVEPLUS) {
+						linestate = LINESTATE_NOTHINGSPECIAL;
+					}
+				}
+			}
+			IO_put_byte(byte);
 		}
 		if (hibit)
 			IO_put_string(REV_END);

contrib/toacme/src/version.c

@@ -1,15 +1,15 @@
 // ToACME - converts other source codes to ACME format.
-// Copyright (C) 1999-2015 Marco Baye
+// Copyright (C) 1999-2019 Marco Baye
 // Have a look at "main.c" for further info
 //
 // Version
 
-#define RELEASE_NUMBER	"0.13"		// change before release (FIXME)
-#define CHANGE_DATE	"16 Feb"		// change before release
-#define CHANGE_YEAR	"2016"		// change before release
+#define RELEASE_NUMBER	"0.15"		// change before release (FIXME)
+#define CHANGE_DATE	"25 Apr"		// change before release
+#define CHANGE_YEAR	"2019"		// change before release
 #define HOME_PAGE	"http://sourceforge.net/projects/acme-crossass/"
 //			"http://home.pages.de/~mac_bacon/smorbrod/acme/"
-#define FILE_TAG	";ACME 0.95.6"	// check before release
+#define FILE_TAG	";ACME 0.96.4"	// check before release
 
 #include <stdio.h>
 #include <string.h>

contrib/toacme/src/win/setRelease.sh

@@ -0,0 +1,58 @@
+#/bin/bash
+#
+# Get release and create RC-File
+#
+function DEBUG()
+{
+    [ "$_DEBUG" == "on" ] &&  $@
+}
+function pause()
+{
+    read -p "Weiter mit Eingabe" $a
+}
+FILE="resource.rc"
+RELEASE=`grep -m1 "define RELEASE"  ../version.c  | cut -f2 | tr -d '"'`
+DEBUG echo $RELEASE
+VERSION=${RELEASE//./,},0
+DEBUG echo $VERSION
+FILEVERSION=\""$RELEASE ${CODENAME//\"/}"\"
+DEBUG echo $FILEVERSION
+CHANGE_YEAR=`grep -m1 "define CHANGE_YEAR"  ../version.c  | cut -f2 | tr -d '"'`
+DEBUG echo $CHANGE_YEAR
+
+cat << EndOfFile > $FILE
+// Iconfile (64/32/16)
+ID ICON "Logo.ico"
+
+// Infos for windows
+1 VERSIONINFO
+FILEVERSION     $VERSION
+PRODUCTVERSION  $VERSION
+FILEFLAGSMASK   0x3fL
+FILEFLAGS       0x0L
+FILEOS          0x40004L
+FILETYPE        0x2L
+FILESUBTYPE     0x0L
+BEGIN
+  BLOCK "StringFileInfo"
+  BEGIN
+    BLOCK "040904E4"
+    BEGIN
+      VALUE "CompanyName", "Smørbrød Software"
+      VALUE "FileDescription", "ToAcme Converter"
+      VALUE "FileVersion", $FILEVERSION
+      VALUE "InternalName", "ToACME Converter"
+      VALUE "LegalCopyright", "Copyright © $CHANGE_YEAR Marco Baye"
+      VALUE "OriginalFilename", "toacme.exe"
+      VALUE "ProductName", "ToACME Converter"
+      VALUE "ProductVersion", $FILEVERSION
+      VALUE "PorductLicence","GNU General Public License"
+      VALUE "WindowsPort","Dirk Höpfner hoeppie@gmx.de"
+    END
+  END
+  BLOCK "VarFileInfo"
+  BEGIN
+    VALUE "Translation", 0x409, 1252
+  END
+END
+EndOfFile

docs/65816.txt

@@ -12,10 +12,10 @@ ACME.
 
 
 ----------------------------------------------------------------------
-Section:   Command aliases for "long" JMPs and JSRs
+Section:   Aliases for "long" JMPs and JSRs
 ----------------------------------------------------------------------
 
-In addition to the commands JMP and JSR, the 65816 processor also
+In addition to the mnemonics JMP and JSR, the 65816 processor also
 knows JML and JSL, which are JMP and JSR using new (long) addressing
 modes. ACME also accepts the new addressing modes when using the old
 mnemonics JMP and JSR, but the old addressing modes cannot be used
@@ -30,12 +30,13 @@ According to WDC's official syntax for 65816 assembly language, the
 argument order of the MVN and MVP instructions differs between
 assembly language and machine code.
 To copy bytes from bank $ab to bank $cd, use the following statement:
-    mvn $ab, $cd ; source bank $ab, destination bank $cd
+    mvn $ab, $cd        ; source bank $ab, destination bank $cd
+or
+    mvn #$ab, #$cd      ; source bank $ab, destination bank $cd
 ACME will then produce the following machine code:
-    $54 $cd $ab  ; opcode mvn, destination bank $cd, source bank $ab
+    $54 $cd $ab ; opcode mvn, destination bank $cd, source bank $ab
 
-ACME 0.05 and earlier did it the wrong way. Several other assemblers
-still do. Make sure your sources are correct.
+ACME 0.05 and earlier did it the wrong way.
 
 
 ----------------------------------------------------------------------
@@ -54,13 +55,13 @@ this at any time using the following pseudo opcodes:
 		!rs	; switch to short index registers
 
 Please note that ACME, unlike some other assemblers, does *not* track
-SEP/REP commands: I don't like that method - it fails when
+SEP/REP instructions: I don't like that method - it fails when
 encountering PLPs, for example. So if it doesn't work reliably in the
 first place, why use it? :)
 
 If you don't like that you always have to use a pseudo opcode
-alongside SEP/REP commands, then have a look at the file <65816/std.a>
-(in the library). There are some predefined macros that you can use.
+alongside SEP/REP instructions, then have a look at the library file
+<65816/std.a> which has some predefined macros you can use.
 
 
 ----------------------------------------------------------------------

docs/AddrModes.txt

@@ -7,12 +7,12 @@
                        --- addressing modes ---
 
 
-If a command can be used with different addressing modes, ACME has to
-decide which one to use. Several commands of the 6502 CPU can be used
-with either "absolute" addressing or "zeropage-absolute" addressing.
-The former one means there's a 16-bit argument, the latter one means
-there's an 8-bit argument.
-And the 65816 CPU even knows some commands with 24-bit addressing...
+If an instruction can be used with different addressing modes, ACME
+has to decide which one to use. Several instructions of the 6502 CPU
+can be used with either "absolute" addressing or "zeropage-absolute"
+addressing. The former one means there's a 16-bit argument, the latter
+one means there's an 8-bit argument.
+And the 65816 CPU even has some instructions with 24-bit addressing...
 
 So how does ACME know which addressing mode to use?
 The simple approach is to always use the smallest possible argument,
@@ -29,11 +29,10 @@ The two exceptions are:
 
 
 
-
 ***   1) Symbols are defined too late
 
 If ACME cannot figure out the argument value in the first pass, it
-assumes that the command uses 16-bit addressing.
+assumes that the instruction uses 16-bit addressing.
 
 If it later finds out that the argument only needs 8 bits, ACME gives
 a warning ("using oversized addressing mode") and continues. However,
@@ -43,8 +42,7 @@ These problems can be solved by defining the symbols *before* using
 them, so that the value can be figured out in the first pass. If this
 is not possible, you can use the postfix method, effectively exactly
 defining what addressing mode to use. The postfix method is described
-in a separate paragraph below.
-
+in a separate section below.
 
 
 
@@ -85,10 +83,12 @@ will be assembled to
   ad fd 00	; lda $00fd
   8f ff 00 00	; sta $0000ff
 
-The other possibility is to use the postfix method (described in the
-next paragraph).
+This feature can be disabled using the "--ignore-zeroes" CLI switch.
 
 
+The other possibility is to use the postfix method (described in the
+next section).
+
 
 
 ***   The postfix method
@@ -129,7 +129,7 @@ will be assembled to
   8c fd 00	; sty $00fd
   8f ff 00 00	; sta $0000ff
 
-Postfixes given directly after the command have higher priority than
+Postfixes added directly to the mnemonic have higher priority than
 those given to the argument. As you can see, you can add the postfix
 to the symbol definition as well (equivalent to leading zeros).
 
@@ -138,8 +138,7 @@ gives the high byte and "^" gives the bank byte of a value) to any
 value will clear the argument's Force Bits 2 and 3 and set Force
 Bit 1 instead. So "lda <symbol" will use 8-bit addressing, regardless
 of the symbol's Force Bits. Of course, you can change this by
-postfixing the command again... :)
-
+postfixing the instruction again... :)
 
 
 
@@ -149,7 +148,7 @@ You don't need to read this paragraph just to use ACME, I only
 included it for completeness' sake. This is a description of ACME's
 strategy for finding the addressing mode to use:
 
-First, ACME checks whether the command has any postfix. If it has,
+First, ACME checks whether the instruction has any postfix. If it has,
 ACME acts upon it. So postfixes have the highest priority.
 
 Otherwise, ACME checks whether the argument has any Force Bits set

docs/AllPOs.txt

@@ -22,17 +22,18 @@ Purpose:	Insert 8-bit values.
 Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	"!08", "!by", "!byte"
 Examples:	!08 127, symbol, -128	; output some values
-		!by 14, $3d, %0110, &304, <*, "c"
+		!by 14, $3d, %0110, &304, <*, 'c'
 		!byte 3 - 4, symbol1 XOR symbol2, 2 ^ tz, (3+4)*7
 
 
 Call:		!16 EXPRESSION [, EXPRESSION]*
 Purpose:	Insert 16-bit values in chosen CPU's byte order.
 Parameters:	EXPRESSION: Any formula the value parser accepts.
-Aliases:	"!wo", "!word" (and because all currently supported CPUs are
-		little-endian, "!le16" is in fact another alias)
+Aliases:	"!wo", "!word" (and because all currently supported
+		CPUs are little-endian, "!le16" is in fact another
+		alias)
 Examples:	!16 65535, symbol, -32768	; output some values
-		!wo 14, $4f35, %100101010010110, &36304, *, "c"
+		!wo 14, $4f35, %100101010010110, &36304, *, 'c'
 		!word 3000 - 4, a1 AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 Call:		!le16 EXPRESSION [, EXPRESSION]*
@@ -41,7 +42,7 @@ Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None (but because all currently supported CPUs are
 		little-endian, "!16/!wo/!word" are in fact aliases)
 Examples:	!le16 65535, symbol, -32768	; output some values
-		!le16 14, $4f35, %100101010010110, &36304, *, "c"
+		!le16 14, $4f35, %100101010010110, &36304, *, 'c'
 		!le16 3000 - 4, a1 AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 Call:		!be16 EXPRESSION [, EXPRESSION]*
@@ -49,7 +50,7 @@ Purpose:	Insert 16-bit values in big-endian byte order.
 Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None
 Examples:	!be16 65535, symbol, -32768	; output some values
-		!be16 14, $4f35, %100101010010110, &36304, *, "c"
+		!be16 14, $4f35, %100101010010110, &36304, *, 'c'
 		!be16 3000 - 4, a1 AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 
@@ -59,7 +60,7 @@ Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None (but because all currently supported CPUs are
 		little-endian, "!le24" is in fact an alias)
 Examples:	!24 16777215, symbol, -8388608, 14, $6a4f35
-		!24 %10010110100101010010110, &47336304, *, "c"
+		!24 %10010110100101010010110, &47336304, *, 'c'
 		!24 300000 - 4, a1 AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 Call:		!le24 EXPRESSION [, EXPRESSION]*
@@ -68,7 +69,7 @@ Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None (but because all currently supported CPUs are
 		little-endian, "!24" is in fact an alias)
 Examples:	!le24 16777215, symbol, -8388608, 14, $6a4f35
-		!le24 %10010110100101010010110, &47336304, *, "c"
+		!le24 %10010110100101010010110, &47336304, *, 'c'
 		!le24 300000 - 4, a1 AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 Call:		!be24 EXPRESSION [, EXPRESSION]*
@@ -76,7 +77,7 @@ Purpose:	Insert 24-bit values in big-endian byte order.
 Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None
 Examples:	!be24 16777215, symbol, -8388608, 14, $6a4f35
-		!be24 %10010110100101010010110, &47336304, *, "c"
+		!be24 %10010110100101010010110, &47336304, *, 'c'
 		!be24 300000 - 4, a1 AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 
@@ -86,7 +87,7 @@ Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None (but because all currently supported CPUs are
 		little-endian, "!le32" is in fact an alias)
 Examples:	!32 $7fffffff, symbol, -$80000000, 14, $46a4f35
-		!32 %1001011010010101001011010010, &4733630435, *, "c"
+		!32 %1001011010010101001011010010, &4733630435, *, 'c'
 		!32 300000 - 4, a AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 Call:		!le32 EXPRESSION [, EXPRESSION]*
@@ -95,7 +96,7 @@ Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None (but because all currently supported CPUs are
 		little-endian, "!32" is in fact an alias)
 Examples:	!le32 $7fffffff, symbol, -$80000000, 14, $46a4f35
-		!le32 %1001011010010101001011010010, &4733630435, *, "c"
+		!le32 %1001011010010101001011010010, &4733630435, *, 'c'
 		!le32 300000 - 4, a AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 Call:		!be32 EXPRESSION [, EXPRESSION]*
@@ -103,10 +104,27 @@ Purpose:	Insert 32-bit values in big-endian byte order.
 Parameters:	EXPRESSION: Any formula the value parser accepts.
 Aliases:	None
 Examples:	!be32 $7fffffff, symbol, -$80000000, 14, $46a4f35
-		!be32 %1001011010010101001011010010, &4733630435, *, "c"
+		!be32 %1001011010010101001011010010, &4733630435, *, 'c'
 		!be32 300000 - 4, a AND a2, 2 ^ tz, (3+4)*70, l1 & .j2
 
 
+Call:		!hex PAIRS_OF_HEX_DIGITS
+Purpose:	Insert byte values with a minimum of additional
+		syntax. This pseudo opcode was added for easier
+		writing of external source code generator tools.
+Parameters:	PAIRS_OF_HEX_DIGITS: Just hexadecimal digits, without
+		any "0x" or "$" prefix. Spaces and TABs are allowed,
+		but not needed to separate the byte values.
+Aliases:	"!h"
+Examples:	!h f0 f1 f2 f3 f4 f5 f6 f7	; insert values 0xf0..0xf7
+		!h f0f1f2f3 f4f5f6f7		; insert values 0xf0..0xf7
+		!h f0f1f2f3f4f5f6f7		; insert values 0xf0..0xf7
+		!h f0f 1f2	; ERROR: space inside pair!
+		!h 0x00, $00	; ERROR: "0x", "," and "$" are forbidden!
+		!h SOME_SYMBOL	; ERROR: symbols are forbidden!
+		!h ABCD	; insert value 0xAB, then 0xCD (CAUTION, big-endian)
+
+
 Call:		!fill AMOUNT [, VALUE]
 Purpose:	Fill amount of memory with value.
 Parameters:	AMOUNT: Any formula the value parser accepts, but it
@@ -118,6 +136,18 @@ Examples:	!fi 256, $ff	; reserve 256 bytes
 		!fill 2		; reserve two bytes
 
 
+Call:		!skip AMOUNT
+Purpose:	Advance in output buffer without starting a new
+		segment.
+Parameters:	AMOUNT: Any formula the value parser accepts, but it
+		must be solvable even in the first pass (this
+		limitation will hopefully be lifted in a future
+		release).
+Aliases:	None
+Examples:	!skip BUFSIZE	; reserve some bytes
+		!skip 5		; reserve five bytes
+
+
 Call:		!align ANDVALUE, EQUALVALUE [, FILLVALUE]
 Purpose:	Fill memory until a matching address is reached. ACME
 		outputs FILLVALUE until "program counter AND ANDVALUE"
@@ -128,7 +158,7 @@ Parameters:	ANDVALUE: Any formula the value parser accepts, but it
 		it must be solvable even in the first pass.
 		FILLVALUE: Any formula the value parser accepts. If it
 		is omitted, a default value is used (currently 234,
-		that's the 6502 CPU's NOP command).
+		that's the opcode of the 6502 CPU's NOP instruction).
 Examples:	!align 255, 0	; align to page (256 bytes)
 		!align 63, 0	; align to C64 sprite block (64 bytes)
 
@@ -351,23 +381,25 @@ Section:   Flow control
 
 Call:		!if CONDITION { BLOCK } [ else { BLOCK } ]
 Purpose:	Conditional assembly. If the given condition is true,
-		the first block of statements will be parsed;
-		if it isn't, the second block will be parsed instead
-		(if present).
+		the matching block of statements will be parsed;
+		if no condition is true, the ELSE block (if present)
+		will be parsed.
 Parameters:	CONDITION: Any formula the value parser accepts, but
 		it must be solvable even in the first pass.
 		BLOCK: A block of assembler statements.
-Examples:	!text "Black", 0	; Choose wording according to
-		!if country = uk {	; content of "country" symbol.
+Examples:	; Choose word according to "country" symbol:
+		!if country = uk {
 			!text "Grey"
+		} else if country = fr {
+			!text "Gris"
+		} else if country = de {
+			!text "Grau"
 		} else {
 			!text "Gray"
 		}
-		!byte 0
-		!text "White", 0
 
-		; Insert debug commands if symbol "debug" is not zero:
-		!if debug { lda #"z":jsr char_output }
+		; Insert debug code depending on symbol "debug":
+		!if debug { lda #'z':jsr char_output }
 
 
 Call:		!ifdef SYMBOL { BLOCK } [ else { BLOCK } ]
@@ -410,12 +442,26 @@ Examples:	; this was taken from <6502/std.a>:
 			; further instances will be skipped.
 		}
 
+		; include at most one driver source code:
+		!ifdef RAM_REU {
+			!src "driver_reu.a"
+		} else ifdef RAM_GEORAM {
+			!src "driver_georam.a"
+		} else ifdef RAM_VDCRAM {
+			!src "driver_vdcram.a"
+		} else ifdef RAM_SUPERRAM {
+			!src "driver_superram.a"
+		} else {
+			!src "driver_noram.a"
+		}
+
 
 Call:		!for SYMBOL, START, END { BLOCK }
 Purpose:	Looping assembly. The block of statements will be
 		parsed a fixed number of times, as specified by the
-		values of START and END. For a more flexible
-		possibility, have a look at "!do" below.
+		values of START and END. For more flexible
+		possibilities, have a look at "!do" and "!while"
+		below.
 Parameters:	SYMBOL: Any valid symbol name.
 		START: Any formula the value parser accepts, but it
 		must be solvable even in the first pass. SYMBOL will 
@@ -465,12 +511,13 @@ Examples:
 Miscellaneous:	The old syntax ("!for SYMBOL, END { BLOCK }" where
 		START was always implied to be 1) is still fully
 		supported, but gives a warning to get people to change
-		to the new syntax. You can disable this warning using
-		the "-Wno-old-for" switch, but then you will get
+		to the new syntax.
+		You can disable this warning using the "--dialect" or
+		the "-Wno-old-for" switches, but then you will get
 		warnings for using the *new* syntax.
-		When migrating your sources, bear in mind that it is
-		no longer possible to skip the block completely by
-		specifying a loop count of zero.
+		When migrating your sources to the current syntax,
+		bear in mind that it is no longer possible to skip the
+		block completely by specifying a loop count of zero.
 		Also note that with the new algorithm, SYMBOL has a
 		different value after the block than during the last
 		loop cycle, while the old algorithm kept that last
@@ -480,9 +527,9 @@ Miscellaneous:	The old syntax ("!for SYMBOL, END { BLOCK }" where
 Call:		!set SYMBOL = VALUE
 Purpose:	Assign given value to symbol even if the symbol
 		already has a different value. Needed for loop
-		counters when using "!do", for example. Only use this
-		opcode for something else if you're sure you *really*
-		know what you are doing... :)
+		counters when using "!do"or "!while", for example.
+		Only use this opcode for something else if you're sure
+		you *really* know what you are doing... :)
 Parameters:	SYMBOL: Any valid symbol name.
 		VALUE: Any formula the value parser accepts.
 Example:	see "!do" below
@@ -522,6 +569,34 @@ Examples:	; a loop with conditions at both start and end
 		!do until 3 = 4 {     } while 3 < 4
 
 
+Call:		!while [CONDITION] { BLOCK }
+Purpose:	Looping assembly. The block of statements can be
+		parsed several times, depending on the given
+		condition.
+		The condition is parsed in every repetition before the
+		actual block. If it isn't met when first checked, the
+		block will be skipped.
+Parameters:	CONDITION: Any formula the value parser accepts, but
+		it must be solvable even in the first pass.
+		BLOCK: A block of assembler statements.
+Examples:	; a loop with a counter
+		!set a = 0			; init loop counter
+		!while a < 6 {
+			lda #a
+			sta label + a
+			!set a = a + 1
+		}
+
+		; a loop depending on program counter
+		!while * < $c000 { nop }
+
+		; a never ending loop - this will cause an error
+		!while 3 < 4 { nop }
+
+		; an empty loop - this will hang ACME
+		!while 3 != 4 {     }
+
+
 Call:		!endoffile
 Purpose:	Stop processing the current source file. Using this
 		pseudo opcode you can add explanatory text inside your
@@ -536,22 +611,24 @@ Example:	rts	; some assembler mnemonic
 		"!eof" is reached.
 
 
-Call:		!warn STRING_VALUE [, STRING_VALUE]*
+Call:		!warn VALUE [, VALUE]*
 Purpose:	Show a warning during assembly.
-Parameters:	STRING_VALUE: Can be either a string given in double
-		quotes or any formula the value parser accepts.
-		Numbers will be output in decimal _and_ hex format.
+Parameters:	VALUE: Can be either a string given in double quotes
+		or any formula the value parser accepts.
+		Integer numbers will be output in both decimal _and_
+		hex formats.
 Example:	!if * > $a000 {
 			!warn "Program reached ROM: ", * - $a000, " bytes overlap."
 		}
 
 
-Call:		!error STRING_VALUE [, STRING_VALUE]*
+Call:		!error VALUE [, VALUE]*
 Purpose:	Generate an error during assembly (therefore, no
 		output file will be generated).
-Parameters:	STRING_VALUE: Can be either a string given in double
-		quotes or any formula the value parser accepts.
-		Numbers will be output in decimal _and_ hex format.
+Parameters:	VALUE: Can be either a string given in double quotes
+		or any formula the value parser accepts.
+		Integer numbers will be output in both decimal _and_
+		hex formats.
 Example:	rts	; end of some function
 	start	!source "colors.a"
 	end	!if end - start > 256 {
@@ -559,12 +636,13 @@ Example:	rts	; end of some function
 		}
 
 
-Call:		!serious STRING_VALUE [, STRING_VALUE]*
+Call:		!serious VALUE [, VALUE]*
 Purpose:	Generate a serious error, immediately stopping
 		assembly.
-Parameters:	STRING_VALUE: Can be either a string given in double
-		quotes or any formula the value parser accepts.
-		Numbers will be output in decimal _and_ hex format.
+Parameters:	VALUE: Can be either a string given in double quotes
+		or any formula the value parser accepts.
+		Integer numbers will be output in both decimal _and_
+		hex formats.
 Example:	!source "part1.a"	; sets part1_version
 		!source "part2.a"	; sets part2_version
 		!if part1_version != part2_version {
@@ -584,9 +662,9 @@ Parameters:	TITLE: The macro's desired name (same rules as for
 		could want this is beyond me).
 		SYMBOL: The desired name for the parameter value at
 		call time. Normally, these parameter symbols should be
-		local (first character a dot), as different macro
-		calls will almost for sure have different parameter
-		values.
+		local (first character a '.' or a '@'), as different
+		macro calls will almost for sure have different
+		parameter values.
 		If you prefix SYMBOL with a '~' character, it will be
 		called by reference, not by value: Changing the value
 		inside the macro will result in the "outer" symbol to
@@ -713,8 +791,10 @@ Purpose:	Set program counter to given value and start new
 		issued. Because some people do this overlapping
 		on purpose, the warnings can be suppressed using
 		modifier keywords.
-		Future versions of ACME may issue errors instead of
-		warnings.
+		Using the "--strict-segments" CLI switch, these
+		warnings can be turned onto errors. Future versions of
+		ACME may do that by default - so if needed, use the
+		modifier keywords.
 Parameters:	EXPRESSION: Any formula the value parser accepts, but
 		it must be solvable even in the first pass.
 		MODIFIER: "overlay" or "invisible" (without quotes):
@@ -771,6 +851,24 @@ Examples:	!to "TinyDemo", cbm	; define output file + format
 		; Useful if you want to store your code in an EPROM.
 
 
+Call:		!xor EXPRESSION [ { BLOCK } ]
+Purpose:	Change the value to XOR all output bytes with (the
+		value defaults to zero on startup). This "encryption"
+		facility was added to compensate for the shortcomings
+		of the "!scrxor" pseudo opcode, which only XORs
+		strings and characters, but not numbers.
+		When used with block syntax, the previously chosen
+		value is restored afterwards.
+Parameters:	EXPRESSION: Any formula the value parser accepts.
+		BLOCK: A block of assembler statements.
+Examples:	; first as normal screencodes:
+		!scr "Hello everybody...", GROUPLOGOCHAR
+		; and now as inverted screencodes:
+		!xor $80 {
+		!scr "Hello everybody...", GROUPLOGOCHAR
+		}
+		
+
 ----------------------------------------------------------------------
 Section:   Offset assembly
 ----------------------------------------------------------------------
@@ -802,6 +900,13 @@ Examples:	ldx #.shifted_end - .shifted_start
 		}
 	.shifted_end
 
+Miscellaneous:	If you need to convert a label or the program counter
+		from its "pseudopc" to its "real" value, you can do
+		that using the "&" operator. Given the example above,
+		the symbol ".target" will evaluate to the value $0400,
+		but "&.target" will evaluate to the same value as
+		".shifted_start" will.
+
 
 ----------------------------------------------------------------------
 Section:   CPU support pseudo opcodes (especially 65816 support)
@@ -811,23 +916,25 @@ Call:		!cpu KEYWORD [ { BLOCK } ]
 Purpose:	Select the processor to produce code for. If this PO
 		isn't used, ACME defaults to the 6502 CPU (or to the
 		one selected by the "--cpu" command line option).
-		ACME will give errors if you try to assemble commands
-		the chosen CPU does not have. You can change the
-		chosen CPU at any time. When used with block syntax,
-		the previously chosen CPU value is restored
-		afterwards.
+		ACME will give errors if you try to assemble
+		instructions the chosen CPU does not support. You can
+		change the chosen CPU at any time. When used with
+		block syntax, the previously chosen CPU value is
+		restored afterwards.
 Parameters:	KEYWORD: Currently valid keywords are:
-			6502	for the original MOS 6502
-			6510	6502 plus undocumented opcodes
-			65c02	6502 plus BRA,PHX/Y,PLX/Y,STZ,TRB/TSB
-			r65c02	65c02 plus BBRx, BBSx, RMBx, SMBx
-			w65c02	r65c02 plus STP/WAI
-			65816	65c02 plus 16/24-bit extensions
-			65ce02	r65c02 plus Z reg, long branches, ...
-			4502	65ce02 with MAP instead of AUG
-			c64dtv2	6502 plus BRA/SAC/SIR plus some of the
+		6502		for the original MOS 6502
+		nmos6502	6502 plus undocumented opcodes
+		6510		(alias for "nmos6502")
+		65c02		6502 plus BRA,PHX/Y,PLX/Y,STZ,TRB/TSB
+		r65c02		65c02 plus BBRx, BBSx, RMBx, SMBx
+		w65c02		r65c02 plus STP/WAI
+		65816		65c02 plus 16/24-bit extensions
+		65ce02		r65c02 plus Z reg, long branches, ...
+		4502		65ce02 with MAP instead of AUG
+		m65		4502 plus 32-bit extensions
+		c64dtv2		6502 plus BRA/SAC/SIR plus some of the
 				undocumented opcodes
-			See "docs/cputypes.txt" for more info.
+			See "docs/cputypes/all.txt" for more info.
 		BLOCK: A block of assembler statements.
 Examples:	!if cputype = $65c02 {
 			!cpu 65c02 {	; temporarily allow 65c02 stuff
@@ -840,7 +947,7 @@ Examples:	!if cputype = $65c02 {
 			pla
 		}
 		rts
-		!cpu 65816	; allow 65816 commands from here on
+		!cpu 65816	; now allow instructions of 65816 cpu
 
 
 Call:		!al [ { BLOCK } ]
@@ -888,6 +995,8 @@ Parameters:	BLOCK: A block of assembler statements
 		If no block is given, only the current statement will
 		be affected, which should then be an explicit symbol
 		definition.
+		To make use of this feature, you need to use the
+		"-Wtype-mismatch" CLI switch.
 Aliases:	"!addr"
 Examples:	!addr	k_chrout = $ffd2	; this is an address
 			CLEAR = 147		; but this is not
@@ -913,6 +1022,10 @@ Purpose:	Use PetSCII as the text conversion table. Now
 		superseded by the "!convtab" pseudo opcode.
 Old usage:	!cbm		; gives "use !ct pet instead" error
 Now use:	!convtab pet	; does the same without error
+		If you just want to assemble an old source code
+		without touching it, use the "--dialect" CLI switch:
+		Using "--dialect 0.94.6" or earlier will assemble this
+		pseudo opcode without throwing an error.
 
 
 Call:		!subzone [TITLE] { BLOCK }
@@ -927,6 +1040,10 @@ Old usage:	!subzone graphics {
 Now use:	!zone graphics {
 			!source "graphics.a"
 		}
+		If you just want to assemble an old source code
+		without touching it, use the "--dialect" CLI switch:
+		Using "--dialect 0.94.6" or earlier will assemble this
+		pseudo opcode without throwing an error.
 
 
 Call:		!realpc
@@ -940,3 +1057,8 @@ Old usage:	!pseudopc $0400
 Now use:	!pseudopc $0400 {
 			; imagine some code here...
 		}
+		If you just want to assemble an old source code
+		without touching it, use the "--dialect" CLI switch:
+		Using "--dialect 0.94.6" or earlier will assemble this
+		pseudo opcode without throwing an error.
+		Using "--dialect 0.85", not even a warning is thrown.

docs/Changes.txt

@@ -12,6 +12,121 @@ platform used. There should be another help file in this archive
 outlining the platform specific changes.
 
 
+----------------------------------------------------------------------
+Section:   New in release 0.97
+----------------------------------------------------------------------
+
+FINALLY strings can be assigned to symbols!
+    "anything in double quotes" is a string, while characters in
+    single quotes are, just as before, immediately converted to their
+    character code. Go and read "docs/Upgrade.txt" to learn about
+    compatibility issues.
+Added backslash escaping in all string literals:
+    \0 is a null byte, \t is a TAB, \n is a line feed, \r is a
+    carriage return, \" is a double quote, \' is a single quote and
+    \\ is a backslash. Go and read "docs/Upgrade.txt" to learn about
+    compatibility issues.
+Added "--dialect" CLI switch:
+    Because string symbols and backslash escaping introduce a few
+    incompatibilities to older versions, ACME can now be told to mimic
+    the behavior of older versions. So it's still possible to assemble
+    older sources.
+Added lists:
+    Lists can be used to pass an arbitrary number of arguments to
+    macros, or to store any number of items (including other lists) in
+    a single symbol. Example:   my_list = [1, 2, label, "string", 9]
+Added "len()" operator:
+    This returns the number of elements in a string or list.
+Added "[]" operator (for indexing):
+    This returns a single element from a string or list. Negative
+    indices are supported as well, they access the string/list from
+    the other end. Examples:   a = my_list[2]   b = my_string[-1]
+Added "&" operator:
+    This operator converts labels or the program counter from its
+    value inside a "!pseudopc" block to the value outside of that
+    block. Thanks to markusC64 for the suggestion!
+Added "!while {}" pseudo opcode.
+Added "else if", "else ifdef" and "else ifndef" possibilities.
+Added "M65" cpu:
+    This instruction set includes the MEGA65 extensions, namely 32-bit
+    pointers and 32-bit data operations using prefix bytes.
+Added "NMOS6502" as an alias for "6510" cpu.
+Improved NMOS6502/6510 mode:
+    DOP and TOP can now also be written as NOP.
+Improved 65816 mode:
+    MVN and MVP can now also be written with '#' before arguments.
+Added "--test" CLI switch:
+    This is for people who want to help test experimental features.
+Improved error messages:
+    "Garbage data at end of statement" now includes the unexpected
+    character.
+    "Symbol not defined" is only output once per symbol.
+Added warning:
+    ACME complains about binary literals with an "unusual" number of
+    digits. Thanks to groepaz for the suggestion!
+    The warning can be disabled using the "-Wno-bin-len" CLI switch.
+Added warning:
+    All mnemonics without indirect addressing now complain about
+    unneeded parentheses.
+Fix: Characters are now unsigned (had been architecture-dependent).
+Improved error handling of "--cpu" and "--format" switches.
+Added opcode table for NMOS6502 cpu to docs.
+Added some test sources.
+Added support for "hashbang" lines (if file starts with a '#'
+    character, the first line is ignored)
+Fixed some minor bugs no-one ever seems to have encountered.
+Rewritten "docs/Upgrade.txt".
+
+
+----------------------------------------------------------------------
+Section:   New in release 0.96.5
+----------------------------------------------------------------------
+
+Allowed C++-style comments via "//". Thanks to awsm for the
+    suggestion (and please accept my apologies for taking so long to
+    release this)
+Added "--ignore-zeroes" CLI switch. This disables the "leading zeroes
+    determine number size" algorithm. Thanks to groepaz for the
+    suggestion.
+Added "--strict-segments" CLI switch. This changes warnings about
+    overlapping memory segments into errors. Thanks to groepaz for the
+    suggestion.
+Added 6502 family tree to docs/cpu_types/all.txt
+
+
+----------------------------------------------------------------------
+Section:   New in release 0.96.4
+----------------------------------------------------------------------
+
+Bugfix: Removed warnings about zero page wrap-around for the 65816's
+    24-bit pointers (because wrap-around does not actually happen).
+    Thanks to Johann Klasek for reporting this.
+Added "!xor" pseudo opcode to compensate for the shortcomings of the
+    "!scrxor" pseudo opcode. Thanks to spider-j for the initial bug
+    report.
+Added "-I" CLI switch to add search paths for input files. Thanks to
+    peiselulli for the suggestion.
+
+
+----------------------------------------------------------------------
+Section:   New in release 0.96.3
+----------------------------------------------------------------------
+
+Added "!h"/"!hex" pseudo opcode: Now external source code generator
+    tools can easily put data in sources with minimal syntax overhead.
+Added "!skip" pseudo opcode: "!skip N" works like "*=*+N" without
+    starting a new segment.
+Added "cheap locals": Labels with '@' prefix have automatic scoping,
+    bounded by the preceding and the following global labels.
+Added "--fullstop" CLI switch to change pseudo opcode prefix from '!'
+    to '.' (so other assemblers' sources need less conversion work)
+Fixed a bug where expressions like "1)+1" crashed ACME. Thanks to
+    Bitbreaker for reporting this.
+Added warning when using zp-indirect addressing modes where argument
+    is $ff because pointer wraps around to $00. Thanks to Gerrit for
+    the suggestion.
+
+
 ----------------------------------------------------------------------
 Section:   New in release 0.96.2
 ----------------------------------------------------------------------
@@ -100,8 +215,9 @@ Section:   New in release 0.95.3
 
 Added "c64dtv2" cpu type so you can use its SIR, SAC and BRA opcodes;
     along with the undocumented ("illegal") opcodes of the 6510.
-Added Martin Piper's "--msvc" patch so error output can be configured
-    to be in Visual Studio format.
+Added "--msvc" CLI switch so error output can be configured to be in
+    Visual Studio format. Thanks to Martin Piper for writing this
+    patch!
 Merged third-party patch (who wrote it?) to output label dump in VICE
     format. Still needs work to be configurable about the types of
     symbols actually output.

docs/Errors.txt

@@ -38,15 +38,29 @@ Assembling buggy JMP($xxff) instruction
     location ARGUMENT + 1, but from ARGUMENT - 255. Therefore ACME
     issues this warning if you are about to generate such an
     instruction.
-    Note that this warning is only given for CPU types 6502 and 6510,
-    because 65c02 and 65816 have been fixed in this respect.
+    Note that this warning is only given for some CPU types (6502,
+    nmos6502/6510, c64dtv2) because later ones like 65c02 and 65816
+    have been fixed in this regard.
 
+Assembling unstable ANE #NONZERO instruction
 Assembling unstable LXA #NONZERO instruction
-    This warning is only ever given for CPU type 6510. LXA is one of
-    the undocumented ("illegal") opcodes of this CPU (opcode 0xab),
-    and it only works reliably if its argument is zero. Therefore ACME
-    issues this warning if you are about to generate this instruction
-    with a non-zero argument.
+    These warnings are only ever given for CPU type nmos6502 (6510).
+    ANE and LXA are undocumented ("illegal") opcodes of this CPU, and
+    they only work reliably if the argument is zero or the accumulator
+    contains 0xff.
+    Therefore ACME issues these warnings if it is about to generate
+    these instructions with a non-zero argument.
+
+Binary literal without any digits.
+Hex literal without any digits.
+    A special literal was started, but then no digits followed. Expect
+    this to become an error in future!
+
+Binary literal with strange number of digits.
+    This warning is given if the number of digits in a binary literal
+    is not a multiple of four. This is useful when you meant to write
+    %#....... but actually wrote %#........ by mistake. See? :P
+    You can disable this warning using the CLI switch "-Wno-bin-len".
 
 Bug in ACME, code follows
     A situation has been encountered implying there is a bug in ACME.
@@ -57,7 +71,7 @@ C-style "==" comparison detected.
 
 Converted to integer for binary logic operator.
     Applying binary logic to float values does not make much sense,
-    therefore floats will be converted to integer in this case.
+    therefore floats will be converted to integer in such cases.
 
 "EOR" is deprecated; use "XOR" instead.
     This means the operator, not the mnemonic.
@@ -65,7 +79,9 @@ Converted to integer for binary logic operator.
 Found old "!for" syntax.
     Please update your sources to use the new "!for" syntax. See
     AllPOs.txt for details.
-    You can suppress this warning using the "-Wno-old-for" switch.
+    You can suppress this warning using the "--dialect" or the
+    "-Wno-old-for" CLI switch.
+    ("-Wno-old-for" does _exactly_ the same as "--dialect 0.94.8")
 
 Found new "!for" syntax.
     When using the "-Wno-old-for" switch to disable the warning about
@@ -96,33 +112,35 @@ Label name starts with a shift-space character.
     warning is issued.
 
 Memory already initialised.
-    The "!initmem" command was given more than once (or in addition to
-    the "--initmem" command line option). Only use it once.
+    The "!initmem" pseudo opcode was given more than once, or in
+    addition to the "--initmem" command line option. Only use it once.
 
 Output file already chosen.
-    The "!to" command was given more than once (or in addition to the
-    "--outfile" command line option). Only use it once.
+    The "!to" pseudo opcode was given more than once, or in addition
+    to the "--outfile" command line option. Only use it once.
 
 Segment reached another one, overwriting it.
     The program counter has just reached the start of another segment.
     Because some people might want to assemble "onto" a binary file
-    that was loaded before, this warning can be switched off using
-    modifier keywords when changing the program counter via "* =".
-    Future versions of ACME might throw an error instead of a warning
-    in this case.
+    that was loaded before, this warning can be inhibited using
+    modifier keywords when changing the program counter via "*=".
+    For extra safety you can also turn this warning into an error
+    using the "--strict-segments" CLI switch. In future versions of
+    ACME this might become the default.
 
 Segment starts inside another one, overwriting it.
-    The given value in a "* =" command is located inside another
+    The given value in a "*=" assignment is located inside another
     segment. Because some people might want to assemble "onto" a
-    binary file that was loaded before, this warning can be switched
-    off using modifier keywords when changing the program counter via
-    "* =".
-    Future versions of ACME might throw an error instead of a warning
-    in this case.
+    binary file that was loaded before, this warning can be inhibited
+    using modifier keywords when changing the program counter via
+    "*=".
+    For extra safety you can also turn this warning into an error
+    using the "--strict-segments" CLI switch. In future versions of
+    ACME this might become the default.
 
 Symbol list file name already chosen.
-    The "!sl" command was given more than once (or in addition to the
-    "--symbollist" command line option). Only use it once.
+    The "!sl" pseudo opcode was given more than once, or in addition
+    to the "--symbollist" command line option. Only use it once.
 
 Used "!to" without file format indicator. Defaulting to "cbm".
     Now that "!to" can be given a file format keyword (either "plain"
@@ -130,19 +148,19 @@ Used "!to" without file format indicator. Defaulting to "cbm".
     works though.
 
 Using oversized addressing mode.
-    ACME just assembled a command using an addressing mode that was
-    larger than needed. This only happens if ACME could not work out
-    the argument's value in the first pass, therefore assuming a 16-
-    bit addressing mode. If, in a later pass, ACME finds out that the
-    argument is small enough to fit in 8 bits, then this warning is
-    shown. If you define all your zeropage symbols *before* they are
-    first used, this shouldn't happen. If you know that a specific
+    ACME just assembled an instruction using an addressing mode that
+    was larger than needed. This only happens if ACME could not work
+    out the argument's value in the first pass, therefore assuming a
+    16-bit addressing mode. If, in a later pass, ACME finds out that
+    the argument is small enough to fit in 8 bits, then this warning
+    is shown. If you define all your zeropage symbols *before* they
+    are first used, this shouldn't happen. If you know that a specific
     argument fits in 8 bits, you can force ACME to use 8 bits
-    addressing by postfixing the command with "+1". Example:
+    addressing by postfixing the mnemonic with "+1". Example:
 		lda+1 label
     ACME will then use an 8-bit addressing mode, regardless of whether
     the label is known or not. If the label value happens to be too
-    large to fit in 8 bits, ACME will show an error of course (To
+    large to fit in 8 bits, ACME will show an error of course (to
     always truncate a value to 8 bits, use the '<' operator).
     More about the postfixing method can be found in "AddrModes.txt".
 
@@ -155,6 +173,11 @@ Wrong type for loop's END value - must match type of START value.
     In "!for" loops, START and END must have the same type, which then
     gets used for the loop counter.
 
+Zeropage pointer wraps around from $ff to $00
+    A zeropage-indirect addressing mode uses $ff as the argument. The
+    6502 will then fetch the second pointer byte from $00 instead of
+    $0100, therefore this warning is issued.
+
 ...called from here.
     If warnings and/or errors are output during a macro call, messages
     with this text are added to display the call stack (because you
@@ -173,19 +196,35 @@ Section:   Errors during assembly
 "!cbm" is obsolete; use "!ct pet" instead.
     This is given when the now obsolete "!cbm" pseudo opcode is
     encountered.
+    If you want to assemble an old source code without first updating
+    it, you can use the "--dialect" CLI switch to make ACME mimic an
+    older version.
 
 "!pseudopc/!realpc" is obsolete; use "!pseudopc {}" instead.
     This is given when one of the now obsolete !pseudopc/!realpc
     pseudo opcodes is encountered.
+    If you want to assemble an old source code without first updating
+    it, you can use the "--dialect" CLI switch to make ACME mimic an
+    older version.
 
 "!subzone {}" is obsolete; use "!zone {}" instead.
     This is given when the now obsolete "!subzone" pseudo opcode is
     encountered.
+    If you want to assemble an old source code without first updating
+    it, you can use the "--dialect" CLI switch to make ACME mimic an
+    older version.
 
 !error: ...
     This is given when the pseudo opcode "!error" is executed. The
     actual message varies according to the pseudo opcode's arguments.
 
+After ELSE, expected block or IF/IFDEF/IFNDEF.
+    There is something strange after ELSE: It must be "if", "ifdef",
+    "ifndef" or an opening brace.
+
+Argument out of range.
+    You called arcsin/arccos with something not in the [-1, 1] range.
+
 Cannot open input file.
     ACME had problems opening an input file ("!bin", "!convtab" or
     "!src"). Maybe you mistyped its name.
@@ -194,35 +233,62 @@ Conversion table incomplete.
     The conversion table file is too small. It needs to be exactly 256
     bytes in size.
 
+CPU does not support this addressing mode for this mnemonic.
+    The given mnemonic cannot be combined with the given addressing
+    mode on the CPU you have chosen.
+
+CPU does not support this postfix for this mnemonic.
+    The given mnemonic cannot be combined with the addressing mode
+    indicated by the given postfix, at least not on the CPU you have
+    chosen.
+
 Division by zero.
     Guess what - you attempted to divide by zero.
 
+Expected ELSE or end-of-statement.
+    There is something after the closing brace of an IF block that is
+    not an ELSE.
+
+Expected end-of-statement after ELSE block.
+    There is something after the closing brace of an ELSE block.
+
 Exponent is negative.
     Using negative exponents only give sensible results when using
     floating point maths.
 
+Expression did not return a number.
+    An expression returned a string or a list but a number (integer or
+    float) was expected.
+
 File name quotes not found ("" or <>).
     File names have to be given in quotes. Either "" quoting for files
     located in the current directory or <> quoting for library files.
 
+Force bits can only be given to numbers.
+    You tried to give a force bit to a symbol and then assign a string
+    or list to it.
+
 Found '}' instead of end-of-file.
     ACME encountered a '}' character when it expected the file to end
     instead (because no blocks were open).
 
-Garbage data at end of statement.
-    There are still arguments when there should not be any more.
+Garbage data at end of statement (unexpected 'CHAR').
+    There are still arguments when there should not be any more. The
+    given character is the one where end-of-line was expected.
 
-Illegal combination of command and addressing mode.
-    The given command cannot be used with the given addressing mode on
-    the CPU you have chosen.
-
-Illegal combination of command and postfix.
-    The given command cannot be used with the addressing mode
-    indicated by the given postfix.
+Hex digits are not given in pairs.
+    The two digits of a hex byte are separated by another character,
+    or there is an odd number of digits.
 
 Illegal postfix.
     You used a postfix other than "+1", "+2" or "+3".
 
+Index is undefined.
+    You attempted an indexing operation with some undefined symbol.
+
+Index out of range.
+    The value for an indexing operation wasn't in the allowed range.
+
 Macro already defined.
     Macros can only be defined once. If you define a macro twice, ACME
     will help you find the definitions by giving a warning for the
@@ -239,6 +305,10 @@ Macro parameter twice.
     The same symbol name is used two (or more) times in the same macro
     parameter list.
 
+Negative size argument.
+    The size argument of "!bin" or "!skip" must be zero or positive,
+    but cannot be negative.
+
 Negative value - cannot choose addressing mode.
     Because the argument is a negative value, ACME does not know what
     addressing mode (8 bits, 16 bits, on a 65816 even 24 bits) to use.
@@ -246,10 +316,18 @@ Negative value - cannot choose addressing mode.
     your program to use positive addresses instead.
 
 No string given.
-    ACME expects a string but doesn't find it.
+    ACME expects a string but doesn't find it, or the string is empty.
 
+Number does not fit in N bits.
 Number out of range.
-    A value is too high or too low.
+    A value is too high or too low to be stored in 8/16/24 bits.
+    This can also mean the desired addressing mode is not available,
+    as in "sty $e000, x".
+
+Operation not supported: Cannot apply "OP" to "TYPE".
+Operation not supported: Cannot apply "OP" to "TYPE" and "TYPE".
+    You tried to use an operator on the wrong type(s) of argument(s),
+    like indexing a float or negating a string.
 
 Program counter is unset.
     You didn't set the program counter, so ACME didn't know where to
@@ -261,27 +339,35 @@ Quotes still open at end of line.
 Source file contains illegal character.
     Your source code file contained a null byte.
 
+String length is not 1.
+    You tried something like LDA#"X" with an illegal string length.
+
 Symbol already defined.
-    You defined a symbol that already had a different value. To change
-    a symbol's value, use the "!set" pseudo opcode.
+    You defined a symbol that already had a different type or value.
+    To change a symbol's type or value, use the "!set" pseudo opcode.
 
 Syntax error.
     Guess what - there's a syntax error.
 
 Target not in bank (0xTARGET).
     You tried to branch to an address not in the 0x0000..0xffff range.
-    Relative addressing (branch commands or PER) cannot leave the
+    Relative addressing (branch instructions or PER) cannot leave the
     current code bank of 64 KiB.
 
 Target out of range (N; M too far).
-    Branch commands use relative addressing, which only has a limited
-    range. You exceeded it. N is the attempted offset, M is the
-    difference to the limit - so if you succeed in optimizing M bytes
-    away, the code would assemble.
+    Branch instructions use relative addressing, which only has a
+    limited range. You exceeded it. N is the attempted offset, M is
+    the difference to the limit - so if you succeed in optimizing M
+    bytes away, the code would assemble.
+
+The chosen CPU uses opcode 0xXY as a prefix code, do not use this mnemonic!
+    The mnemonic is valid, but should not be used on this CPU. If you
+    know better, you can get around this error like this:
+	!cpu ANY_OTHER_CPU { PROBLEMATIC_MNEMONIC }
 
 There's more than one character.
-    You used a text string in an arithmetic expression, but the string
-    contained more than a single character.
+    You used a text string containing more than one character in a
+    situation where only a string with length one is allowed.
 
 Too late for postfix.
     You can only postfix symbols at the start, before they are used for
@@ -290,27 +376,57 @@ Too late for postfix.
 Too many '('.
     A formula ends before all parentheses were closed.
 
-Too many ')'.
-    There are more closing than opening parentheses in a formula.
+Un-pseudopc operator '&' can only be applied to labels.
+    You tried to apply the operator '&' to something that is not a
+    label. This operator only works on labels and on '*' (the program
+    counter), it cannot be used on other objects.
+
+Un-pseudopc operator '&' has no !pseudopc context.
+    You either tried to apply the operator '&' to something that is
+    not an implicitly defined label, but the result of an explicit
+    symbol assignment (like the result of a calculation).
+    Or you applied the operator to a label that was defined outside of
+    a !pseudopc block, or, more generally, the number of '&'
+    characters used was larger than the number of !pseudopc blocks
+    around the definition.
 
 Unknown encoding.
-    You used the "!convtab" command with a keyword ACME does not know.
+    You used the "!convtab" pseudo opcode with a keyword ACME does not
+    know.
+
+Unknown function.
+    You used a mathematical function ACME does not know.
 
 Unknown operator.
     You used an arithmetic/logical operator ACME does not know.
 
 Unknown output format.
-    You used the "!to" command with a keyword ACME does not know.
+    You used the "!to" pseudo opcode with a format specifier ACME does
+    not know.
 
 Unknown processor.
-    You used the "!cpu" command with a keyword ACME does not know.
+    You used the "!cpu" pseudo opcode with a cpu specifier ACME does
+    not know.
 
 Unknown pseudo opcode.
-    You have mistyped a "!" command.
+    You have mistyped the keyword after "!".
 
-Unknown "* =" segment modifier.
+Unknown "*=" segment modifier.
     You used a modifier keyword ACME does not know.
 
+Unsupported backslash sequence.
+    The character following the backslash was not one of the allowed
+    ones. Backslash escaping was added in release 0.97 of ACME.
+    If you want to assemble an old source code without first updating
+    it, you can use the "--dialect" CLI switch to make ACME mimic an
+    older version.
+
+Unterminated index spec.
+    An index was started with '[' but did not end with ']'.
+
+Unterminated list.
+    A list was started with '[' but did not end with ']'.
+
 Value not defined (SYMBOL NAME).
     A value could not be worked out. Maybe you mistyped a symbol name.
     Whether this is given as a "normal" or as a serious error depends
@@ -330,8 +446,9 @@ Found end-of-file instead of '}'.
     (because there was at least one block left open).
 
 Loop count is negative.
-    You used the "!for" command with a negative loop count (getting
-    this error is only possible when using the now deprecated syntax).
+    You used the "!for" pseudo opcode with a negative loop count
+    (getting this error is only possible when using the now deprecated
+    syntax).
 
 Macro already defined.
     Macros can only be defined once. If you define a macro twice, ACME
@@ -341,7 +458,7 @@ Macro already defined.
 
 Missing '{'.
     ACME didn't find the expected '{' character. Remember that '{'
-    characters must be given on the same line as the command they
+    characters must be given on the same line as the keyword they
     belong to.
 
 Out of memory.
@@ -359,14 +476,17 @@ Syntax error.
 
 Too deeply nested. Recursive macro calls?
     The only reason for ACME to have a limit on macro call nesting
-    at all is to find infinite recursions. Current limit is 64.
+    at all is to find infinite recursions.
+    The default limit is 64, this can be changed using the
+    "--maxdepth" CLI switch.
 
 Too deeply nested. Recursive "!source"?
     The only reason for ACME to still have a limit on "!source"
-    nesting at all is to find infinite recursions. Current limit is
-    64.
+    nesting at all is to find infinite recursions.
+    The default limit is 64, this can be changed using the
+    "--maxdepth" CLI switch.
 
-Value not yet defined.
+Value not defined.
     A value could not be worked out. Maybe you mistyped a symbol name.
     Whether this is given as a "normal" or as a serious error depends
     on the currently parsed pseudo opcode.
@@ -389,45 +509,89 @@ No output file specified (use the "-o" option or the "!to" pseudo opcode).
 Section:   Bugs in ACME
 ----------------------------------------------------------------------
 
-    The warning "Bug in ACME, code follows" is always followed by a
-    serious error message, stopping assembly. The second message
-    actually gives a hint about the bug's location in the source code.
-    If you ever get this combination of warning and serious error,
-    please send me an e-mail and tell me about it. If possible,
-    include a piece of source code that triggers it.
+The warning "Bug in ACME, code follows" is always followed by a
+serious error message, stopping assembly. The second message actually
+gives a hint about the bug's location in the source code.
+If you ever get this combination of warning and serious error, please
+send me an e-mail and tell me about it. If possible, include a piece
+of source code that triggers it.
+Please don't get this wrong - there are no known bugs. I just left
+some debugging code in place in case there is a bug I failed to notice
+during testing. In practice, this warning is not expected to be given
+at all. That's the reason why I want to be notified if it *does*
+decide to show up.
 
-    Please don't get this wrong - there are no known bugs. I just left
-    some debugging code in place in case there is a bug I failed to
-    notice during testing. In practice, this warning is not expected
-    to be given at all. That's the reason why I want to be notified if
-    it *does* decide to show up.
+The hint messages are of no real interest to the end user, but here
+they are for completeness' sake:
 
-    The hint messages are of no real interest to the end user, but here
-    they are for completeness' sake.
 
-IllegalGroupIndex
-    The mnemonic tree contains a group that I didn't add.
+ArgStackEmpty
+    There was no data for a monadic operator to work on.
+
+ArgStackNotEmpty
+    The expression parser has finished though there are still
+    arguments left to process.
+
+ExtendingListWithItself
+    There were multiple references to the same list.
+
+ForceBitZero
+    The function to handle force bits was called without a force bit.
 
 IllegalBlockTerminator
     A RAM block (macro or loop) was terminated incorrectly.
 
-IllegalOperatorHandle
-    The expression parser found an operator that does not exist.
+IllegalGroupIndex
+    The mnemonic tree contains a group that I didn't add.
+
+IllegalIfMode
+    A sanity check in the if/ifdef/ifndef/else code failed.
 
 IllegalImmediateMode
     The mnemonic tree contains invalid info about the size of immediate
     arguments.
 
-OperandStackNotEmpty
-    The expression parser has finished though there are still operands
-    left to parse.
+IllegalInputSource
+    Input is taken neither from a file nor from a RAM block.
+
+IllegalNumberTypeX
+    A number was neither INT nor FLOAT nor UNDEFINED.
+
+IllegalObjectType
+    A symbol is used that is neither number nor list nor string.
+
+IllegalOperatorId
+IllegalOperatorGroup
+    The expression parser found an operator that does not exist.
+
+IllegalSymbolNameLength
+    A sanity check on string lengths failed.
+
+NotEnoughArgs
+    There was not enough data for a dyadic operator to work on.
+
+NullTypeObject
+ObjectHasNullType
+    A symbol is used that does not have a type (number/list/string)
+    associated with it.
+
+OperatorIsNotDyadic
+OperatorIsNotMonadic
+    A function was passed the wrong type of operator.
 
 OperatorStackNotEmpty
     The expression parser has finished though there are still
-    operators left to parse.
+    operators left to process.
+
+PartialEscapeSequence
+    Buffered data ended on a backslash, which shouldn't be possible.
+
+SecondArgIsNotAnInt
+    A sanity check failed: An argument should have been converted to
+    integer but wasn't.
 
 StrangeInputMode
     The input state machine has reached a state that does not exist.
 
-StrangeParenthesis
+StrangeOperator
     The expression parser found a non-existing operator.

docs/Example.txt

@@ -30,4 +30,6 @@ Just in case you wonder:
     Please *don't* look at it. :)
 
     "trigono.o" is a simple example written to test the floating-point
-    capabilities.
+    capabilities. Because floats are prone to rounding errors, there
+    are two different "expected" outputs: these were generated on
+    different architectures, they only differ in one byte ($7f/$80).

docs/Floats.txt

@@ -17,7 +17,7 @@ with floats and returns a float. Applies to sin(), cos(), tan(),
 arcsin(), arccos(), arctan() and float(): These are always computed in
 float mode and always return floats.
 
-b) if a maths operation is useles when done with floats, it is done
+b) if a maths operation is useless when done with floats, it is done
 with integers and returns an integer. Applies to NOT, AND, OR, XOR,
 MOD, DIV, LSR, lowbyteof, highbyteof, bankbyteof and int(). These are
 always computed in integer mode and always return integers.

docs/Help.txt

@@ -6,15 +6,15 @@
 
                           - free  software -
 
-                       (C) 1998-2016 Marco Baye
+                       (C) 1998-2020 Marco Baye
 
 
 ----------------------------------------------------------------------
 Section:   Copyright
 ----------------------------------------------------------------------
 
-ACME - a crossassembler for producing 6502/6510/65c02/65816 code.
-Copyright (C) 1998-2016 Marco Baye
+ACME - a crossassembler for producing 6502/65c02/65816 code.
+Copyright (C) 1998-2020 Marco Baye
 The ACME icon was designed by Wanja "Brix" Gayk
 
 This program is free software; you can redistribute it and/or modify
@@ -39,9 +39,9 @@ Section:   Introduction
 
 ACME is a crossassembler for the 65xx range of processors. It knows
 about the standard 6502, the 65c02 and the 65816. It also supports
-the undocumented ("illegal") opcodes of the 6510 processor (a 6502-
-variant that is used in the Commodore C=64), and the extensions added
-in the C64DTV2.
+the undocumented ("illegal") opcodes of the NMOS versions of the 6502,
+like the 6510 variant that is used in the Commodore C=64, and it also
+supports extensions to the intruction set done by other parties.
 
 This text and the other files in the same directory only describe the
 basic functions independent of the platform used. There should be
@@ -61,14 +61,16 @@ The files in the docs directory and what they contain:
     Help.txt       ...is this text.
     Illegals.txt   Support for undocumented opcodes
     Lib.txt        Information about the library
-    QuickRef.txt   All the basic stuff about ACME
+    QuickRef.txt   All the basic stuff about ACME       <- START HERE!
     Source.txt     How to compile ACME
     Upgrade.txt    Incompatibilities to earlier versions
-    cputypes.txt   Instruction sets of target CPUs
+    cputypes/      Instruction sets of target CPUs
 
-IMPORTANT: If you upgrade from ACME 0.05 or earlier, don't forget to
-read the file "Upgrade.txt" - release 0.07 and all later ones are
-slightly incompatible to 0.05 and earlier.
+IMPORTANT: If you upgrade from an earlier version of ACME, don't
+forget to read the files "Changes.txt" and "Upgrade.txt". Adding new
+features can not always be done in a 100% compatible way, so newer
+versions may behave slightly differently. To solve this problem, the
+"--dialect" CLI switch can be used.
 
 If you want to start using ACME right away, read the file
 "QuickRef.txt", it contains the main help text.
@@ -79,15 +81,14 @@ Section:   What it can and does
 ----------------------------------------------------------------------
 
 ACME is a crossassembler.
-ACME can produce code for the 6502, 6510, 65c02 and 65816 processors.
+ACME can produce code for the 6502, 65c02 and 65816 processors.
 It does this *fast*.
 It can produce at most 64 KBytes of code.
 You can use global labels, local labels and anonymous labels.
 It is fast.
 You can use global and local macros.
 You can use conditional assembly.
-You can use looping assembly (There are two ways to do this; a very
-    simple and a very flexible one).
+You can use looping assembly.
 You can include other source files.
 You can include binary files (either whole or parts) directly into the
     output.
@@ -100,6 +101,7 @@ ACME's maths parser has no problems concerning parentheses and
     indirect addressing modes.
 ACME's maths parser knows a shit load of different operations.
 ACME supports both integer and floating point maths operations.
+In addition to numbers, symbols can also hold strings or lists.
 You can dump the global symbols into a file.
 ACME supports a library of commonly used macros and symbols.
 It always takes as many passes as are needed.

docs/Illegals.txt

@@ -8,7 +8,7 @@
 
 
 In release 0.87, support for some of the undocumented opcodes of the
-6502 processor was added.
+NMOS 6502 processor was added.
 In release 0.89, some more were added.
 In release 0.94.8, another one was added (lxa).
 In release 0.95.3, C64DTV2 support was added, which includes these
@@ -23,15 +23,14 @@ opcodes (mnemonics in parentheses are used by other sources):
  mnemonic       |  8  8,x 8,y 16 16,x 16,y (8,x) (8),y | performs:
 ----------------+--------------------------------------+-----------
  slo (aso)      | 07  17      0f  1f   1b   03    13   | asl + ora
- rla            | 27  37      2f  3f   3b   23    33   | rol + and
+ rla (rln)      | 27  37      2f  3f   3b   23    33   | rol + and
  sre (lse)      | 47  57      4f  5f   5b   43    53   | lsr + eor
- rra            | 67  77      6f  7f   7b   63    73   | ror + adc
+ rra (rrd)      | 67  77      6f  7f   7b   63    73   | ror + adc
  sax (axs, aax) | 87      97  8f            83         | stx + sta
  lax            | a7      b7  af       bf   a3    b3   | ldx + lda
  dcp (dcm)      | c7  d7      cf  df   db   c3    d3   | dec + cmp
  isc (isb, ins) | e7  f7      ef  ff   fb   e3    f3   | inc + sbc
  las (lar, lae) |                      bb              | A,X,S = {addr} & S
-These five are said to be unstable:
  tas (shs, xas) |                      9b              | S = A & X    {addr} = A&X& {H+1}
  sha (axa, ahx) |                      9f         93   | {addr} = A & X & {H+1}
  shx (xas, sxa) |                      9e              | {addr} = X & {H+1}
@@ -40,31 +39,37 @@ These five are said to be unstable:
                 |         addressing mode         |
  mnemonic       | implied #8    8   8,x  16  16,x | performs:
 ----------------+---------------------------------+-----------------------
- anc            |         0b*                     | A = A & arg, then C=N
- asr (alr)      |         4b                      | A = A & arg, then lsr
+ anc (ana, anb) |         0b*                     | A = A & arg, then C=N
+ alr/asr        |         4b                      | A = A & arg, then lsr
  arr            |         6b                      | A = A & arg, then ror
  sbx (axs, sax) |         cb                      | X = (A & X) - arg
  dop (nop, skb) |   80**  80   04   14            | skips next byte
  top (nop, skw) |   0c**                 0c   1c  | skips next two bytes
+ nop (skb, skw) |   ea    80   04   14   0c   1c  | see the two lines above
  jam (kil, hlt) |   02                            | crash (wait for reset)
 These two are somewhat unstable, because they involve an arbitrary value:
- ane (xaa)      |         8b***                   | A = (A | ??) & X & arg
+ ane (xaa, axm) |         8b***                   | A = (A | ??) & X & arg
  lxa (lax, atx) |         ab***                   | A,X = (A | ??) & arg
 
 Example:
-		!cpu 6510	; activate additional mnemonics...
+		!cpu nmos6502	; activate additional mnemonics...
 		lax (some_zp_label,x)	; ...and use them. No, this
 		dcp (other_zp_label),y	; example does not make sense.
 
 *) Up until ACME version 0.95.1, anc#8 generated opcode 0x2b. Since
 ACME version 0.95.2, anc#8 generates opcode 0x0b. Both opcodes work
-the same way on a real 6510 CPU, but they do not work on the C64DTV2.
+the same way on a real NMOS 6502 CPU, but they do not work on the
+C64DTV2.
+Using the "--dialect" CLI switch does not change the generated opcode!
 
 **) Note that DOP ("double nop") and TOP ("triple nop") can be used
 with implied addressing, but the generated opcodes are those for
-immediate and 16-bit absolute addressing, respectively. Using dop/top
-with x-indexed addressing might have its uses when timing is critical
-(crossing a page border adds a penalty cycle).
+immediate and 16-bit absolute addressing, respectively, This way DOP
+and TOP can be used to skip the following one- or two-byte
+instruction.
+Using DOP/TOP with x-indexed addressing might have its uses when
+timing is critical (crossing a page border adds a penalty cycle).
+Unless using implied addressing, DOP/TOP can also be written as NOP.
 
 ***) ANE and LXA first perform an ORA with an arbitrary(!) value and
 then perform an AND with the given argument. So they are unstable and
@@ -75,10 +80,8 @@ ACME will output a warning if these opcodes get assembled with a
 nonzero argument.
 
 There is no guarantee that these opcodes actually work on a given 6502
-(or 6510, or 8500, or 8502) CPU. But as far as I know, nobody ever
-found an unmodified C64/C128 where these illegals didn't work. That's
-why I used "6510" as the CPU keyword instead of "6502illegal" or
-something like that.
+(or 6510, or 8500, or 8501, or 8502) CPU. But as far as I know, nobody
+ever found an unmodified C64/C128 where these illegals didn't work.
 
 These illegals will definitely *not* work on 65c02 and 65816 CPUs. But
 I really should not have to tell you that ;)
@@ -88,7 +91,7 @@ people use different names for them. I hope my choices are not too
 exotic for your taste.
 
 Just for the sake of completeness: Here are all the remaining opcodes
-(the ones ACME won't generate even with "6510" cpu):
+(the ones ACME won't generate even with "nmos6502" cpu chosen):
 
 Opcode|  Description                                          C64DTV2
 ------+--------------------------------------------------------------
@@ -133,4 +136,10 @@ For more information about what these opcodes do, see these documents:
   Extra Instructions Of The 65XX Series CPU, Adam Vardy, 27 Sept. 1996
   6502 Undocumented Opcodes, by Freddy Offenga, 5/17/1997
   AAY64 (All About Your 64)
-  NMOS 6510 Unintended Opcodes
+
+...but the most comprehensive work is:
+
+    "No More Secrets - NMOS 6510 Unintended Opcodes"
+
+Download it from    https://csdb.dk/release/?id=185341
+or ask google for the latest version.

docs/QuickRef.txt

@@ -106,6 +106,10 @@ Then there are local symbols (their names starting with a '.'
 character). These can only be accessed from inside the macro or zone
 they were defined in (for more about macros and zones, see the file
 "AllPOs.txt").
+There are also "cheap locals": their names start with an '@'.
+The area where these can be accessed is limited automatically by the
+previous and the following global label (cheap locals are "cheap"
+because you don't have to put in any extra work to limit their range).
 And then there are anonymous labels (their names being sequences of
 either '-' or '+' characters). They are also local (bound to their
 macro/zone), but in addition to that, the "-" labels can only be used
@@ -113,6 +117,7 @@ for backward references, while the "+" labels can only be used for
 forward references.
 In contrast to global and local labels, anonymous labels can not be
 defined explicitly (as in SYMBOL = VALUE).
+Each macro call automatically gets its own scope for local symbols.
 
 Save the given example source code to a file called "tiny.a" and start
 acme by typing
@@ -149,7 +154,7 @@ found in the file "AllPOs.txt". Here's just a short overview:
 !convtab   !pet   !raw   !scr   !scrxor   !text
 ...for converting and outputting strings.
 
-!do   !endoffile   !for   !if   !ifdef   !ifndef   !set
+!do   !endoffile   !for   !if   !ifdef   !ifndef   !set   !while
 ...for flow control; looping assembly and conditional assembly.
 
 !binary   !source   !to
@@ -170,7 +175,7 @@ found in the file "AllPOs.txt". Here's just a short overview:
 !warn   !error   !serious
 ...for generating warnings, errors and serious errors.
 
-!addr
+!address
 ...to mark symbols as addresses, for the optional type check system.
 
 
@@ -187,33 +192,57 @@ Available options are:
         This is more or less useless, because the help is also shown
         if ACME is run without any arguments at all.
 
-    -f, --format FORMAT    set output file format ("plain", "cbm" or "apple")
+    -f, --format FORMAT    set output file format
+        Use this with a bogus format type to get a list of all
+        supported ones (as of writing: "plain", "cbm" and "apple")
     -o, --outfile FILE     set output file name
         Output file name and format can also be given using the "!to"
         pseudo opcode. If the format is not specified, "!to" defaults
         to "cbm", while the command line option defaults to "plain".
 
+    -r, --report           set report file name
+        This creates a text listing containing the original line
+        number, the resulting memory address, the byte value(s) put
+        there and the original text line from the source file.
+
     -l, --symbollist FILE  set symbol list file name
         This can also be given using the "!symbollist"/"!sl" pseudo
         opcode. The switch was called "--labeldump" in older versions,
         that name still works, too.
 
-    --setpc NUMBER         set program counter
-        This can also be given in the source code using "* = NUMBER".
+    --vicelabels FILE      set file name for label dump in VICE format
+        The resulting file uses a format suited for the VICE emulator.
 
-    --cpu CPU_TYPE         set processor type
+    --setpc VALUE          set program counter
+        This can also be given in the source code using "* = VALUE".
+
+    --cpu CPU_TYPE         set target processor
         This can be changed in the source code using the "!cpu" pseudo
         opcode. Defaults to 6502.
+        Use this with a bogus cpu type to get a list of all supported
+        ones.
 
-    --initmem NUMBER       define 'empty' memory
+    --initmem VALUE        define 'empty' memory
         This can also be given using the "!initmem" pseudo opcode.
         Defaults to zero.
 
     --maxerrors NUMBER     set number of errors before exiting
         If not given, defaults to 10.
 
-    --maxdepth NUMBER      set recursion depth for macro calls and the
-        "!source" pseudo opcode. If not given, defaults to 64.
+    --maxdepth NUMBER      set recursion depth for macro calls and !src
+        The default value for this is 64.
+
+    --ignore-zeroes        do not determine number size by leading zeroes
+        Normally, using leading zeroes forces ACME to generate
+        oversized addressing modes, like 3-byte absolute instructions
+        instead of 2-byte zero page instructions.
+        Using this CLI switch disables this behavior.
+
+    --strict-segments      turn segment overlap warnings into errors
+        When changing the program counter, segment overlap warnings may
+        be generated. Using this CLI switch turns those warnings into
+        errors (which is recommended).
+        This strict behavior may become the default in future releases!
 
     -vDIGIT                set verbosity level
         Sets how much additional informational output is generated.
@@ -239,6 +268,12 @@ Available options are:
         "-DSYSTEM=128" could build the C128 version of the software
         (using conditional assembly in your source code file).
 
+    -I PATH/TO/DIR         add search path for input files
+        This option allows to add a directory to the search list for
+        input files. If an input file cannot be found in the current
+        working directory, all directories in the search list are
+        tried (the first match is used).
+
     -W                     fine-tune amount and type of warnings
         -Wno-label-indent
             Disables warnings about labels not being in the leftmost
@@ -246,6 +281,11 @@ Available options are:
         -Wno-old-for
             Disables warnings about the old "!for" syntax and at the
             same time enables warnings about the _new_ "!for" syntax.
+            Internally, this does exactly the same as what happens
+            when the "--dialect 0.94.8" CLI switch is used...
+        -Wno-bin-len
+            Do not complain about unusual number of digits in a binary
+            literal.
         -Wtype-mismatch
             Enables type checking system (warns about wrong types).
 
@@ -253,6 +293,28 @@ Available options are:
         With this option, errors are written to the standard output
         stream instead of to the standard error stream.
 
+    --msvc                 output errors in MS VS format
+        This changes the format of the error output to that used by
+        a certain commercial IDE.
+
+    --color                uses ANSI color codes for error output
+        If your terminal emulation supports ANSI escape codes, use
+        this option to have warnings and errors displayed in color.
+
+    --fullstop             use '.' as pseudo opcode prefix
+        This changes the prefix character used to mark pseudo opcodes
+        from '!' to '.' (so sources intended for other assemblers can
+        be converted with less effort).
+
+    --dialect VERSION      behave like different version
+        This CLI switch tells ACME to mimic the behavior of an older
+        version. Use this with a bogus version to get a list of all
+        supported ones.
+
+    --test                 enable experimental features
+        This is for people who want to help test new features before
+        they are officially announced.
+
     -V, --version          show version and exit.
 
 Platform-specific versions of ACME might offer more options.
@@ -261,61 +323,119 @@ given on the command line.
 
 
 ----------------------------------------------------------------------
-Section:   The maths parser
+Section:   The expression parser
 ----------------------------------------------------------------------
 
 ACME has a relatively powerful maths parser. This parser is used
-whenever ACME expects to read a numerical value. Supported operations
-include addition, subtraction, multiplication, divisions, comparisons,
-shifts, negation, boolean operations and some assembler-specific stuff
-like extracting the "low byte", the "high byte" or the "bank byte"
-of a value.
-Calculations are done using either signed 32-bit integer arithmetic or
-floating point arithmetic using the C "double" data type. Symbol
-values are stored the same way.
+whenever ACME expects to read a value. Supported operations include
+addition, subtraction, multiplication, divisions, comparisons, shifts,
+negation, boolean operations and some assembler-specific stuff like
+extracting the "low byte", the "high byte" or the "bank byte" of a
+value.
+Calculations are done using either signed (at least 32-bit) integer
+arithmetic or floating point arithmetic using the C "double" data
+type. Symbol values are stored the same way.
+
+
+This is a list of the value formats currently known by ACME:
+
+Examples        Notes
+---------------------------------------------------------------------
+128             a decimal value, integer
+128.5           a decimal value, floating point
+$d011           hexadecimal values are indicated by either a
+0xffd2              leading "$" or a leading "0x".
+&1701           an octal value, indicated by "&"
+%1010           binary values are indicated by either a leading "%"
+%....#...           or a leading "0b". In binary values, you can
+0b01100110          substitute the characters "0" and "1" by "." and
+                    "#" respectively. This way the values are much
+                    more readable, especially when building bitmapped
+                    objects (like C64 sprites or fonts) in your source
+                    code.
+'p'             single characters in single quotes are converted to
+                    their character code. The actual numeric value
+                    depends on the current conversion table chosen
+                    using the "!ct" pseudo opcode.
+"player 2"      double quotes indicate text strings. See below for
+                    more information on single vs. double quotes.
+[2, 3, 5, 7]    brackets indicate lists. These are useful to group
+[0, [x, y], 9]      data, for example when passing an arbitrary number
+                    of arguments to a macro.
+poll_joy2       a global symbol
+.fail           a local symbol, indicated by leading "."
+@loop           a "cheap local", indicated by leading "@"
+*               the current program counter. During offset assembly,
+                    "*" gives the value of the "Pseudo PC". Just to
+                    make sure: The value of the program counter is
+                    always the value that was valid at the start of
+                    the current statement, so
+                        !word *, *, *, *
+                    will give the same value four times. I think most
+                    assemblers do it this way.
+
+In older versions of ACME, 'x' and "x" were the same thing, namely the
+character code of the letter x using the currently selected encoding
+table.
+Since release 0.97, anything in single quotes gives the character code
+(as before), while anything in double quotes is treated as a string
+object. To be compatible to those older versions, ACME keeps accepting
+one-char strings in a lot of places where actually single characters
+are expected.
+
 
 This is a list of the operators currently known by ACME:
 
- Priority    Example      Meaning                   Alias
-------------------------------------------------------------
-       14     sin(v)      Trigonometric sine function
-       14     cos(v)      Trigonometric cosine function
-       14     tan(v)      Trigonometric tangent function
-       14     arcsin(v)   Inverse of sin()
-       14     arccos(v)   Inverse of cos()
-       14     arctan(v)   Inverse of tan()
-       14     address(v)  Mark as address           addr(v)
-       14     int(v)      Convert to integer
-       14     float(v)    Convert to float
-       13        !  v     Complement of             NOT
-       12     v  ^  w     To the power of
-       11        -  v     Negate
-       10     v  *  w     Multiply
-       10     v  /  w     Divide
-       10     v DIV w     Integer-Divide
-       10     v  %  w     Remainder of DIV          MOD
-        9     v  +  w     Add
-        9     v  -  w     Subtract
-        8     v <<  w     Shift left                ASL, LSL
-        8     v >>  w     Arithmetic shift right    ASR
-        8     v >>> w     Logical shift right       LSR
-        7        <  v     Lowbyte of
-        7        >  v     Highbyte of
-        7        ^  v     Bankbyte of
-        6     v <=  w     Lower or equal
-        6     v  <  w     Lower than
-        6     v >=  w     Higher or equal
-        6     v  >  w     Higher than
-        5     v !=  w     Not equal                 <>,  ><
-        4     v  =  w     Equal
-        3     v  &  w     Bit-wise AND              AND
-        2                 Bit-wise exclusive OR     XOR
-        1     v  |  w     Bit-wise OR               OR
+Priority Example      Meaning                       Alias         Note
+----------------------------------------------------------------------
+ 16      is_number(v) these three functions return 1               *3
+ 16      is_list(v)       if v is the correct symbol               *3
+ 16      is_string(v)         type and 0 otherwise                 *3
+ 16      len(v)       length of list or string                     *2
+ 16      sin(v)       trigonometric sine function
+ 16      cos(v)       trigonometric cosine function
+ 16      tan(v)       trigonometric tangent function
+ 16      arcsin(v)    inverse of sin()
+ 16      arccos(v)    inverse of cos()
+ 16      arctan(v)    inverse of tan()
+ 16      address(v)   mark as address               addr(v)
+ 16      int(v)       convert to integer
+ 16      float(v)     convert to float
+ 15      &symbol      "unpseudopc" symbol (see docs on "!pseudopc")
+ 14       v[w]        access v with index w                        *2
+ 13         !  v      bit-wise complement           NOT
+ 12      v  ^  w      to the power of
+ 11         -  v      negate
+ 10      v  *  w      multiply
+ 10      v  /  w      divide
+ 10      v DIV w      integer divide
+ 10      v  %  w      remainder of DIV              MOD
+  9      v  +  w      add                                          *3
+  9      v  -  w      subtract
+  8      v <<  w      shift left                    ASL, LSL
+  8      v >>  w      arithmetic shift right        ASR
+  8      v >>> w      logical shift right           LSR
+  7         <  v      low byte of
+  7         >  v      high byte of
+  7         ^  v      bank byte of
+  6      v <=  w      lower or equal
+  6      v  <  w      lower than
+  6      v >=  w      higher or equal
+  6      v  >  w      higher than
+  5      v !=  w      not equal                     <>,  ><        *3
+  4      v  =  w      equal                                        *3
+  3      v  &  w      bit-wise AND                  AND
+  2                   bit-wise exclusive OR         XOR
+  1      v  |  w      bit-wise OR                   OR
+
+Notes:
+"*2" means this operator only works on lists and strings.
+"*3" means this operator works on all three data types (numbers, lists
+and strings).
+All other operators only work on numbers.
 
 Operations with higher priority are done first. Of course you can
-change this using parentheses. If you prefer the aliases over the
-shorthand characters, note that they must be written in capital
-letters.
+change this using parentheses.
 Note that though there are operators to extract the "low byte", the
 "high byte" and the "bank byte", there is no operator to extract the
 fourth byte. If you want to access that, shift it down using ">>>" or
@@ -332,41 +452,8 @@ Calculating 0^0 (zero to the power of zero) will give 1. If
 you don't know why I'm telling you this, ask a mathematician. :)
 
 
-This is a list of the value formats currently known by ACME:
-
-Examples        Notes
----------------------------------------------------------------------
-128             a decimal value, integer
-128.5           a decimal value, floating point
-$d011           hexadecimal values are indicated by either a
-0xffd2              leading "$" or a leading "0x".
-&1701           an octal value, indicated by "&"
-%010010         binary values are indicated by either a leading "%"
-%....#...           or a leading "0b". In binary values, you can
-0b01100110          substitute the characters "0" and "1" by "." and
-                    "#" respectively. This way the values are much
-                    more readable, especially when building bitmapped
-                    objects (like C64 sprites or fonts) in your source
-                    code.
-"p"             character values are indicated by double or single
-'q'                 quotes. The actual numeric value depends on the
-                    current conversion table (none/petscii/screen),
-                    chosen using the "!ct" pseudo opcode.
-poll_joy2       a global symbol
-.fail           a local symbol, indicated by leading dot
-*               the current program counter. During offset assembly,
-                    "*" gives the value of the "Pseudo PC". Just to
-                    make sure: The value of the program counter is
-                    always the value that was valid at the start of
-                    the current statement, so
-                        !word *, *, *, *
-                    will give the same value four times. I think most
-                    assemblers do it this way.
-
-
-
 ----------------------------------------------------------------------
-Section:   Almost, but not quite, entirely useless syntax
+Section:   Almost, but not quite, complete syntax
 ----------------------------------------------------------------------
 
 Every ACME source code file consists of a non-negative number of
@@ -376,7 +463,7 @@ or CRLF characters.
 Every line consists of a non-negative number of "statements" and an
 optional comment. Statements have to be separated from each other
 using colon (":") characters, the comment has to be prefixed with a
-semicolon (";") character.
+semicolon (";") character or two slashes ("//").
 
 Every statement consists of an optional "label" and an optional
 "command". These are separated from each other using any number of
@@ -386,15 +473,19 @@ issued (to spot typing errors - see Errors.txt for more info).
 Every symbol name consists of these characters: "a" to "z", "A" to
 "Z", "0" to "9", the underscore character "_" and all characters with
 values beyond 127. The first character must not be a digit though. But
-it can be a dot ("."), making the symbol a local one. Two other
-possibilities for label names are "all-characters-are-minus" (then it
-is an anonymous backward label) and "all-characters-are-plus" (then it
-is an anonymous forward label).
+it can be '.' or '@', making the symbol a local one.
+Local symbols beginning with '.' are only valid inside the current
+zone (marked using the "!zone" pseudo opcode) or the current macro.
+Local symbols beginning with '@' are only valid between the enclosing
+global labels (or inside the current macro).
+Two other possibilities for label names are "all-characters-are-minus"
+(then it is an anonymous backward label) and "all-characters-are-plus"
+(then it is an anonymous forward label).
 
 Every command is one of the following:
-    An assembler opcode
+    An assembler mnemonic with an optional argument
     A pseudo opcode, beginning with a "!" character
-    A symbol definition (symbol=value)
+    An explicit symbol definition (SYMBOL = VALUE)
     A pc definition, beginning with a "*" character
     A macro call, beginning with a "+" character
 ...and the syntax of those things varies. :)

docs/Upgrade.txt

@@ -4,116 +4,197 @@
 
          ...the ACME Crossassembler for Multiple Environments
 
-                    --- compatibility problems ---
+                --- upgrading from earlier versions ---
 
 
 If you haven't used ACME before, you don't need to read this text.
-It is only of use to people who upgraded from ACME 0.05 (or earlier)
-to ACME 0.07 (or later).
 
-You might encounter some slight incompatibilities: I have done a few
-changes to ACME's workings.
-Because backwards compatibility is the root of all evil (*g*), I did
-not include any possibility to enforce the old behaviour. If one of
-the following changes applies to your source files, assemble them with
-this new release of ACME and then compare new and old output files.
-
-Sorry for this inconvenience, but at least I think that there won't be
-any further changes in the future.
 
 
 ----------------------------------------------------------------------
-Section:   Offset assembly / segment assembly
+Upgrading from earlier releases to ACME release 0.97
 ----------------------------------------------------------------------
 
+a) Single quotes vs. double quotes:
+Since "anything in double quotes" is now considered to be a string,
+problems can arise when trying to do a calculation with a character
+code. Here are some examples:
+	lda #' '	; loads 32 like before (ASCII code of space)
+	lda #' ' + 1	; loads 33 like before (32 plus one)
+	lda #" "	; loads 32 like before (ASCII code of space)
+	lda #" " + 1	; used to load 33, now fails with error!
+The third example still works, because 1-char-strings are treated
+just like single characters when returned by the expression parser as
+an argument for a mnemonic.
+However, the fourth example now fails because the expression parser
+tries to add a string to an integer, which is an undefined operation.
+Some examples for a related problem:
+	a = ' '		; a is 32 (ASCII code of space)
+	b = ' ' + 1	; b is 33 (32 plus one)
+	c = "!"		; c used to be 33, now it's a 1-char string
+	d = "!" + "!"	; d used to be 66, now it's a 2-char string
+If you do not get any errors when compiling your old sources, you do
+not need to worry about this problem.
+If you _do_ get errors, just use single quotes instead of double
+quotes. If you had to use double quotes because the quoted character
+itself is a single quote, write '\'' instead (backslash escaping, see
+below).
+
+b) Backslash escaping:
+Backslashes in single or double quotes are now used as escape
+characters. You need to replace any backslash in older sources with a
+sequence of two backslashes, so "some\string" becomes "some\\string",
+and a single character '\' becomes '\\'.
+If you have used backslashes as directory separators in path names (in
+Windows/DOS environments), these also need changing - but instead of
+using a double backslash, just use a single forward slash ('/')
+instead. This has the added benefit of making the sources platform-
+independent (*and* it's compatible to older ACME releases as well).
+
+c) Character values are now unsigned:
+When parsing a character in single quotes, ACME returns its character
+code, according to the chosen encoding (raw/petscii/screencode). If
+this resulted in a byte with its most significant bit set, the actual
+number was architecture-dependent. Here's an example:
+		!ct pet	; choose PetSCII encoding
+		x = 'A'	; PetSCII 'A' is 0xc1, so MSB is set
+Now x was either -63 or +193, depending on the host cpu architecture.
+Since release 0.97, this example will give 193 on all architectures.
+In most cases, this is not a problem, because the actual bit pattern
+of the lower eight bits is the same. But if you have written any code
+where the numerical value of a PetSCII character is used for 
+computations _in_the_source_code_, please check those computations.
+
+Use the "--dialect 0.94.12" CLI switch to get the old behavior
+concerning a) double quotes and b) backslashes. There is no way to get
+the old behavior concerning c) character values, because, as explained
+above, the old behavior was architecture-dependent, which is a bad
+idea(tm).
+
+
+
+----------------------------------------------------------------------
+Upgrading from earlier releases to ACME release 0.95.2
+----------------------------------------------------------------------
+
+In 6510 mode, ACME now outputs 0x0b instead of 0x2b when assembling
+the undocumented ("illegal") ANC #imm8 instruction. Both opcodes do
+the same thing, this was only changed because all other mnemonics use
+the lowest-numbered possible opcode as well.
+Forcing the old behavior via the "--dialect" switch is not supported.
+
+
+
+----------------------------------------------------------------------
+Upgrading from earlier releases to ACME release 0.94.12
+----------------------------------------------------------------------
+
+The pseudo opcode "!for" has a new syntax. The old syntax still works,
+but gives a warning.
+You can use the "--dialect 0.94.8" CLI switch to get the old behavior.
+
+
+----------------------------------------------------------------------
+Upgrading from earlier releases to ACME release 0.94.8
+----------------------------------------------------------------------
+
+The pseudo opcodes "!cbm", "!subzone" and "!realpc" no longer give
+warnings, but have now been disabled.
+You can use the "--dialect 0.94.6" CLI switch to get the old behavior.
+
+
+
+----------------------------------------------------------------------
+Upgrading from earlier releases to ACME release 0.94.6
+----------------------------------------------------------------------
+
+The "to-the-power-of" operator ('^') is now right-associative, so
+    b^c^d    will now give    b^(c^d)    instead of    (b^c)^d
+If you have never used the operator in this way, you don't need to
+worry about it.
+You can use the "--dialect 0.86" CLI switch to get the old behavior.
+
+
+
+----------------------------------------------------------------------
+Upgrading from earlier releases to ACME release 0.89
+----------------------------------------------------------------------
+
+The "logical shift right" operator has been changed. Note: This is
+about ACME's expression parser and has nothing to do with the 6502
+mnemonic called "LSR".
+Older versions were supposed to work like this:
+	a = b >> c	; alias "LSR", do a logical shift right
+But what they actually did depended on the compiler that was used to
+create the ACME binary: many binaries did an "arithmetic shift right"
+instead. This has now been fixed and changed to:
+	a = b >> c	; alias "ASR", do an arithmetic shift right
+	a = b >>> c	; alias "LSR", do a logical shift right
+If you have never applied the old ">>"/"LSR" operator to a negative
+value, you do not need to worry about this. If you have, please check
+what you expected to happen in those instances (arithmetic or logical
+shift) and update your source codes accordingly (use either ">>"/"ASR"
+or ">>>"/"LSR").
+Forcing the old behavior via the "--dialect" switch is not possible,
+because as explained above, the old behavior was compiler-dependent
+anyway.
+
+
+
+----------------------------------------------------------------------
+Upgrading from earlier releases to ACME release 0.07
+----------------------------------------------------------------------
+
+Re-defining the program counter via "* = NEW_VALUE" no longer starts
+offset assembly. Instead, ACME will change its pointer into the output
+buffer to the given value, so you can write your code in distinct
+segments. These segments can be given in any order. After assembly,
+ACME stores everything from the lowest address used to the highest
+address used. Have a look at "AllPOs.txt" for an example on how to use
+this facility.
+
 Offset assembly is now done using a new pseudo opcode called
 "!pseudopc". Have a look at "AllPOs.txt" for further information on
 its syntax and usage.
-The old way of just redefining the program counter by using more than
-one "* = EXPRESSION" statements does something totally different now:
-Whenever the program counter is redefined, ACME will actually change
-its pointer into the output buffer, so you can write your code in
-distinct segments. These segments can be given in any order. After
-assembly, ACME stores everything from the lowest address used to the
-highest address used. Have a look at "AllPOs.txt" for an example on
-how to use this facility.
 
+The pseudo opcode "!end" has been removed. Use "!eof" instead.
 
-----------------------------------------------------------------------
-Section:   Argument order of MVP/MVN
-----------------------------------------------------------------------
+The mnemonic BIT can no longer be assembled without any argument. If
+you want to insert the opcode only to mask the next instruction, use
+	!src <6502/std.a>
+to get the definitions for these two macros:
+	+bit8	; output $24 to mask following 1-byte instruction
+	+bit16	; output $2c to mask following 2-byte instruction
 
-The syntax of the 65816 opcodes MVN and MVP is usually given as
+When using the 65816 cpu, ACME now uses the correct argument order for
+the MVN and MVP mnemonics, which is:
+	mnemonic source_bank, destination_bank
 
-    MVN source_bank, destination_bank
-
-All previous versions of ACME did it the other way round: First the
-destination bank, then the source bank. This has been fixed, ACME now
-uses the syntax given above.
-
-
-----------------------------------------------------------------------
-Section:   Typecast
-----------------------------------------------------------------------
-
-You can use leading zeros to make ACME use a bigger addressing mode
-than needed. Until now, this did not work when using labels. The
-source code
-
-	label1 = $fa
+Using leading zeroes in hexadecimal or binary values makes ACME use
+bigger addressing modes than needed. This has now been extended to
+symbols as well:
 	label2 = $00fa
-
-		lda $fa
-		lda $00fa
-		lda label1
 		lda label2
+will be assembled to:
+    ad fa 00    lda $00fa
 
-was assembled to:
+Forcing the old behavior via the "--dialect" switch is not supported.
 
-		lda $fa
-		lda $00fa
-		lda $fa
-		lda $fa
-
-Release 0.07 of ACME now correctly assembles the given source code to:
-
-		lda $fa
-		lda $00fa
-		lda $fa
-		lda $00fa
 
 
 ----------------------------------------------------------------------
-Section:   !endoffile
+Upgrading from earlier releases to ACME release 0.04 beta
 ----------------------------------------------------------------------
 
-Previous versions of ACME knew a pseudo opcode called "!end" that
-marks the end of a source code file. Because the word "end" doesn't
-actually specify *what* is about to end, I changed this to
-"!endoffile". You can also use a short version, called "!eof". The old
-PO "!end" no longer works.
+The pseudo opcode "!module" has been removed. Use "!zone" instead.
+Forcing the old behavior via the "--dialect" switch is not supported.
+
 
 
 ----------------------------------------------------------------------
-Section:   Using the BIT command without parameters
+Upgrading from earlier releases to ACME release 0.03 beta
 ----------------------------------------------------------------------
 
-Release 0.07 of ACME will complain if you try to assemble BIT without
-any parameter. Previous versions did just output the byte $2c - a
-commonly known trick to mask the following 2-byte command on the 6502
-processor. If you still want to do this, use
+It is no longer possible to have more than one label in a single line.
+Forcing the old behavior via the "--dialect" switch is not supported.
 
-		!src <6502/std.a>	; parse library file
-
-to include some standard macros. Then you can use
-
-		+bit8	; output $24 to mask following 1-byte command
-
-and
-
-		+bit16	; output $2c to mask following 2-byte command
-
-respectively.
-
-
-That's all. Again, sorry for the inconvenience...

docs/cputypes/all.txt

@@ -7,32 +7,69 @@
                           --- cpu types ---
 
 
-ACME supports the following cpu types:
+ACME supports the following cpu types (shown here as a sort of family
+tree):
+
+ 6502                   standard
+ |
+ |\_nmos6502 (=6510)    + undocumented opcodes
+ |
+ |\_c64dtv2             + BRA/SAC/SIR and some (not all!) undocumented
+ |
+  \_65c02               + BRA/PHX/PHY/PLX/PLY/STZ/TRB/TSB/...
+    |
+    |\_65816            16 bit regs, 24 bit address space, ...
+    |
+     \_r65c02           + bit manipulation instructions
+       |
+       |\_w65c02        + STP/WAI
+       |
+        \_65ce02        + Z reg, long branches, ...
+          |
+           \_4502       + MAP/EOM
+             |
+              \_m65     + 32-bit pointers, 32-bit 'Q' register
 
 
 
+	!cpu 6502
 
-***   6502
-
-This is the instruction set of the original NMOS 6502 designed by MOS
-(later CSG).
+This is the official instruction set of the original NMOS 6502 CPU
+designed by MOS (later CSG).
 There are 151 documented opcodes.
 ACME does not use "A" to indicate "accumulator addressing"; just write
 the mnemonic without any argument: "LSR" will work, "LSR A" won't.
 
 
 
-***   6510
+	!cpu nmos6502
 
-This is the 6502 variant used in the C64 computer. It uses the same
-instruction set as the 6502, but in addition to that, ACME supports
-most of the undocumented opcodes as well.
+This instruction set includes the undocumented ("illegal") opcodes of
+the NMOS 6502.
 See "docs/Illegals.txt" for more info.
 
 
 
+	!cpu 6510
 
-***   65c02
+This is an alias for "nmos6502", because the 6510 cpu (as used in the
+C64 computer) is a variant of this type.
+
+
+
+	!cpu c64dtv2
+
+This is the cpu in version 2 of the C64DTV. It uses a superset of the
+6502 instruction set. Features:
+- new instructions:
+	BRA near_target		branch always
+	SAC #$12		set accumulator mapping
+	SIR #$12		set index register mapping
+- support for some (but not all!) of the undocumented opcodes.
+
+
+
+	!cpu 65c02
 
 This is the CMOS re-design of the 6502. It seems to have also been
 available from Rockwell, GTE/CMD and others. Features:
@@ -63,33 +100,7 @@ There are 178 documented opcodes.
 
 
 
-***   r65c02
-
-This is a superset of 65c02, probably originally by Rockwell. It adds
-bit manipulation instructions:
-	BBR0 $12, near_target	branch on bit reset in zp
-	BBS0 $12, near_target	branch on bit set in zp
-	RMB0 $12		reset memory bit in zp
-	SMB0 $12		set memory bit in zp
-The digit in the mnemonic is the bit number, therefore it must be in
-the 0..7 range.
-Chips with this instruction set seem to have been available from
-Rockwell, GTE/CMD and others.
-There are 210 documented opcodes.
-
-
-
-***   w65c02
-
-This is a superset of r65c02, originating at WDC. It adds two new
-instructions:
-	STP			stop (wait for reset)
-	WAI			wait for interrupt
-There are 212 documented opcodes.
-
-
-
-***   65816
+	!cpu 65816
 
 This is a superset of 65c02, originally designed by WDC (it seems to
 have been available from GTE/CMD as well). Features:
@@ -103,7 +114,33 @@ See "docs/65816.txt" for more info.
 
 
 
-***   65ce02
+	!cpu r65c02
+
+This is a superset of 65c02, probably originally by Rockwell. It adds
+bit manipulation instructions:
+	BBR4 $12, near_target	branch on bit reset in zp
+	BBS5 $12, near_target	branch on bit set in zp
+	RMB6 $12		reset memory bit in zp
+	SMB7 $12		set memory bit in zp
+The digit in the mnemonic is the bit number, therefore it must be in
+the 0..7 range.
+Chips with this instruction set seem to have been available from
+Rockwell, GTE/CMD and others.
+There are 210 documented opcodes.
+
+
+
+	!cpu w65c02
+
+This is a superset of r65c02, originating at WDC. It adds two new
+instructions:
+	STP			stop (wait for reset)
+	WAI			wait for interrupt
+There are 212 documented opcodes.
+
+
+
+	!cpu 65ce02
 
 This is a superset of r65c02, originating at CSG. Features:
 - Z register
@@ -123,7 +160,7 @@ unconditional") instead. ACME accepts both mnemonics.
 
 
 
-***   4502
+	!cpu 4502
 
 This is basically the same as 65ce02, but
 - MAP replaces AUG
@@ -133,12 +170,55 @@ There are 256 documented opcodes.
 
 
 
-***   c64dtv2
+	!cpu m65
 
-This is the cpu in version 2 of the C64DTV. It uses a superset of the
-6502 instruction set. Features:
-- new instructions:
-	BRA near_target		branch always
-	SAC #$12		set accumulator mapping
-	SIR #$12		set index register mapping
-- support for some of the undocumented opcodes.
+This is a superset of 4502 specified by the MEGA65 project. It uses
+NEG:NEG and NOP as prefix bytes to extend the instruction set.
+Features:
+- "quad mode" (32-bit data operations on virtual register 'Q')
+- "long mode" (32-bit pointer addressing for existing mnemonics)
+- "quad" and "long" modes can be combined
+quad mode introduces several new mnemonics:
+	LDQ/STQ/CPQ		like LDA/STA/CMP
+	ADCQ/SBCQ		like ADC/SBC
+	ANDQ/EORQ/ORQ		like AND/EOR/ORA
+	ASLQ/LSRQ/ROLQ/RORQ	like ASL/LSR/ROL/ROR
+	INQ/DEQ			like INC/DEC
+	BITQ			like BIT
+	ASRQ			like ASR
+  The new mnemonics support most of the addressing modes of the
+  original mnemonics with these exceptions:
+	- there is no immediate addressing
+	- indirect-Z-indexed addressing becomes indirect addressing
+	- all other indexed addressing modes can only really be used
+	with read-modify-write instructions or LDQ, because otherwise
+	a part of the 'Q' value would be used as the index.
+  CAUTION: The STQ instruction clobbers the N and Z flags!
+  There is no "real" Q register, instead A/X/Y/Z are combined to form
+  the Q register (A holds lsb, Z holds msb), except for read-modify-
+  write instructions, where the 32-bit operation is performed without
+  using A/X/Y/Z.
+  To load a 32-bit immediate constant into the Q register, use the
+  +movq macro from the <m65/std.a> library file.
+long mode brings a single new addressing mode for eight mnemonics:
+	LDA [$12], z		contents of $12/$13/$14/$15
+	STA [$12], z		plus z form the address
+	CMP [$12], z
+	ADC [$12], z
+	SBC [$12], z
+	AND [$12], z
+	EOR [$12], z
+	ORA [$12], z
+quad and long modes combined result in another addressing mode for
+  eight of the new mnemonics:
+	LDQ [$12]		contents of $12/$13/$14/$15
+	STQ [$12]		form the address
+	CPQ [$12]
+	ADCQ [$12]
+	SBCQ [$12]
+	ANDQ [$12]
+	EORQ [$12]
+	ORQ [$12]
+The NOP mnemonic is disabled for this instruction set because its
+opcode is re-used internally as a prefix byte.
+CAUTION: The !align pseudo opcode still inserts NOPs.

docs/cputypes/cpu 4502.txt

@@ -0,0 +1,98 @@
+
+  4502 opcode table
+
+There are no more undocumented opcodes.
+All differences to the 65ce02 are marked using '!' signs (only
+opcodes 5c and ea anyway)
+
+
+00  brk             01  ora (zp, x)     02  cle             03  see
+04  tsb zp          05  ora zp          06  asl zp          07  rmb0 zp
+08  php             09  ora #imm8       0a  asl             0b  tsy
+0c  tsb abs16       0d  ora abs16       0e  asl abs16       0f  bbr0 zp, rel8
+10  bpl rel8        11  ora (zp), y     12  ora (zp), z     13  bpl rel16
+14  trb zp          15  ora zp, x       16  asl zp, x       17  rmb1 zp
+18  clc             19  ora abs16, y    1a  inc             1b  inz
+1c  trb abs16       1d  ora abs16, x    1e  asl abs16, x    1f  bbr1 zp, rel8
+
+20  jsr abs16       21  and (zp, x)     22  jsr (abs16)     23  jsr (abs16, x)
+24  bit zp          25  and zp          26  rol zp          27  rmb2 zp
+28  plp             29  and #imm8       2a  rol             2b  tys
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f  bbr2 zp, rel8
+30  bmi rel8        31  and (zp), y     32  and (zp), z     33  bmi rel16
+34  bit zp, x       35  and zp, x       36  rol zp, x       37  rmb3 zp
+38  sec             39  and abs16, y    3a  dec             3b  dez
+3c  bit abs16, x    3d  and abs16, x    3e  rol abs16, x    3f  bbr3 zp, rel8
+
+40  rti             41  eor (zp, x)     42  neg             43  asr
+44  asr zp          45  eor zp          46  lsr zp          47  rmb4 zp
+48  pha             49  eor #imm8       4a  lsr             4b  taz
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f  bbr4 zp, rel8
+50  bvc rel8        51  eor (zp), y     52  eor (zp), z     53  bvc rel16
+54  asr zp, x       55  eor zp, x       56  lsr zp, x       57  rmb5 zp
+58  cli             59  eor abs16, y    5a  phy             5b  tab
+5c! map             5d  eor abs16, x    5e  lsr abs16, x    5f  bbr5 zp, rel8
+
+60  rts             61  adc (zp, x)     62  rtn #imm8       63  bsr rel16
+64  stz zp          65  adc zp          66  ror zp          67  rmb6 zp
+68  pla             69  adc #imm8       6a  ror             6b  tza
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f  bbr6 zp, rel8
+70  bvs rel8        71  adc (zp), y     72  adc (zp), z     73  bvs rel16
+74  stz zp, x       75  adc zp, x       76  ror zp, x       77  rmb7 zp
+78  sei             79  adc abs16, y    7a  ply             7b  tba
+7c  jmp (abs16, x)  7d  adc abs16, x    7e  ror abs16, x    7f  bbr7 zp, rel8
+
+80  bra rel8        81  sta (zp, x)     82  sta (o8, s), y  83  bra rel16
+84  sty zp          85  sta zp          86  stx zp          87  smb0 zp
+88  dey             89  bit #imm8       8a  txa             8b  sty abs16, x
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f  bbs0 zp, rel8
+90  bcc rel8        91  sta (zp), y     92  sta (zp), z     93  bcc rel16
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97  smb1 zp
+98  tya             99  sta abs16, y    9a  txs             9b  stx abs16, y
+9c  stz abs16       9d  sta abs16, x    9e  stz abs16, x    9f  bbs1 zp, rel8
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3  ldz #imm8
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7  smb2 zp
+a8  tay             a9  lda #imm8       aa  tax             ab  ldz abs16
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af  bbs2 zp, rel8
+b0  bcs rel8        b1  lda (zp), y     b2  lda (zp), z     b3  bcs rel16
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7  smb3 zp
+b8  clv             b9  lda abs16, y    ba  tsx             bb  ldz abs16, x
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf  bbs3 zp, rel8
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2  cpz #imm8       c3  dew zp
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7  smb4 zp
+c8  iny             c9  cmp #imm8       ca  dex             cb  asw abs16
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf  bbs4 zp, rel8
+d0  bne rel8        d1  cmp (zp), y     d2  cmp (zp), z     d3  bne rel16
+d4  cpz zp          d5  cmp zp, x       d6  dec zp, x       d7  smb5 zp
+d8  cld             d9  cmp abs16, y    da  phx             db  phz
+dc  cpz abs16       dd  cmp abs16, x    de  dec abs16, x    df  bbs5 zp, rel8
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2  lda (o8, s), y  e3  inw zp
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7  smb6 zp
+e8  inx             e9  sbc #imm8       ea! nop/eom         eb  row abs16
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef  bbs6 zp, rel8
+f0  beq rel8        f1  sbc (zp), y     f2  sbc (zp), z     f3  beq rel16
+f4  phw #imm16      f5  sbc zp, x       f6  inc zp, x       f7  smb7 zp
+f8  sed             f9  sbc abs16, y    fa  plx             fb  plz
+fc  phw abs16       fd  sbc abs16, x    fe  inc abs16, x    ff  bbs7 zp, rel8
+
+
+#imm8:  8-bit immediate value
+#imm16: 16-bit immediate value (only for opcode f4)
+zp:     8-bit zeropage address
+abs16:  16-bit absolute address
+rel8:   8-bit relative address offset
+rel16:  16-bit relative address offset
+o8:     8-bit offset (for stack-relative addressing, see opcodes 82 and e2)
+
+
+The 4502 is a superset of the 65ce02. Opcode 5c (originally a "4-byte NOP
+reserved for future expansion") has been changed to the "map" instruction,
+now using implied addressing.
+"map" uses the contents of the A/X/Y/Z registers to alter the memory map.
+Because this might also change the current stack memory, all interrupts are
+then implicitly disabled until an "eom" instruction gets executed.
+"eom" stands for "end of mapping", but this is actually just a new alias for
+opcode ea, the 6502-standard NOP.

docs/cputypes/cpu 6502.txt

@@ -0,0 +1,84 @@
+
+  6502 opcode table
+
+All empty entries are undocumented opcodes.
+Notice there are no documented opcodes in the fourth column.
+
+
+00  brk             01  ora (zp, x)     02                  03
+04                  05  ora zp          06  asl zp          07
+08  php             09  ora #imm8       0a  asl             0b
+0c                  0d  ora abs16       0e  asl abs16       0f
+10  bpl rel8        11  ora (zp), y     12                  13
+14                  15  ora zp, x       16  asl zp, x       17
+18  clc             19  ora abs16, y    1a                  1b
+1c                  1d  ora abs16, x    1e  asl abs16, x    1f
+
+20  jsr abs16       21  and (zp, x)     22                  23
+24  bit zp          25  and zp          26  rol zp          27
+28  plp             29  and #imm8       2a  rol             2b
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f
+30  bmi rel8        31  and (zp), y     32                  33
+34                  35  and zp, x       36  rol zp, x       37
+38  sec             39  and abs16, y    3a                  3b
+3c                  3d  and abs16, x    3e  rol abs16, x    3f
+
+40  rti             41  eor (zp, x)     42                  43
+44                  45  eor zp          46  lsr zp          47
+48  pha             49  eor #imm8       4a  lsr             4b
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f
+50  bvc rel8        51  eor (zp), y     52                  53
+54                  55  eor zp, x       56  lsr zp, x       57
+58  cli             59  eor abs16, y    5a                  5b
+5c                  5d  eor abs16, x    5e  lsr abs16, x    5f
+
+60  rts             61  adc (zp, x)     62                  63
+64                  65  adc zp          66  ror zp          67
+68  pla             69  adc #imm8       6a  ror             6b
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f
+70  bvs rel8        71  adc (zp), y     72                  73
+74                  75  adc zp, x       76  ror zp, x       77
+78  sei             79  adc abs16, y    7a                  7b
+7c                  7d  adc abs16, x    7e  ror abs16, x    7f
+
+80                  81  sta (zp, x)     82                  83
+84  sty zp          85  sta zp          86  stx zp          87
+88  dey             89                  8a  txa             8b
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f
+90  bcc rel8        91  sta (zp), y     92                  93
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97
+98  tya             99  sta abs16, y    9a  txs             9b
+9c                  9d  sta abs16, x    9e                  9f
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7
+a8  tay             a9  lda #imm8       aa  tax             ab
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af
+b0  bcs rel8        b1  lda (zp), y     b2                  b3
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7
+b8  clv             b9  lda abs16, y    ba  tsx             bb
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2                  c3
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7
+c8  iny             c9  cmp #imm8       ca  dex             cb
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf
+d0  bne rel8        d1  cmp (zp), y     d2                  d3
+d4                  d5  cmp zp, x       d6  dec zp, x       d7
+d8  cld             d9  cmp abs16, y    da                  db
+dc                  dd  cmp abs16, x    de  dec abs16, x    df
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2                  e3
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7
+e8  inx             e9  sbc #imm8       ea  nop             eb
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef
+f0  beq rel8        f1  sbc (zp), y     f2                  f3
+f4                  f5  sbc zp, x       f6  inc zp, x       f7
+f8  sed             f9  sbc abs16, y    fa                  fb
+fc                  fd  sbc abs16, x    fe  inc abs16, x    ff
+
+
+#imm8:  8-bit immediate value
+zp:     8-bit zeropage address
+abs16:  16-bit absolute address
+rel8:   8-bit relative address offset

docs/cputypes/cpu 65816.txt

@@ -0,0 +1,164 @@
+
+  65816 opcode table
+
+There are no more undocumented opcodes.
+All differences to the 65c02 are marked using a '+' sign, except
+for the #imm8 -> #imm change.
+
+
+00  brk             01  ora (dp, x)     02+ cop imm8        03+ ora offs8, s
+04  tsb dp          05  ora dp          06  asl dp          07+ ora [dp]
+08  php             09  ora #imm        0a  asl             0b+ phd
+0c  tsb abs16       0d  ora abs16       0e  asl abs16       0f+ ora abs24
+10  bpl rel8        11  ora (dp), y     12  ora (dp)        13+ ora (offs8, s), y
+14  trb dp          15  ora dp, x       16  asl dp, x       17+ ora [dp], y
+18  clc             19  ora abs16, y    1a  inc             1b+ tcs
+1c  trb abs16       1d  ora abs16, x    1e  asl abs16, x    1f+ ora abs24, x
+
+20  jsr abs16       21  and (dp, x)     22+ jsr abs24       23+ and offs8, s
+24  bit dp          25  and dp          26  rol dp          27+ and [dp]
+28  plp             29  and #imm        2a  rol             2b+ pld
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f+ and abs24
+30  bmi rel8        31  and (dp), y     32  and (dp)        33+ and (offs8, s), y
+34  bit dp, x       35  and dp, x       36  rol dp, x       37+ and [dp], y
+38  sec             39  and abs16, y    3a  dec             3b+ tsc
+3c  bit abs16, x    3d  and abs16, x    3e  rol abs16, x    3f+ and abs24, x
+
+40  rti             41  eor (dp, x)     42+ wdm             43+ eor offs8, s
+44+ mvp src, dst    45  eor dp          46  lsr dp          47+ eor [dp]
+48  pha             49  eor #imm        4a  lsr             4b+ phk
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f+ eor abs24
+50  bvc rel8        51  eor (dp), y     52  eor (dp)        53+ eor (offs8, s), y
+54+ mvn src, dst    55  eor dp, x       56  lsr dp, x       57+ eor [dp], y
+58  cli             59  eor abs16, y    5a  phy             5b+ tcd
+5c+ jmp abs24       5d  eor abs16, x    5e  lsr abs16, x    5f+ eor abs24, x
+
+60  rts             61  adc (dp, x)     62+ per rel16       63+ adc offs8, s
+64  stz dp          65  adc dp          66  ror dp          67+ adc [dp]
+68  pla             69  adc #imm        6a  ror             6b+ rtl
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f+ adc abs24
+70  bvs rel8        71  adc (dp), y     72  adc (dp)        73+ adc (offs8, s), y
+74  stz dp, x       75  adc dp, x       76  ror dp, x       77+ adc [dp], y
+78  sei             79  adc abs16, y    7a  ply             7b+ tdc
+7c  jmp (abs16, x)  7d  adc abs16, x    7e  ror abs16, x    7f+ adc abs24, x
+
+80  bra rel8        81  sta (dp, x)     82+ brl rel16       83+ sta offs8, s
+84  sty dp          85  sta dp          86  stx dp          87+ sta [dp]
+88  dey             89  bit #imm        8a  txa             8b+ phb
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f+ sta abs24
+90  bcc rel8        91  sta (dp), y     92  sta (dp)        93+ sta (offs8, s), y
+94  sty dp, x       95  sta dp, x       96  stx dp, y       97+ sta [dp], y
+98  tya             99  sta abs16, y    9a  txs             9b+ txy
+9c  stz abs16       9d  sta abs16, x    9e  stz abs16, x    9f+ sta abs24, x
+
+a0  ldy #imm        a1  lda (dp, x)     a2  ldx #imm        a3+ lda offs8, s
+a4  ldy dp          a5  lda dp          a6  ldx dp          a7+ lda [dp]
+a8  tay             a9  lda #imm        aa  tax             ab+ plb
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af+ lda abs24
+b0  bcs rel8        b1  lda (dp), y     b2  lda (dp)        b3+ lda (offs8, s), y
+b4  ldy dp, x       b5  lda dp, x       b6  ldx dp, y       b7+ lda [dp], y
+b8  clv             b9  lda abs16, y    ba  tsx             bb+ tyx
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf+ lda abs24, x
+
+c0  cpy #imm        c1  cmp (dp, x)     c2+ rep #imm8       c3+ cmp offs8, s
+c4  cpy dp          c5  cmp dp          c6  dec dp          c7+ cmp [dp]
+c8  iny             c9  cmp #imm        ca  dex             cb+ wai
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf+ cmp abs24
+d0  bne rel8        d1  cmp (dp), y     d2  cmp (dp)        d3+ cmp (offs8, s), y
+d4+ pei (dp)        d5  cmp dp, x       d6  dec dp, x       d7+ cmp [dp], y
+d8  cld             d9  cmp abs16, y    da  phx             db+ stp
+dc+ jmp [abs16]     dd  cmp abs16, x    de  dec abs16, x    df+ cmp abs24, x
+
+e0  cpx #imm        e1  sbc (dp, x)     e2+ sep #imm8       e3+ sbc offs8, s
+e4  cpx dp          e5  sbc dp          e6  inc dp          e7+ sbc [dp]
+e8  inx             e9  sbc #imm        ea  nop             eb+ xba
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef+ sbc abs24
+f0  beq rel8        f1  sbc (dp), y     f2  sbc (dp)        f3+ sbc (offs8, s), y
+f4+ pea abs16       f5  sbc dp, x       f6  inc dp, x       f7+ sbc [dp], y
+f8  sed             f9  sbc abs16, y    fa  plx             fb+ xce
+fc+ jsr (abs16, x)  fd  sbc abs16, x    fe  inc abs16, x    ff+ sbc abs24, x
+
+
+#imm:     immediate value (8 or 16 bits, depends on processor status)
+#imm8:    8-bit immediate value
+dp:       8-bit direct page address
+offs8:    8-bit offset (for stack-relative addressing)
+abs16:    16-bit absolute address
+abs24:    24-bit absolute address
+rel8:     8-bit relative address offset
+rel16:    16-bit relative address offset
+src, dst: two 8-bit bank values
+	(CAUTION: assembler expects "mnemonic src, dst" syntax,
+	but machine language order is actually "opcode dst src")
+
+
+The instruction set of the 65816 is a much extended superset of that
+of the 65c02. Among the improvements are:
+- the register widths can be switched to 16 bits
+- addresses can now be 24 bits wide
+- zero page is now called direct page, with an arbitrary base address
+- lots of new addressing modes
+- block transfer instructions
+
+New mnemonics
+-------------
+	02  cop imm8		coprocessor operation
+	6b  rtl			return long (fetches 24-bit address from stack)
+	82  brl rel16		branch long (16-bit offset)
+
+	0b  phd			push direct page register
+	4b  phk			push program bank register
+	8b  phb			push data bank register
+	62  per rel16		push effective relative address
+	d4  pei (dp)		push effective indirect address
+	f4  pea abs16		push effective absolute address
+
+	2b  pld			pull direct page register
+	ab  plb			pull data bank register
+
+	1b  tcs			transfer C to stack pointer
+	3b  tsc			transfer stack pointer to C
+	5b  tcd			transfer C to direct page register
+	7b  tdc			transfer direct page register to C
+	('C' means the whole 16-bit accumulator, even in 8-bit mode)
+	9b  txy			transfer X to Y
+	bb  tyx			transfer Y to X
+
+	eb  xba			exchange high and low bytes of accumulator
+	fb  xce			exchange Carry and Emulation bits
+
+	c2  rep #imm8		clear bits in status register
+	e2  sep #imm8		set bits in status register
+
+	cb  wai			wait for interrupt
+	db  stp			wait for reset
+
+	42  wdm			(reserved for future expansion)
+
+	block transfers:
+	44  mvp src, dst	move previous (decrementing addresses)
+	54  mvn src, dst	move next (incrementing addresses)
+	the arguments are bank numbers. block size minus one must be in 16-bit
+	accumulator. X holds source address, Y holds target address.
+	after these instructions, data bank register is set to "dst".
+	if blocks overlap:
+	when moving a block to a higher address, use mvp and put the highest
+	addresses in X/Y.
+	when moving a block to a lower address, use mvn and put the lowest
+	adresses in X/Y.
+
+New addressing modes for existing mnemonics
+-------------------------------------------
+	22  jsr abs24
+	5c  jmp abs24
+	dc  jmp [abs16]
+	fc  jsr (abs16, x)
+
+	ora and eor adc sta lda cmp sbc
+	-------------------------------
+	03  23  43  63  83  a3  c3  e3  offs8, s
+	07  27  47  67  87  a7  c7  e7  [dp]
+	0f  2f  4f  6f  8f  af  cf  ef  abs24
+	13  33  53  73  93  b3  d3  f3  (offs8, s), y
+	17  37  57  77  97  b7  d7  f7  [dp], y
+	1f  3f  5f  7f  9f  bf  df  ff  abs24, x

docs/cputypes/cpu 65c02.txt

@@ -0,0 +1,130 @@
+
+  65c02 opcode table
+
+All empty entries are undocumented opcodes.
+Notice there are still no documented opcodes in the fourth column.
+All differences to the original 6502 are marked using a '+' sign.
+
+
+00  brk             01  ora (zp, x)     02                  03
+04+ tsb zp          05  ora zp          06  asl zp          07
+08  php             09  ora #imm8       0a  asl             0b
+0c+ tsb abs16       0d  ora abs16       0e  asl abs16       0f
+10  bpl rel8        11  ora (zp), y     12+ ora (zp)        13
+14+ trb zp          15  ora zp, x       16  asl zp, x       17
+18  clc             19  ora abs16, y    1a+ inc             1b
+1c+ trb abs16       1d  ora abs16, x    1e  asl abs16, x    1f
+
+20  jsr abs16       21  and (zp, x)     22                  23
+24  bit zp          25  and zp          26  rol zp          27
+28  plp             29  and #imm8       2a  rol             2b
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f
+30  bmi rel8        31  and (zp), y     32+ and (zp)        33
+34+ bit zp, x       35  and zp, x       36  rol zp, x       37
+38  sec             39  and abs16, y    3a+ dec             3b
+3c+ bit abs16, x    3d  and abs16, x    3e  rol abs16, x    3f
+
+40  rti             41  eor (zp, x)     42                  43
+44                  45  eor zp          46  lsr zp          47
+48  pha             49  eor #imm8       4a  lsr             4b
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f
+50  bvc rel8        51  eor (zp), y     52+ eor (zp)        53
+54                  55  eor zp, x       56  lsr zp, x       57
+58  cli             59  eor abs16, y    5a+ phy             5b
+5c                  5d  eor abs16, x    5e  lsr abs16, x    5f
+
+60  rts             61  adc (zp, x)     62                  63
+64+ stz zp          65  adc zp          66  ror zp          67
+68  pla             69  adc #imm8       6a  ror             6b
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f
+70  bvs rel8        71  adc (zp), y     72+ adc (zp)        73
+74+ stz zp, x       75  adc zp, x       76  ror zp, x       77
+78  sei             79  adc abs16, y    7a+ ply             7b
+7c+ jmp (abs16, x)  7d  adc abs16, x    7e  ror abs16, x    7f
+
+80+ bra rel8        81  sta (zp, x)     82                  83
+84  sty zp          85  sta zp          86  stx zp          87
+88  dey             89+ bit #imm8       8a  txa             8b
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f
+90  bcc rel8        91  sta (zp), y     92+ sta (zp)        93
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97
+98  tya             99  sta abs16, y    9a  txs             9b
+9c+ stz abs16       9d  sta abs16, x    9e+ stz abs16, x    9f
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7
+a8  tay             a9  lda #imm8       aa  tax             ab
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af
+b0  bcs rel8        b1  lda (zp), y     b2+ lda (zp)        b3
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7
+b8  clv             b9  lda abs16, y    ba  tsx             bb
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2                  c3
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7
+c8  iny             c9  cmp #imm8       ca  dex             cb
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf
+d0  bne rel8        d1  cmp (zp), y     d2+ cmp (zp)        d3
+d4                  d5  cmp zp, x       d6  dec zp, x       d7
+d8  cld             d9  cmp abs16, y    da+ phx             db
+dc                  dd  cmp abs16, x    de  dec abs16, x    df
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2                  e3
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7
+e8  inx             e9  sbc #imm8       ea  nop             eb
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef
+f0  beq rel8        f1  sbc (zp), y     f2+ sbc (zp)        f3
+f4                  f5  sbc zp, x       f6  inc zp, x       f7
+f8  sed             f9  sbc abs16, y    fa+ plx             fb
+fc                  fd  sbc abs16, x    fe  inc abs16, x    ff
+
+
+#imm8: 8-bit immediate value
+zp:    8-bit zeropage address
+abs16: 16-bit absolute address
+rel8:  8-bit relative address offset
+
+
+The 65c02 is the CMOS re-design of the 6502. It has a few
+improvements:
+
+New mnemonics
+-------------
+	test and (re-)set bits against accumulator:
+		04  tsb zp
+		0c  tsb abs16
+		14  trb zp
+		1c  trb abs16
+	push/pull x/y:
+		5a  phy
+		7a  ply
+		da  phx
+		fa  plx
+	store zero:
+		64  stz zp
+		74  stz zp, x
+		9c  stz abs16
+		9e  stz abs16, x
+	branch always:
+		80  bra rel8
+
+New addressing modes for existing mnemonics
+-------------------------------------------
+	zp indirect addressing without indexing:
+		12  ora (zp)
+		32  and (zp)
+		52  eor (zp)
+		72  adc (zp)
+		92  sta (zp)
+		b2  lda (zp)
+		d2  cmp (zp)
+		f2  sbc (zp)
+	implied ("accumulator") addressing:
+		1a  inc
+		3a  dec
+	x-indexed indirect addressing:
+		7c  jmp (abs16, x)
+	three more addressing modes for BIT:
+		34  bit zp, x
+		3c  bit abs16, x
+		89  bit #imm8

docs/cputypes/cpu 65ce02.txt

@@ -0,0 +1,162 @@
+
+  65ce02 opcode table
+
+There are no more undocumented opcodes.
+All differences to the r65c02 are marked using '+' or '!' signs:
+'!' marks (backward-compatible) changes to existing instructions.
+'+' marks new instructions.
+
+
+00  brk             01  ora (zp, x)     02+ cle             03+ see
+04  tsb zp          05  ora zp          06  asl zp          07  rmb0 zp
+08  php             09  ora #imm8       0a  asl             0b+ tsy
+0c  tsb abs16       0d  ora abs16       0e  asl abs16       0f  bbr0 zp, rel8
+10  bpl rel8        11  ora (zp), y     12! ora (zp), z     13+ bpl rel16
+14  trb zp          15  ora zp, x       16  asl zp, x       17  rmb1 zp
+18  clc             19  ora abs16, y    1a  inc             1b+ inz
+1c  trb abs16       1d  ora abs16, x    1e  asl abs16, x    1f  bbr1 zp, rel8
+
+20  jsr abs16       21  and (zp, x)     22+ jsr (abs16)     23+ jsr (abs16, x)
+24  bit zp          25  and zp          26  rol zp          27  rmb2 zp
+28  plp             29  and #imm8       2a  rol             2b+ tys
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f  bbr2 zp, rel8
+30  bmi rel8        31  and (zp), y     32! and (zp), z     33+ bmi rel16
+34  bit zp, x       35  and zp, x       36  rol zp, x       37  rmb3 zp
+38  sec             39  and abs16, y    3a  dec             3b+ dez
+3c  bit abs16, x    3d  and abs16, x    3e  rol abs16, x    3f  bbr3 zp, rel8
+
+40  rti             41  eor (zp, x)     42+ neg             43+ asr
+44+ asr zp          45  eor zp          46  lsr zp          47  rmb4 zp
+48  pha             49  eor #imm8       4a  lsr             4b+ taz
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f  bbr4 zp, rel8
+50  bvc rel8        51  eor (zp), y     52! eor (zp), z     53+ bvc rel16
+54+ asr zp, x       55  eor zp, x       56  lsr zp, x       57  rmb5 zp
+58  cli             59  eor abs16, y    5a  phy             5b+ tab
+5c+ aug             5d  eor abs16, x    5e  lsr abs16, x    5f  bbr5 zp, rel8
+
+60  rts             61  adc (zp, x)     62+ rtn #imm8       63+ bsr rel16
+64! stz zp          65  adc zp          66  ror zp          67  rmb6 zp
+68  pla             69  adc #imm8       6a  ror             6b+ tza
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f  bbr6 zp, rel8
+70  bvs rel8        71  adc (zp), y     72! adc (zp), z     73+ bvs rel16
+74! stz zp, x       75  adc zp, x       76  ror zp, x       77  rmb7 zp
+78  sei             79  adc abs16, y    7a  ply             7b+ tba
+7c  jmp (abs16, x)  7d  adc abs16, x    7e  ror abs16, x    7f  bbr7 zp, rel8
+
+80  bra rel8        81  sta (zp, x)     82+ sta (o8, s), y  83+ bra rel16
+84  sty zp          85  sta zp          86  stx zp          87  smb0 zp
+88  dey             89  bit #imm8       8a  txa             8b+ sty abs16, x
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f  bbs0 zp, rel8
+90  bcc rel8        91  sta (zp), y     92! sta (zp), z     93+ bcc rel16
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97  smb1 zp
+98  tya             99  sta abs16, y    9a  txs             9b+ stx abs16, y
+9c! stz abs16       9d  sta abs16, x    9e! stz abs16, x    9f  bbs1 zp, rel8
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3+ ldz #imm8
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7  smb2 zp
+a8  tay             a9  lda #imm8       aa  tax             ab+ ldz abs16
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af  bbs2 zp, rel8
+b0  bcs rel8        b1  lda (zp), y     b2! lda (zp), z     b3+ bcs rel16
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7  smb3 zp
+b8  clv             b9  lda abs16, y    ba  tsx             bb+ ldz abs16, x
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf  bbs3 zp, rel8
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2+ cpz #imm8       c3+ dew zp
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7  smb4 zp
+c8  iny             c9  cmp #imm8       ca  dex             cb+ asw abs16
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf  bbs4 zp, rel8
+d0  bne rel8        d1  cmp (zp), y     d2! cmp (zp), z     d3+ bne rel16
+d4+ cpz zp          d5  cmp zp, x       d6  dec zp, x       d7  smb5 zp
+d8  cld             d9  cmp abs16, y    da  phx             db+ phz
+dc+ cpz abs16       dd  cmp abs16, x    de  dec abs16, x    df  bbs5 zp, rel8
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2+ lda (o8, s), y  e3+ inw zp
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7  smb6 zp
+e8  inx             e9  sbc #imm8       ea  nop             eb+ row abs16
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef  bbs6 zp, rel8
+f0  beq rel8        f1  sbc (zp), y     f2! sbc (zp), z     f3+ beq rel16
+f4+ phw #imm16      f5  sbc zp, x       f6  inc zp, x       f7  smb7 zp
+f8  sed             f9  sbc abs16, y    fa  plx             fb+ plz
+fc+ phw abs16       fd  sbc abs16, x    fe  inc abs16, x    ff  bbs7 zp, rel8
+
+
+#imm8:  8-bit immediate value
+#imm16: 16-bit immediate value (only for opcode f4)
+zp:     8-bit zeropage address
+abs16:  16-bit absolute address
+rel8:   8-bit relative address offset
+rel16:  16-bit relative address offset
+o8:     8-bit offset (for stack-relative addressing, see opcodes 82 and e2)
+
+
+The 65ce02 is a superset of the r65c02. It has several improvements:
+
+new Z register
+--------------
+The "stz" mnemonic no longer means "store zero", but "store Z register"
+	(see opcodes 64, 74, 9c, 9e)
+The "(zp)" addressing mode now becomes "(zp), z"
+	(see opcodes 12, 32, 52, 72, 92, b2, d2, f2)
+Z defaults to zero after reset, so these changes are backward compatible
+until Z is loaded with a non-zero value.
+New instructions for this register:
+	1b  inz			increment Z
+	3b  dez			decrement Z
+	4b  taz			transfer A to Z
+	6b  tza			transfer Z to A
+	a3  ldz #imm8
+	ab  ldz abs16
+	bb  ldz abs16, x
+	c2  cpz #imm8
+	d4  cpz zp
+	db  phz			push Z
+	dc  cpz abs16
+	fb  plz			pull Z
+
+16-bit stack pointer
+--------------------
+	02  cle			clear stack extend disable
+	03  see			set stack extend disable
+	0b  tsy			transfer stack_ptr_high to Y
+	2b  tys			transfer Y to stack_ptr_high
+
+16-bit branches
+---------------
+	13  bpl rel16
+	33  bmi rel16
+	53  bvc rel16
+	63  bsr rel16		relative jsr, "branch to subroutine"
+	73  bvs rel16
+	83  bra rel16		relative jmp
+	93  bcc rel16
+	b3  bcs rel16
+	d3  bne rel16
+	f3  beq rel16
+To use these in ACME, use the mnemonics lbpl, lbmi, lbvc, ...
+(except for "bsr", because there is no 8-bit version of it anyway)
+
+new addressing modes for existing instructions
+----------------------------------------------
+	22  jsr (abs16)
+	23  jsr (abs16, x)
+	82  sta (offset8, s), y
+	8b  sty abs16, x
+	9b  stx abs16, y
+	e2  lda (offset8, s), y
+
+new instructions
+----------------
+	42  neg			negate A
+	43  asr
+	44  asr zp
+	54  asr zp, x
+	5b  tab
+	5c  aug			"4-byte NOP reserved for future expansion"
+	62  rtn #imm8
+	7b  tba
+	c3  dew zp
+	cb  asw abs16
+	e3  inw zp
+	eb  row abs16
+	f4  phw #imm16
+	fc  phw abs16

docs/cputypes/cpu m65.txt

@@ -0,0 +1,120 @@
+
+  m65 opcode table(s)
+
+The m65 instruction set extends the 4502 instruction set using prefix bytes.
+Therefore, the "normal" opcode table is the same as for the 4502 cpu (see that
+file), so this file only contains information about the extensions.
+
+
+"quad mode" allows 32-bit data operations using a virtual register called 'Q'.
+The mnemonics aslq/lsrq/rolq/rorq/inq/deq have five addressing modes.
+The mnemonic ldq has eight addressing modes in quad mode, and a ninth when
+combined with long mode.
+The mnemonics stq/cpq/adcq/sbcq/andq/eorq/orq have three addressing modes in
+quad mode, and a fourth when combined with long mode.
+The mnemonic bitq has two addressing modes.
+The mnemonic asrq has three addressing modes.
+This mode is entered after a NEG:NEG (42 42) prefix, the following opcode is
+then taken from this table:
+
+00                  01                  02                  03
+04                  05 orq zp           06 aslq zp          07
+08                  09                  0a aslq             0b
+0c                  0d orq abs16        0e aslq abs16       0f
+10                  11                  12 orq (zp)         13
+14                  15                  16 aslq zp, x       17
+18                  19                  1a inq              1b
+1c                  1d                  1e aslq abs16, x    1f
+
+20                  21                  22                  23
+24 bitq zp          25 andq zp          26 rolq zp          27
+28                  29                  2a rolq             2b
+2c bitq abs16       2d andq abs16       2e rolq abs16       2f
+30                  31                  32 andq (zp)        33
+34                  35                  36 rolq zp, x       37
+38                  39                  3a deq              3b
+3c                  3d                  3e rolq abs16, x    3f
+
+40                  41                  42                  43 asrq
+44 asrq zp          45 eorq zp          46 lsrq zp          47
+48                  49                  4a lsrq             4b
+4c                  4d eorq abs16       4e lsrq abs16       4f
+50                  51                  52 eorq (zp)        53
+54 asrq zp, x       55                  56 lsrq zp, x       57
+58                  59                  5a                  5b
+5c                  5d                  5e lsrq abs16, x    5f
+
+60                  61                  62                  63
+64                  65 adcq zp          66 rorq zp          67
+68                  69                  6a rorq             6b
+6c                  6d adcq abs16       6e rorq abs16       6f
+70                  71                  72 adcq (zp)        73
+74                  75                  76 rorq zp, x       77
+78                  79                  7a                  7b
+7c                  7d                  7e rorq abs16, x    7f
+
+80                  81                  82                  83
+84                  85 stq zp           86                  87
+88                  89                  8a                  8b
+8c                  8d stq abs16        8e                  8f
+90                  91                  92 stq (zp)         93
+94                  95                  96                  97
+98                  99                  9a                  9b
+9c                  9d                  9e                  9f
+
+a0                  a1                  a2                  a3
+a4                  a5 ldq zp           a6                  a7
+a8                  a9                  aa                  ab
+ac                  ad ldq abs16        ae                  af
+b0                  b1 ldq (zp), y      b2 ldq (zp)         b3
+b4                  b5 ldq zp, x        b6                  b7
+b8                  b9 ldq abs16, y     ba                  bb
+bc                  bd ldq abs16, x     be                  bf
+
+c0                  c1                  c2                  c3
+c4                  c5 cpq zp           c6 deq zp           c7
+c8                  c9                  ca                  cb
+cc                  cd cpq abs16        ce deq abs16        cf
+d0                  d1                  d2 cpq (zp)         d3
+d4                  d5                  d6 deq zp, x        d7
+d8                  d9                  da                  db
+dc                  dd                  de deq abs16, x     df
+
+e0                  e1                  e2 ldq (zp, s), y   e3
+e4                  e5 sbcq zp          e6 inq zp           e7
+e8                  e9                  ea                  eb
+ec                  ed sbcq abs16       ee inq abs16        ef
+f0                  f1                  f2 sbcq (zp)        f3
+f4                  f5                  f6 inq zp, x        f7
+f8                  f9                  fa                  fb
+fc                  fd                  fe inq abs16, x     ff
+
+zp:     8-bit zeropage address
+abs16:  16-bit absolute address
+
+
+"long mode" adds an addressing mode using 32-bit pointers for eight existing
+mnemonics. This mode is entered after a NOP (ea) prefix, the following opcode
+should then be one of these:
+
+12 ora [zp], z      32 and [zp], z      52 eor [zp], z      72 adc [zp], z
+92 sta [zp], z      b2 lda [zp], z      d2 cmp [zp], z      f2 sbc [zp], z
+
+
+"quad" and "long" modes can be combined to have 32-bit data access using a
+32-bit pointer. This adds another addressing mode for eight of the new
+mnemonics. This mode is entered after a NEG:NEG:NOP (42 42 ea) prefix, the
+following opcode should then be one of these:
+
+12 orq [zp]         32 andq [zp]        52 eorq [zp]        72 adcq [zp]
+92 stq [zp]         b2 ldq [zp]         d2 cpq [zp]         f2 sbcq [zp]
+
+
+Because the addressing modes are changed a bit by the prefix codes, here are
+some of the unsupported combinations just for comparison (these result in
+"Illegal combination of command and addressing mode"):
+	lda (zp)	; 65c02 knew this, but 65ce02 added z index!
+	lda [zp]	; long mode also expects z index!
+	ldq #imm	; quad mode has no immediate addressing!
+	ldq (zp), z	; quad mode does not use z index!
+	ldq [zp], z	; quad and long modes combined do not use z index!

docs/cputypes/cpu nmos6502.txt

@@ -0,0 +1,86 @@
+
+  nmos6502 opcode table
+
+This table includes all of the unintended ("illegal") opcodes. These are
+marked using '+' or '!' signs:
+'+' means the instruction is supported by ACME,
+'!' means ACME will use a different (but functionally equivalent) opcode.
+
+
+00  brk             01  ora (zp, x)     02+ jam             03+ slo (zp, x)
+04+ nop zp          05  ora zp          06  asl zp          07+ slo zp
+08  php             09  ora #imm8       0a  asl             0b+ anc #imm8
+0c+ nop abs16       0d  ora abs16       0e  asl abs16       0f+ slo abs16
+10  bpl rel8        11  ora (zp), y     12! jam             13+ slo (zp), y
+14+ nop zp, x       15  ora zp, x       16  asl zp, x       17+ slo zp, x
+18  clc             19  ora abs16, y    1a! nop             1b+ slo abs16, y
+1c+ nop abs16, x    1d  ora abs16, x    1e  asl abs16, x    1f+ slo abs16, x
+
+20  jsr abs16       21  and (zp, x)     22! jam             23+ rla (zp, x)
+24  bit zp          25  and zp          26  rol zp          27+ rla zp
+28  plp             29  and #imm8       2a  rol             2b! anc #imm8
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f+ rla abs16
+30  bmi rel8        31  and (zp), y     32! jam             33+ rla (zp), y
+34! nop zp, x       35  and zp, x       36  rol zp, x       37+ rla zp, x
+38  sec             39  and abs16, y    3a! nop             3b+ rla abs16, y
+3c! nop abs16, x    3d  and abs16, x    3e  rol abs16, x    3f+ rla abs16, x
+
+40  rti             41  eor (zp, x)     42! jam             43+ sre (zp, x)
+44! nop zp          45  eor zp          46  lsr zp          47+ sre zp
+48  pha             49  eor #imm8       4a  lsr             4b+ alr #imm8
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f+ sre abs16
+50  bvc rel8        51  eor (zp), y     52! jam             53+ sre (zp), y
+54! nop zp, x       55  eor zp, x       56  lsr zp, x       57+ sre zp, x
+58  cli             59  eor abs16, y    5a! nop             5b+ sre abs16, y
+5c! nop abs16, x    5d  eor abs16, x    5e  lsr abs16, x    5f+ sre abs16, x
+
+60  rts             61  adc (zp, x)     62! jam             63+ rra (zp, x)
+64! nop zp          65  adc zp          66  ror zp          67+ rra zp
+68  pla             69  adc #imm8       6a  ror             6b+ arr #imm8
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f+ rra abs16
+70  bvs rel8        71  adc (zp), y     72! jam             73+ rra (zp), y
+74! nop zp, x       75  adc zp, x       76  ror zp, x       77+ rra zp, x
+78  sei             79  adc abs16, y    7a! nop             7b+ rra abs16, y
+7c! nop abs16, x    7d  adc abs16, x    7e  ror abs16, x    7f+ rra abs16, x
+
+80+ nop #imm8       81  sta (zp, x)     82! nop #imm8       83+ sax (zp, x)
+84  sty zp          85  sta zp          86  stx zp          87+ sax zp
+88  dey             89! nop #imm8       8a  txa             8b+ ane #imm8
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f+ sax abs16
+90  bcc rel8        91  sta (zp), y     92! jam             93+ sha (zp), y
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97+ sax zp, y
+98  tya             99  sta abs16, y    9a  txs             9b+ tas abs16, y
+9c+ shy abs16, x    9d  sta abs16, x    9e+ shx abs16, y    9f+ sha abs16, y
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3+ lax (zp, x)
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7+ lax zp
+a8  tay             a9  lda #imm8       aa  tax             ab+ lxa #imm8
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af+ lax abs16
+b0  bcs rel8        b1  lda (zp), y     b2! jam             b3+ lax (zp), y
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7+ lax zp, y
+b8  clv             b9  lda abs16, y    ba  tsx             bb+ las abs16, y
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf+ lax abs16, y
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2! nop #imm8       c3+ dcp (zp, x)
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7+ dcp zp
+c8  iny             c9  cmp #imm8       ca  dex             cb+ sbx #imm8
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf+ dcp abs16
+d0  bne rel8        d1  cmp (zp), y     d2! jam             d3+ dcp (zp), y
+d4! nop zp, x       d5  cmp zp, x       d6  dec zp, x       d7+ dcp zp, x
+d8  cld             d9  cmp abs16, y    da! nop             db+ dcp abs16, y
+dc! nop abs16, x    dd  cmp abs16, x    de  dec abs16, x    df+ dcp abs16, x
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2! nop #imm8       e3+ isc (zp, x)
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7+ isc zp
+e8  inx             e9  sbc #imm8       ea  nop             eb! sbc #imm8
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef+ isc abs16
+f0  beq rel8        f1  sbc (zp), y     f2! jam             f3+ isc (zp), y
+f4! nop zp, x       f5  sbc zp, x       f6  inc zp, x       f7+ isc zp, x
+f8  sed             f9  sbc abs16, y    fa! nop             fb+ isc abs16, y
+fc! nop abs16, x    fd  sbc abs16, x    fe  inc abs16, x    ff+ isc abs16, x
+
+
+#imm8:  8-bit immediate value
+zp:     8-bit zeropage address
+abs16:  16-bit absolute address
+rel8:   8-bit relative address offset

docs/cputypes/cpu r65c02.txt

@@ -0,0 +1,94 @@
+
+  r65c02 opcode table
+
+All empty entries are undocumented opcodes.
+All differences to the 65c02 (all in the fourth column) are marked
+using a '+' sign.
+
+
+00  brk             01  ora (zp, x)     02                  03
+04  tsb zp          05  ora zp          06  asl zp          07+ rmb0 zp
+08  php             09  ora #imm8       0a  asl             0b
+0c  tsb abs16       0d  ora abs16       0e  asl abs16       0f+ bbr0 zp, rel8
+10  bpl rel8        11  ora (zp), y     12  ora (zp)        13
+14  trb zp          15  ora zp, x       16  asl zp, x       17+ rmb1 zp
+18  clc             19  ora abs16, y    1a  inc             1b
+1c  trb abs16       1d  ora abs16, x    1e  asl abs16, x    1f+ bbr1 zp, rel8
+
+20  jsr abs16       21  and (zp, x)     22                  23
+24  bit zp          25  and zp          26  rol zp          27+ rmb2 zp
+28  plp             29  and #imm8       2a  rol             2b
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f+ bbr2 zp, rel8
+30  bmi rel8        31  and (zp), y     32  and (zp)        33
+34  bit zp, x       35  and zp, x       36  rol zp, x       37+ rmb3 zp
+38  sec             39  and abs16, y    3a  dec             3b
+3c  bit abs16, x    3d  and abs16, x    3e  rol abs16, x    3f+ bbr3 zp, rel8
+
+40  rti             41  eor (zp, x)     42                  43
+44                  45  eor zp          46  lsr zp          47+ rmb4 zp
+48  pha             49  eor #imm8       4a  lsr             4b
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f+ bbr4 zp, rel8
+50  bvc rel8        51  eor (zp), y     52  eor (zp)        53
+54                  55  eor zp, x       56  lsr zp, x       57+ rmb5 zp
+58  cli             59  eor abs16, y    5a  phy             5b
+5c                  5d  eor abs16, x    5e  lsr abs16, x    5f+ bbr5 zp, rel8
+
+60  rts             61  adc (zp, x)     62                  63
+64  stz zp          65  adc zp          66  ror zp          67+ rmb6 zp
+68  pla             69  adc #imm8       6a  ror             6b
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f+ bbr6 zp, rel8
+70  bvs rel8        71  adc (zp), y     72  adc (zp)        73
+74  stz zp, x       75  adc zp, x       76  ror zp, x       77+ rmb7 zp
+78  sei             79  adc abs16, y    7a  ply             7b
+7c  jmp (abs16, x)  7d  adc abs16, x    7e  ror abs16, x    7f+ bbr7 zp, rel8
+
+80  bra rel8        81  sta (zp, x)     82                  83
+84  sty zp          85  sta zp          86  stx zp          87+ smb0 zp
+88  dey             89  bit #imm8       8a  txa             8b
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f+ bbs0 zp, rel8
+90  bcc rel8        91  sta (zp), y     92  sta (zp)        93
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97+ smb1 zp
+98  tya             99  sta abs16, y    9a  txs             9b
+9c  stz abs16       9d  sta abs16, x    9e  stz abs16, x    9f+ bbs1 zp, rel8
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7+ smb2 zp
+a8  tay             a9  lda #imm8       aa  tax             ab
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af+ bbs2 zp, rel8
+b0  bcs rel8        b1  lda (zp), y     b2  lda (zp)        b3
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7+ smb3 zp
+b8  clv             b9  lda abs16, y    ba  tsx             bb
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf+ bbs3 zp, rel8
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2                  c3
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7+ smb4 zp
+c8  iny             c9  cmp #imm8       ca  dex             cb
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf+ bbs4 zp, rel8
+d0  bne rel8        d1  cmp (zp), y     d2  cmp (zp)        d3
+d4                  d5  cmp zp, x       d6  dec zp, x       d7+ smb5 zp
+d8  cld             d9  cmp abs16, y    da  phx             db
+dc                  dd  cmp abs16, x    de  dec abs16, x    df+ bbs5 zp, rel8
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2                  e3
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7+ smb6 zp
+e8  inx             e9  sbc #imm8       ea  nop             eb
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef+ bbs6 zp, rel8
+f0  beq rel8        f1  sbc (zp), y     f2  sbc (zp)        f3
+f4                  f5  sbc zp, x       f6  inc zp, x       f7+ smb7 zp
+f8  sed             f9  sbc abs16, y    fa  plx             fb
+fc                  fd  sbc abs16, x    fe  inc abs16, x    ff+ bbs7 zp, rel8
+
+
+#imm8: 8-bit immediate value
+zp:    8-bit zeropage address
+abs16: 16-bit absolute address
+rel8:  8-bit relative address offset
+
+
+The r65c02 is a superset of the 65c02. It adds bit manipulation instructions:
+	smbB zp			set bit in zp location
+	rmbB zp			reset bit in zp location
+	bbsB zp, rel8		branch if bit is set in zp location
+	bbrB zp, rel8		branch if bit is reset in zp location
+The 'B' in the mnemonic is the bit number, therefore it must be in
+the 0..7 range.

docs/cputypes/cpu w65c02.txt

@@ -0,0 +1,90 @@
+
+  w65c02 opcode table
+
+All empty entries are undocumented opcodes.
+All differences to the r65c02 are marked using a '+' sign (only two
+opcodes anyway: cb and db)
+
+
+00  brk             01  ora (zp, x)     02                  03
+04  tsb zp          05  ora zp          06  asl zp          07  rmb0 zp
+08  php             09  ora #imm8       0a  asl             0b
+0c  tsb abs16       0d  ora abs16       0e  asl abs16       0f  bbr0 zp, rel8
+10  bpl rel8        11  ora (zp), y     12  ora (zp)        13
+14  trb zp          15  ora zp, x       16  asl zp, x       17  rmb1 zp
+18  clc             19  ora abs16, y    1a  inc             1b
+1c  trb abs16       1d  ora abs16, x    1e  asl abs16, x    1f  bbr1 zp, rel8
+
+20  jsr abs16       21  and (zp, x)     22                  23
+24  bit zp          25  and zp          26  rol zp          27  rmb2 zp
+28  plp             29  and #imm8       2a  rol             2b
+2c  bit abs16       2d  and abs16       2e  rol abs16       2f  bbr2 zp, rel8
+30  bmi rel8        31  and (zp), y     32  and (zp)        33
+34  bit zp, x       35  and zp, x       36  rol zp, x       37  rmb3 zp
+38  sec             39  and abs16, y    3a  dec             3b
+3c  bit abs16, x    3d  and abs16, x    3e  rol abs16, x    3f  bbr3 zp, rel8
+
+40  rti             41  eor (zp, x)     42                  43
+44                  45  eor zp          46  lsr zp          47  rmb4 zp
+48  pha             49  eor #imm8       4a  lsr             4b
+4c  jmp abs16       4d  eor abs16       4e  lsr abs16       4f  bbr4 zp, rel8
+50  bvc rel8        51  eor (zp), y     52  eor (zp)        53
+54                  55  eor zp, x       56  lsr zp, x       57  rmb5 zp
+58  cli             59  eor abs16, y    5a  phy             5b
+5c                  5d  eor abs16, x    5e  lsr abs16, x    5f  bbr5 zp, rel8
+
+60  rts             61  adc (zp, x)     62                  63
+64  stz zp          65  adc zp          66  ror zp          67  rmb6 zp
+68  pla             69  adc #imm8       6a  ror             6b
+6c  jmp (abs16)     6d  adc abs16       6e  ror abs16       6f  bbr6 zp, rel8
+70  bvs rel8        71  adc (zp), y     72  adc (zp)        73
+74  stz zp, x       75  adc zp, x       76  ror zp, x       77  rmb7 zp
+78  sei             79  adc abs16, y    7a  ply             7b
+7c  jmp (abs16, x)  7d  adc abs16, x    7e  ror abs16, x    7f  bbr7 zp, rel8
+
+80  bra rel8        81  sta (zp, x)     82                  83
+84  sty zp          85  sta zp          86  stx zp          87  smb0 zp
+88  dey             89  bit #imm8       8a  txa             8b
+8c  sty abs16       8d  sta abs16       8e  stx abs16       8f  bbs0 zp, rel8
+90  bcc rel8        91  sta (zp), y     92  sta (zp)        93
+94  sty zp, x       95  sta zp, x       96  stx zp, y       97  smb1 zp
+98  tya             99  sta abs16, y    9a  txs             9b
+9c  stz abs16       9d  sta abs16, x    9e  stz abs16, x    9f  bbs1 zp, rel8
+
+a0  ldy #imm8       a1  lda (zp, x)     a2  ldx #imm8       a3
+a4  ldy zp          a5  lda zp          a6  ldx zp          a7  smb2 zp
+a8  tay             a9  lda #imm8       aa  tax             ab
+ac  ldy abs16       ad  lda abs16       ae  ldx abs16       af  bbs2 zp, rel8
+b0  bcs rel8        b1  lda (zp), y     b2  lda (zp)        b3
+b4  ldy zp, x       b5  lda zp, x       b6  ldx zp, y       b7  smb3 zp
+b8  clv             b9  lda abs16, y    ba  tsx             bb
+bc  ldy abs16, x    bd  lda abs16, x    be  ldx abs16, y    bf  bbs3 zp, rel8
+
+c0  cpy #imm8       c1  cmp (zp, x)     c2                  c3
+c4  cpy zp          c5  cmp zp          c6  dec zp          c7  smb4 zp
+c8  iny             c9  cmp #imm8       ca  dex             cb+ wai
+cc  cpy abs16       cd  cmp abs16       ce  dec abs16       cf  bbs4 zp, rel8
+d0  bne rel8        d1  cmp (zp), y     d2  cmp (zp)        d3
+d4                  d5  cmp zp, x       d6  dec zp, x       d7  smb5 zp
+d8  cld             d9  cmp abs16, y    da  phx             db+ stp
+dc                  dd  cmp abs16, x    de  dec abs16, x    df  bbs5 zp, rel8
+
+e0  cpx #imm8       e1  sbc (zp, x)     e2                  e3
+e4  cpx zp          e5  sbc zp          e6  inc zp          e7  smb6 zp
+e8  inx             e9  sbc #imm8       ea  nop             eb
+ec  cpx abs16       ed  sbc abs16       ee  inc abs16       ef  bbs6 zp, rel8
+f0  beq rel8        f1  sbc (zp), y     f2  sbc (zp)        f3
+f4                  f5  sbc zp, x       f6  inc zp, x       f7  smb7 zp
+f8  sed             f9  sbc abs16, y    fa  plx             fb
+fc                  fd  sbc abs16, x    fe  inc abs16, x    ff  bbs7 zp, rel8
+
+
+#imm8: 8-bit immediate value
+zp:    8-bit zeropage address
+abs16: 16-bit absolute address
+rel8:  8-bit relative address offset
+
+
+The w65c02 is a superset of the r65c02. It only adds two instructions:
+	cb wai		wait for interrupt
+	db stp		wait for reset

src/Makefile

@@ -1,4 +1,5 @@
 CFLAGS		= -O3 -Wall -Wstrict-prototypes
+#CFLAGS		= -O3 -Wall -Wextra -Wstrict-prototypes
 LIBS		= -lm
 CC		= gcc
 RM		= rm
@@ -31,7 +32,7 @@ encoding.o: config.h alu.h acme.h dynabuf.h global.h output.h input.h tree.h enc
 
 flow.o: config.h acme.h alu.h dynabuf.h global.h input.h mnemo.h symbol.h tree.h flow.h flow.c
 
-global.o: config.h platform.h acme.h cpu.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
+global.o: config.h platform.h acme.h cpu.h dynabuf.h encoding.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
 
 input.o: config.h alu.h dynabuf.h global.h section.h symbol.h tree.h input.h input.c
 
@@ -43,7 +44,7 @@ output.o: config.h acme.h alu.h cpu.h dynabuf.h global.h input.h tree.h output.h
 
 platform.o: config.h platform.h platform.c
 
-pseudoopcodes.o: acme.h alu.h input.h macro.h output.h pseudoopcodes.h pseudoopcodes.c
+pseudoopcodes.o: acme.h alu.h global.h input.h macro.h output.h symbol.h pseudoopcodes.h pseudoopcodes.c
 
 section.o: config.h dynabuf.h global.h symbol.h tree.h section.h section.c
 

src/Makefile.dos

@@ -32,7 +32,7 @@ encoding.o: config.h alu.h acme.h dynabuf.h global.h output.h input.h tree.h enc
 
 flow.o: config.h acme.h alu.h dynabuf.h global.h input.h mnemo.h symbol.h tree.h flow.h flow.c
 
-global.o: config.h platform.h acme.h cpu.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
+global.o: config.h platform.h acme.h cpu.h dynabuf.h encoding.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
 
 input.o: config.h alu.h dynabuf.h global.h section.h symbol.h tree.h input.h input.c
 
@@ -44,7 +44,7 @@ output.o: config.h acme.h alu.h cpu.h dynabuf.h global.h input.h tree.h output.h
 
 platform.o: config.h platform.h platform.c
 
-pseudoopcodes.o: acme.h alu.h input.h macro.h output.h pseudoopcodes.h pseudoopcodes.c
+pseudoopcodes.o: acme.h alu.h global.h input.h macro.h output.h symbol.h pseudoopcodes.h pseudoopcodes.c
 
 section.o: config.h dynabuf.h global.h symbol.h tree.h section.h section.c
 

src/Makefile.mingw

@@ -35,7 +35,7 @@ encoding.o: config.h alu.h acme.h dynabuf.h global.h output.h input.h tree.h enc
 
 flow.o: config.h acme.h alu.h dynabuf.h global.h input.h mnemo.h symbol.h tree.h flow.h flow.c
 
-global.o: config.h platform.h acme.h cpu.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
+global.o: config.h platform.h acme.h cpu.h dynabuf.h encoding.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
 
 input.o: config.h alu.h dynabuf.h global.h section.h symbol.h tree.h input.h input.c
 
@@ -47,7 +47,7 @@ output.o: config.h acme.h alu.h cpu.h dynabuf.h global.h input.h tree.h output.h
 
 platform.o: config.h platform.h platform.c
 
-pseudoopcodes.o: acme.h alu.h input.h macro.h output.h pseudoopcodes.h pseudoopcodes.c
+pseudoopcodes.o: acme.h alu.h global.h input.h macro.h output.h symbol.h pseudoopcodes.h pseudoopcodes.c
 
 section.o: config.h dynabuf.h global.h symbol.h tree.h section.h section.c
 

src/Makefile.riscos

@@ -1,4 +1,5 @@
 CFLAGS		= -O3 -mthrowback -mlibscl -Wall -Wstrict-prototypes
+LINKFLAGS	= -mlibscl -Wall
 LIBS		= -lm
 CC		= gcc
 RM		= rm
@@ -13,7 +14,7 @@ OBJS		= acme.o alu.o cliargs.o cpu.o dynabuf.o encoding.o flow.o global.o input.
 all: $(PROGS)
 
 acme: $(OBJS)
-	$(CC) $(CFLAGS) -o !Unsqueezed $(OBJS) $(LIBS)
+	$(CC) $(LINKFLAGS) -o !Unsqueezed $(OBJS) $(LIBS)
 	Squeeze -f -v !Unsqueezed !ACME.!RunImage
 
 acme.o: config.h platform.h acme.h alu.h cpu.h dynabuf.h encoding.h flow.h global.h input.h macro.h mnemo.h output.h pseudoopcodes.h section.h symbol.h version.h acme.h acme.c
@@ -30,7 +31,7 @@ encoding.o: config.h alu.h acme.h dynabuf.h global.h output.h input.h tree.h enc
 
 flow.o: config.h acme.h alu.h dynabuf.h global.h input.h mnemo.h symbol.h tree.h flow.h flow.c
 
-global.o: config.h platform.h acme.h cpu.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
+global.o: config.h platform.h acme.h cpu.h dynabuf.h encoding.h input.h macro.h pseudoopcodes.h section.h symbol.h global.h global.c
 
 input.o: config.h alu.h dynabuf.h global.h section.h symbol.h tree.h input.h input.c
 
@@ -42,7 +43,7 @@ output.o: config.h acme.h alu.h cpu.h dynabuf.h global.h input.h tree.h output.h
 
 platform.o: config.h platform.h platform.c
 
-pseudoopcodes.o: acme.h alu.h input.h macro.h output.h pseudoopcodes.h pseudoopcodes.c
+pseudoopcodes.o: acme.h alu.h global.h input.h macro.h output.h symbol.h pseudoopcodes.h pseudoopcodes.c
 
 section.o: config.h dynabuf.h global.h symbol.h tree.h section.h section.c
 

src/_amiga.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Platform specific stuff (in this case, for AmigaOS)
@@ -13,7 +13,10 @@
 #define PLATFORM_INIT
 
 // convert UNIX-style pathname to Amiga-style pathname (no change)
-#define PLATFORM_CONVERTPATHCHAR(a)	(a)
+//#define PLATFORM_CONVERTPATH(p)
+
+// directory separator for include paths
+#define DIRECTORY_SEPARATOR	'\0'	// actually '/', but paths ending on ':' are ok, so auto-adding '/' is bad)
 
 // string containing the prefix for accessing files from the library tree
 #define PLATFORM_LIBPREFIX		"progdir:acme_lib/"

src/_dos.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Platform specific stuff (in this case, for DOS, OS/2 and Windows)
@@ -34,14 +34,17 @@ void DOS_entry(void)
 }
 
 
-// convert UNIX-style pathname character to DOS-style pathname character
-char DOS_convert_path_char(char byte)
+// convert UNIX-style pathname to DOS-style pathname
+void DOS_convert_path(char *p)
 {
-	if (byte == '/')
-		return '\\';
-	if (byte == '\\')
-		return '/';
-	return byte;
+	while (*p) {
+		if (*p == '/') {
+			*p = '\\';
+		} else if (*p == '\\') {
+			*p = '/';
+		}
+		++p;
+	}
 }
 
 

src/_dos.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Platform specific stuff (in this case, for DOS, OS/2 and Windows)
@@ -16,7 +16,10 @@
 #define PLATFORM_INIT		DOS_entry()
 
 // convert UNIX-style pathname to DOS-style pathname
-#define PLATFORM_CONVERTPATHCHAR(a)	DOS_convert_path_char(a)
+#define PLATFORM_CONVERTPATH(p)	DOS_convert_path(p)
+
+// directory separator for include paths
+#define DIRECTORY_SEPARATOR	'\\'
 
 // string containing the prefix for accessing files from the library tree
 #define PLATFORM_LIBPREFIX	DOS_lib_prefix
@@ -54,8 +57,8 @@ extern char *DOS_lib_prefix;	// header string of library tree
 
 // used as PLATFORM_INIT: reads "ACME" environment variable
 extern void DOS_entry(void);
-// Convert UNIX-style pathname character to DOS-style pathname character
-extern char DOS_convert_path_char(char);
+// Convert UNIX-style pathname to DOS-style pathname
+extern void DOS_convert_path(char *p);
 
 
 #endif

src/_riscos.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Platform specific stuff (in this case, for RISC OS)
@@ -23,7 +23,7 @@
 
 
 // variables
-int	RISCOS_flags	= 0;	// used to store platform-specific flags
+bits	RISCOS_flags	= 0;	// used to store platform-specific flags
 
 
 // exit handler: if throwback was used, de-register now
@@ -46,17 +46,20 @@ void RISCOS_entry(void)
 
 
 // convert UNIX-style pathname to RISC OS-style pathname
-char RISCOS_convert_path_char(char byte)
+void RISCOS_convert_path(char *p)
 {
-	if (byte == '.')
-		return '/';
-	if (byte == '/')
-		return '.';
-	if (byte == '?')
-		return '#';
-	if (byte == '#')
-		return '?';
-	return byte;
+	while (*p) {
+		if (*p == '.') {
+			*p = '/';
+		} else if (*p == '/') {
+			*p = '.';
+		} else if (*p == '?') {
+			*p = '#';
+		} else if (*p == '#') {
+			*p = '?';
+		}
+		++p;
+	}
 }
 
 

src/_riscos.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Platform specific stuff (in this case, for RISC OS)
@@ -15,7 +15,10 @@
 #define PLATFORM_INIT			RISCOS_entry()
 
 // convert UNIX-style pathname to RISC OS-style pathname
-#define PLATFORM_CONVERTPATHCHAR(a)	RISCOS_convert_path_char(a)
+#define PLATFORM_CONVERTPATH(path)	RISCOS_convert_path(path)
+
+// directory separator for include paths
+#define DIRECTORY_SEPARATOR	'\0'	// actually '.', but paths ending on ':' are ok, so auto-adding '.' is bad)
 
 // string containing the prefix for accessing files from the library tree
 #define PLATFORM_LIBPREFIX	"ACME_Lib:"
@@ -55,7 +58,7 @@ do {				\
 
 
 // variables
-extern int	RISCOS_flags;	// Holds platform-specific flags
+extern bits	RISCOS_flags;	// Holds platform-specific flags
 #define RISCOSFLAG_THROWBACK	(1u << 0)	// use throwback protocol
 #define RISCOSFLAG_THROWN	(1u << 1)	// throwback is active
 
@@ -63,11 +66,11 @@ extern int	RISCOS_flags;	// Holds platform-specific flags
 // used as PLATFORM_INIT: registers exit handler
 extern void RISCOS_entry(void);
 // convert UNIX-style pathname to RISC OS-style pathname
-extern char RISCOS_convert_path_char(char);
+extern void RISCOS_convert_path(char *path);
 // setting the created files' types
-extern void RISCOS_set_filetype(const char *, int);
+extern void RISCOS_set_filetype(const char *filename, int type);
 // use DDEUtils module's "Throwback" protocol
-extern void RISCOS_throwback(const char *, int);
+extern void RISCOS_throwback(const char *msg, int type);
 
 
 #endif

src/_std.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Platform specific stuff (in this case, for unknown OSes)
@@ -13,7 +13,10 @@
 #define PLATFORM_INIT		AnyOS_entry()
 
 // convert UNIX-style pathname to AnyOS-style pathname (no change)
-#define PLATFORM_CONVERTPATHCHAR(a)	(a)
+//#define PLATFORM_CONVERTPATH(p)
+
+// directory separator for include paths
+#define DIRECTORY_SEPARATOR	'/'
 
 // string containing the prefix for accessing files from the library tree
 #define PLATFORM_LIBPREFIX	AnyOS_lib_prefix

src/acme.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 //
 // This program is free software; you can redistribute it and/or modify
 // it under the terms of the GNU General Public License as published by
@@ -62,9 +62,15 @@ static const char	arg_vicelabels[]	= "VICE labels filename";
 #define OPTION_VERSION		"version"
 #define OPTION_MSVC		"msvc"
 #define OPTION_COLOR		"color"
+#define OPTION_FULLSTOP		"fullstop"
+#define OPTION_IGNORE_ZEROES	"ignore-zeroes"
+#define OPTION_STRICT_SEGMENTS	"strict-segments"
+#define OPTION_DIALECT		"dialect"
+#define OPTION_TEST		"test"
 // options for "-W"
 #define OPTIONWNO_LABEL_INDENT	"no-label-indent"
 #define OPTIONWNO_OLD_FOR	"no-old-for"
+#define OPTIONWNO_BIN_LEN	"no-bin-len"
 #define OPTIONWTYPE_MISMATCH	"type-mismatch"
 
 
@@ -123,23 +129,29 @@ static void show_help_and_exit(void)
 "  -l, --" OPTION_SYMBOLLIST " FILE  set symbol list file name\n"
 "      --" OPTION_LABELDUMP "        (old name for --" OPTION_SYMBOLLIST ")\n"
 "      --" OPTION_VICELABELS " FILE  set file name for label dump in VICE format\n"
-"      --" OPTION_SETPC " NUMBER     set program counter\n"
+"      --" OPTION_SETPC " VALUE      set program counter\n"
 "      --" OPTION_CPU " CPU          set target processor\n"
-"      --" OPTION_INITMEM " NUMBER   define 'empty' memory\n"
+"      --" OPTION_INITMEM " VALUE    define 'empty' memory\n"
 "      --" OPTION_MAXERRORS " NUMBER set number of errors before exiting\n"
 "      --" OPTION_MAXDEPTH " NUMBER  set recursion depth for macro calls and !src\n"
+"      --" OPTION_IGNORE_ZEROES "    do not determine number size by leading zeroes\n"
+"      --" OPTION_STRICT_SEGMENTS "  turn segment overlap warnings into errors\n"
 "  -vDIGIT                set verbosity level\n"
 "  -DSYMBOL=VALUE         define global symbol\n"
-// as long as there is only one -W option:
-#define OPTIONWNO_LABEL_INDENT	"no-label-indent"
+"  -I PATH/TO/DIR         add search path for input files\n"
+// TODO: replace these:
 "  -W" OPTIONWNO_LABEL_INDENT "      suppress warnings about indented labels\n"
-"  -W" OPTIONWNO_OLD_FOR "           suppress warnings about old \"!for\" syntax\n"
+"  -W" OPTIONWNO_OLD_FOR "           (old, use \"--dialect 0.94.8\" instead)\n"
+"  -W" OPTIONWNO_BIN_LEN "           suppress warnings about lengths of binary literals\n"
 "  -W" OPTIONWTYPE_MISMATCH "        enable type checking (warn about type mismatch)\n"
-// when there are more, use next line and add a separate function:
+// with this line and add a separate function:
 //"  -W                     show warning level options\n"
 "      --" OPTION_USE_STDOUT "       fix for 'Relaunch64' IDE (see docs)\n"
-"      --" OPTION_MSVC "             set output error message format to that of MS Visual Studio\n"
-"      --" OPTION_COLOR "            enable colored error output using ANSI escape codes\n"
+"      --" OPTION_MSVC "             output errors in MS VS format\n"
+"      --" OPTION_COLOR "            uses ANSI color codes for error output\n"
+"      --" OPTION_FULLSTOP "         use '.' as pseudo opcode prefix\n"
+"      --" OPTION_DIALECT " VERSION  behave like different version\n"
+"      --" OPTION_TEST "             enable experimental features\n"
 PLATFORM_OPTION_HELP
 "  -V, --" OPTION_VERSION "          show version and exit\n");
 	exit(EXIT_SUCCESS);
@@ -183,7 +195,7 @@ int ACME_finalize(int exit_code)
 
 	report_close(report);
 	if (symbollist_filename) {
-		fd = fopen(symbollist_filename, FILE_WRITETEXT);
+		fd = fopen(symbollist_filename, FILE_WRITETEXT);	// FIXME - what if filename is given via !sl in sub-dir? fix path!
 		if (fd) {
 			symbols_list(fd);
 			fclose(fd);
@@ -218,7 +230,7 @@ static void save_output_file(void)
 		fputs("No output file specified (use the \"-o\" option or the \"!to\" pseudo opcode).\n", stderr);
 		return;
 	}
-	fd = fopen(output_filename, FILE_WRITEBINARY);
+	fd = fopen(output_filename, FILE_WRITEBINARY);	// FIXME - what if filename is given via !to in sub-dir? fix path!
 	if (fd == NULL) {
 		fprintf(stderr, "Error: Cannot open output file \"%s\".\n",
 			output_filename);
@@ -229,12 +241,13 @@ static void save_output_file(void)
 }
 
 
-// perform a single pass. Returns number of "NeedValue" type errors.
-static int perform_pass(void)
+// increment pass number and perform a single pass
+static void perform_pass(void)
 {
 	FILE	*fd;
 	int	ii;
 
+	++pass.number;
 	// call modules' "pass init" functions
 	Output_passinit();	// disable output, PC undefined
 	cputype_passinit(default_cpu);	// set default cpu type
@@ -244,8 +257,9 @@ static int perform_pass(void)
 	encoding_passinit();	// set default encoding
 	section_passinit();	// set initial zone (untitled)
 	// init variables
-	pass_undefined_count = 0;	// no "NeedValue" errors yet
-	pass_real_errors = 0;	// no real errors yet
+	pass.undefined_count = 0;
+	//pass.needvalue_count = 0;	FIXME - use
+	pass.error_count = 0;
 	// Process toplevel files
 	for (ii = 0; ii < toplevel_src_count; ++ii) {
 		if ((fd = fopen(toplevel_sources[ii], FILE_READBINARY))) {
@@ -254,66 +268,63 @@ static int perform_pass(void)
 			fprintf(stderr, "Error: Cannot open toplevel file \"%s\".\n", toplevel_sources[ii]);
 			if (toplevel_sources[ii][0] == '-')
 				fprintf(stderr, "Options (starting with '-') must be given _before_ source files!\n");
- 			++pass_real_errors;
+ 			++pass.error_count;
 		}
 	}
-	if (pass_real_errors)
+	Output_end_segment();
+/*	TODO:
+	if --save-start is given, parse arg string
+	if --save-limit is given, parse arg string
+*/
+	if (pass.error_count)
 		exit(ACME_finalize(EXIT_FAILURE));
-	else
-		Output_end_segment();
-	return pass_undefined_count;
 }
 
 
 static struct report	global_report;
 // do passes until done (or errors occurred). Return whether output is ready.
-static int do_actual_work(void)
+static boolean do_actual_work(void)
 {
-	int		undefined_prev,	// "NeedValue" errors of previous pass
-			undefined_curr;	// "NeedValue" errors of current pass
+	int	undefs_before;	// number of undefined results in previous pass
 
 	report = &global_report;	// let global pointer point to something
 	report_init(report);	// we must init struct before doing passes
-	if (Process_verbosity > 1)
+	if (config.process_verbosity > 1)
 		puts("First pass.");
-	pass_count = 0;
-	undefined_curr = perform_pass();	// First pass
-	// now pretend there has been a pass before the first one
-	undefined_prev = undefined_curr + 1;
-	// As long as the number of "NeedValue" errors is decreasing but
-	// non-zero, keep doing passes.
-	while (undefined_curr && (undefined_curr < undefined_prev)) {
-		++pass_count;
-		undefined_prev = undefined_curr;
-		if (Process_verbosity > 1)
+	pass.complain_about_undefined = FALSE;	// disable until error pass needed
+	pass.number = -1;	// pre-init, will be incremented by perform_pass()
+	perform_pass();	// first pass
+	// pretend there has been a previous pass, with one more undefined result
+	undefs_before = pass.undefined_count + 1;
+	// keep doing passes as long as the number of undefined results keeps decreasing.
+	// stop on zero (FIXME - zero-check pass.needvalue_count instead!)
+	while (pass.undefined_count && (pass.undefined_count < undefs_before)) {
+		undefs_before = pass.undefined_count;
+		if (config.process_verbosity > 1)
 			puts("Further pass.");
-		undefined_curr = perform_pass();
+		perform_pass();
 	}
 	// any errors left?
-	if (undefined_curr == 0) {
+	if (pass.undefined_count == 0) {	// FIXME - use pass.needvalue_count instead!
 		// if listing report is wanted and there were no errors,
 		// do another pass to generate listing report
 		if (report_filename) {
-			if (Process_verbosity > 1)
+			if (config.process_verbosity > 1)
 				puts("Extra pass to generate listing report.");
 			if (report_open(report, report_filename) == 0) {
-				++pass_count;
 				perform_pass();
 				report_close(report);
 			}
 		}
-		return 1;
+		return TRUE;
 	}
 	// There are still errors (unsolvable by doing further passes),
 	// so perform additional pass to find and show them.
-	if (Process_verbosity > 1)
+	if (config.process_verbosity > 1)
 		puts("Extra pass needed to find error.");
-	// activate error output
-	ALU_optional_notdef_handler = Throw_error;
-
-	++pass_count;
+	pass.complain_about_undefined = TRUE;	// activate error output
 	perform_pass();	// perform pass, but now show "value undefined"
-	return 0;
+	return FALSE;
 }
 
 
@@ -327,30 +338,43 @@ static void keyword_to_dynabuf(const char keyword[])
 }
 
 
-// check output format (the output format tree must be set up at this point!)
-static void set_output_format(void)
+// set output format (the output format tree must be set up at this point!)
+static void set_output_format(const char format_name[])
 {
-	keyword_to_dynabuf(cliargs_safe_get_next("output format"));
-	if (outputfile_set_format()) {
-		fprintf(stderr, "%sUnknown output format (known formats are: %s).\n", cliargs_error, outputfile_formats);
-		exit(EXIT_FAILURE);
+	// caution, name may be NULL!
+	if (format_name) {
+		keyword_to_dynabuf(format_name);
+		if (!outputfile_set_format())
+			return;	// ok
+
+		fputs("Error: Unknown output format.\n", stderr);
+	} else {
+		fputs("Error: No output format specified.\n", stderr);
 	}
+	fprintf(stderr, "Supported formats are:\n\n\t%s\n\n", outputfile_formats);
+	exit(EXIT_FAILURE);
 }
 
 
-// check CPU type (the cpu type tree must be set up at this point!)
-static void set_starting_cpu(void)
+// set CPU type (the cpu type tree must be set up at this point!)
+static void set_starting_cpu(const char cpu_name[])
 {
 	const struct cpu_type	*new_cpu_type;
 
-	keyword_to_dynabuf(cliargs_safe_get_next("CPU type"));
-	new_cpu_type = cputype_find();
-	if (new_cpu_type) {
-		default_cpu = new_cpu_type;
+	// caution, name may be NULL!
+	if (cpu_name) {
+		keyword_to_dynabuf(cpu_name);
+		new_cpu_type = cputype_find();
+		if (new_cpu_type) {
+			default_cpu = new_cpu_type;
+			return;	// ok
+		}
+		fputs("Error: Unknown CPU type.\n", stderr);
 	} else {
-		fprintf(stderr, "%sUnknown CPU type (known types are: %s).\n", cliargs_error, cputype_names);
-		exit(EXIT_FAILURE);
+		fputs("Error: No CPU type specified.\n", stderr);
 	}
+	fprintf(stderr, "Supported types are:\n\n\t%s\n\n", cputype_names);
+	exit(EXIT_FAILURE);
 }
 
 
@@ -363,8 +387,8 @@ static void could_not_parse(const char strange[])
 
 // return signed long representation of string.
 // copes with hexadecimal if prefixed with "$", "0x" or "0X".
-// copes with octal if prefixed with "&".
-// copes with binary if prefixed with "%".
+// copes with octal if prefixed with "&".	FIXME - add "0o" prefix?
+// copes with binary if prefixed with "%".	FIXME - add "0b" prefix!
 // assumes decimal otherwise.
 static signed long string_to_number(const char *string)
 {
@@ -393,22 +417,24 @@ static signed long string_to_number(const char *string)
 
 
 // set program counter
-static void set_starting_pc(void)
+static void set_starting_pc(const char expression[])
 {
-	start_address = string_to_number(cliargs_safe_get_next("program counter"));
+	start_address = string_to_number(expression);
 	if ((start_address > -1) && (start_address < 65536))
 		return;
+
 	fprintf(stderr, "%sProgram counter out of range (0-0xffff).\n", cliargs_error);
 	exit(EXIT_FAILURE);
 }
 
 
 // set initial memory contents
-static void set_mem_contents(void)
+static void set_mem_contents(const char expression[])
 {
-	fill_value = string_to_number(cliargs_safe_get_next("initmem value"));
+	fill_value = string_to_number(expression);
 	if ((fill_value >= -128) && (fill_value <= 255))
 		return;
+
 	fprintf(stderr, "%sInitmem value out of range (0-0xff).\n", cliargs_error);
 	exit(EXIT_FAILURE);
 }
@@ -432,13 +458,64 @@ static void define_symbol(const char definition[])
 }
 
 
+struct dialect {
+	enum version	dialect;
+	const char	*version;
+	const char	*description;
+};
+struct dialect	dialects[]	= {
+	{VER_OLDEST_SUPPORTED,		"0.85",		"(the oldest version supported)"},
+	{VER_DEPRECATE_REALPC,		"0.86",		"\"!realpc\" gives a warning, \"!to\" wants a file format"},
+//	{VER_SHORTER_SETPC_WARNING,	"0.93",		"\"*=\" in offset assembly gives shorter warning but still switches off"},
+	{VER_RIGHTASSOCIATIVEPOWEROF,	"0.94.6",	"\"power of\" is now right-associative"},
+//	{VER_,				"0.94.7",	"empty code segments are no longer included in output file"},
+	{VER_DISABLED_OBSOLETE_STUFF,	"0.94.8",	"\"*=\" works inside \"!pseudopc\", disabled \"!cbm/!realpc/!subzone\""},
+	{VER_NEWFORSYNTAX,		"0.94.12",	"new \"!for\" syntax"},
+//	{VER_,				"0.95.2",	"changed ANC#8 from 0x2b to 0x0b"},
+	{VER_BACKSLASHESCAPING,		"0.97",		"backslash escaping and strings"},
+//	{VER_CURRENT,			"default",	"default"},
+	{VER_FUTURE,			"future",	"enable all experimental features"},
+	{0,				NULL,		NULL}	// NULLs terminate
+};
+
+// choose dialect (mimic behaviour of different version)
+static void set_dialect(const char version[])
+{
+	struct dialect	*dia;
+
+	// caution, version may be NULL!
+	if (version) {
+		// scan array
+		for (dia = dialects; dia->version; ++dia) {
+			if (strcmp(version, dia->version) == 0) {
+				config.wanted_version = dia->dialect;
+				return;	// found
+			}
+		}
+		fputs("Error: Unknown dialect specifier.\n", stderr);
+	} else {
+		fputs("Error: No dialect specified.\n", stderr);
+	}
+	// output table of possible versions and die
+	fputs(
+"Supported dialects are:\n"
+"\n"
+"\tdialect\t\tdescription\n"
+"\t-------\t\t-----------\n", stderr);
+	for (dia = dialects; dia->version; ++dia)
+		fprintf(stderr, "\t%s\t\t%s\n", dia->version, dia->description);
+	fputc('\n', stderr);
+	exit(EXIT_FAILURE);
+}
+
+
 // handle long options (like "--example"). Return unknown string.
 static const char *long_option(const char *string)
 {
 	if (strcmp(string, OPTION_HELP) == 0)
 		show_help_and_exit();
 	else if (strcmp(string, OPTION_FORMAT) == 0)
-		set_output_format();
+		set_output_format(cliargs_get_next());	// NULL is ok (handled like unknown)
 	else if (strcmp(string, OPTION_OUTFILE) == 0)
 		output_filename = cliargs_safe_get_next(name_outfile);
 	else if (strcmp(string, OPTION_LABELDUMP) == 0)	// old
@@ -450,24 +527,35 @@ static const char *long_option(const char *string)
 	else if (strcmp(string, OPTION_REPORT) == 0)
 		report_filename = cliargs_safe_get_next(arg_reportfile);
 	else if (strcmp(string, OPTION_SETPC) == 0)
-		set_starting_pc();
+		set_starting_pc(cliargs_safe_get_next("program counter"));
 	else if (strcmp(string, OPTION_CPU) == 0)
-		set_starting_cpu();
+		set_starting_cpu(cliargs_get_next());	// NULL is ok (handled like unknown)
 	else if (strcmp(string, OPTION_INITMEM) == 0)
-		set_mem_contents();
+		set_mem_contents(cliargs_safe_get_next("initmem value"));
 	else if (strcmp(string, OPTION_MAXERRORS) == 0)
-		max_errors = string_to_number(cliargs_safe_get_next("maximum error count"));
+		config.max_errors = string_to_number(cliargs_safe_get_next("maximum error count"));
 	else if (strcmp(string, OPTION_MAXDEPTH) == 0)
 		macro_recursions_left = (source_recursions_left = string_to_number(cliargs_safe_get_next("recursion depth")));
 //	else if (strcmp(string, "strictsyntax") == 0)
 //		strict_syntax = TRUE;
 	else if (strcmp(string, OPTION_USE_STDOUT) == 0)
-		msg_stream = stdout;
+		config.msg_stream = stdout;
 	else if (strcmp(string, OPTION_MSVC) == 0)
-		format_msvc = TRUE;
-	PLATFORM_LONGOPTION_CODE
+		config.format_msvc = TRUE;
+	else if (strcmp(string, OPTION_FULLSTOP) == 0)
+		config.pseudoop_prefix = '.';
+	else if (strcmp(string, OPTION_IGNORE_ZEROES) == 0)
+		config.honor_leading_zeroes = FALSE;
+	else if (strcmp(string, OPTION_STRICT_SEGMENTS) == 0)
+		config.segment_warning_is_error = TRUE;
+	else if (strcmp(string, OPTION_DIALECT) == 0)
+		set_dialect(cliargs_get_next());	// NULL is ok (handled like unknown)
+	else if (strcmp(string, OPTION_TEST) == 0) {
+		config.wanted_version = VER_FUTURE;
+		config.test_new_features = TRUE;
+	} PLATFORM_LONGOPTION_CODE
 	else if (strcmp(string, OPTION_COLOR) == 0)
-		format_color = TRUE;
+		config.format_color = TRUE;
 	else if (strcmp(string, OPTION_VERSION) == 0)
 		show_version(TRUE);
 	else
@@ -484,24 +572,31 @@ static char short_option(const char *argument)
 		case 'D':	// "-D" define constants
 			define_symbol(argument + 1);
 			goto done;
+		case 'f':	// "-f" selects output format
+			set_output_format(cliargs_get_next());	// NULL is ok (handled like unknown)
+			break;
 		case 'h':	// "-h" shows help
 			show_help_and_exit();
-		case 'f':	// "-f" selects output format
-			set_output_format();
+			break;
+		case 'I':	// "-I" adds an include directory
+			if (argument[1])
+				includepaths_add(argument + 1);
+			else
+				includepaths_add(cliargs_safe_get_next("include path"));
+			goto done;
+		case 'l':	// "-l" selects symbol list filename
+			symbollist_filename = cliargs_safe_get_next(arg_symbollist);
 			break;
 		case 'o':	// "-o" selects output filename
 			output_filename = cliargs_safe_get_next(name_outfile);
 			break;
-		case 'l':	// "-l" selects symbol list filename
-			symbollist_filename = cliargs_safe_get_next(arg_symbollist);
-			break;
 		case 'r':	// "-r" selects report filename
 			report_filename = cliargs_safe_get_next(arg_reportfile);
 			break;
 		case 'v':	// "-v" changes verbosity
-			++Process_verbosity;
+			++config.process_verbosity;
 			if ((argument[1] >= '0') && (argument[1] <= '9'))
-				Process_verbosity = *(++argument) - '0';
+				config.process_verbosity = *(++argument) - '0';
 			break;
 		// platform specific switches are inserted here
 			PLATFORM_SHORTOPTION_CODE
@@ -510,13 +605,16 @@ static char short_option(const char *argument)
 			break;
 		case 'W':	// "-W" tunes warning level
 			if (strcmp(argument + 1, OPTIONWNO_LABEL_INDENT) == 0) {
-				warn_on_indented_labels = FALSE;
+				config.warn_on_indented_labels = FALSE;
 				goto done;
 			} else if (strcmp(argument + 1, OPTIONWNO_OLD_FOR) == 0) {
-				warn_on_old_for = FALSE;
+				config.wanted_version = VER_NEWFORSYNTAX - 1;
+				goto done;
+			} else if (strcmp(argument + 1, OPTIONWNO_BIN_LEN) == 0) {
+				config.warn_bin_mask = 0;
 				goto done;
 			} else if (strcmp(argument + 1, OPTIONWTYPE_MISMATCH) == 0) {
-				warn_on_type_mismatch = TRUE;
+				config.warn_on_type_mismatch = TRUE;
 				goto done;
 			} else {
 				fprintf(stderr, "%sUnknown warning level.\n", cliargs_error);
@@ -536,28 +634,20 @@ done:
 // guess what
 int main(int argc, const char *argv[])
 {
+	config_default(&config);
 	// if called without any arguments, show usage info (not full help)
 	if (argc == 1)
 		show_help_and_exit();
-	msg_stream = stderr;
 	cliargs_init(argc, argv);
-	DynaBuf_init();	// inits *global* dynamic buffer - important, so first
-	// Init platform-specific stuff.
-	// For example, this could read the library path from an
-	// environment variable, which in turn may need DynaBuf already.
+	// init platform-specific stuff.
+	// this may read the library path from an environment variable.
 	PLATFORM_INIT;
-	// prepare a buffer large enough to hold pointers to "-D" switch values
-//	cli_defines = safe_malloc(argc * sizeof(*cli_defines));
 	// handle command line arguments
 	cliargs_handle_options(short_option, long_option);
 	// generate list of files to process
 	cliargs_get_rest(&toplevel_src_count, &toplevel_sources, "No top level sources given");
-	// Init modules (most of them will just build keyword trees)
-	ALU_init();
-	Macro_init();
-	Mnemo_init();
-	Output_init(fill_value);
-	pseudoopcodes_init();	// setup keyword tree for pseudo opcodes
+	// init output buffer
+	Output_init(fill_value, config.test_new_features);
 	if (do_actual_work())
 		save_output_file();
 	return ACME_finalize(EXIT_SUCCESS);	// dump labels, if wanted

src/acme.h

@@ -12,8 +12,8 @@
 
 // Variables
 extern const char	*symbollist_filename;
-extern const char	*output_filename;
-extern const char	*report_filename;
+extern const char	*output_filename;	// TODO - put in "part" struct
+extern const char	*report_filename;	// TODO - put in "part" struct
 // maximum recursion depth for macro calls and "!source"
 extern signed long	macro_recursions_left;
 extern signed long	source_recursions_left;

src/alu.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // ALU stuff (the expression parser)
@@ -10,67 +10,68 @@
 #include "config.h"
 
 
+struct op;
+struct dynabuf;
+struct type {
+	const char	*name;
+	boolean		(*is_defined)(const struct object *self);
+	boolean		(*differs)(const struct object *self, const struct object *other);
+	void		(*assign)(struct object *self, const struct object *new_value, boolean accept_change);
+	void		(*monadic_op)(struct object *self, const struct op *op);
+	void		(*dyadic_op)(struct object *self, const struct op *op, struct object *other);
+	void		(*fix_result)(struct object *self);
+	void		(*print)(const struct object *self, struct dynabuf *db);
+};
+extern struct type	type_number;
+extern struct type	type_list;
+extern struct type	type_string;
+
+struct expression {
+	struct object	result;
+	boolean		is_empty;		// nothing parsed (first character was a delimiter)
+	int		open_parentheses;	// number of parentheses still open
+	boolean		is_parenthesized;	// whole expression was in parentheses (indicating indirect addressing)
+	// TODO - how to return reserved cpu constant (register names)?
+};
+
+
 // constants
 
-// meaning of bits in "flags" of struct result:	TODO - this is only for future "number" result type!
-#define MVALUE_IS_FP	(1u << 8)	// floating point value
-#define MVALUE_INDIRECT	(1u << 7)	// needless parentheses indicate use of indirect addressing modes
-#define MVALUE_EXISTS	(1u << 6)	// 0: expression was empty. 1: there was *something* to parse.	TODO - get rid of this, make "nothing" its own result type instead!
-#define MVALUE_UNSURE	(1u << 5)	// value once was related to undefined
-// expression. Needed for producing the same addresses in all passes; because in
-// the first pass there will almost for sure be labels that are undefined, you
-// can't simply get the addressing mode from looking at the parameter's value.
-#define MVALUE_DEFINED	(1u << 4)	// 0: undefined expression (value will be zero). 1: known result
-#define MVALUE_ISBYTE	(1u << 3)	// value is guaranteed to fit in one byte
-#define MVALUE_FORCE24	(1u << 2)	// value usage forces 24-bit usage
-#define MVALUE_FORCE16	(1u << 1)	// value usage forces 16-bit usage
-#define MVALUE_FORCE08	(1u << 0)	// value usage forces 8-bit usage
-#define MVALUE_FORCEBITS	(MVALUE_FORCE08|MVALUE_FORCE16|MVALUE_FORCE24)
-#define MVALUE_GIVEN	(MVALUE_DEFINED | MVALUE_EXISTS)	// bit mask for fixed values (defined and existing)
-
-
-// create dynamic buffer, operator/function trees and operator/operand stacks
-extern void ALU_init(void);
-// function pointer for "value undefined" error output.
-// set to NULL to suppress those errors,
-// set to Throw_error to show them.
-extern void (*ALU_optional_notdef_handler)(const char *);
-
+// flag bits in number struct:
+#define NO_FORCE_BIT		0		// (just to make source more readable)
+#define NUMBER_FORCES_8		(1u << 0)	// value usage forces 8-bit usage
+#define NUMBER_FORCES_16	(1u << 1)	// value usage forces 16-bit usage
+#define NUMBER_FORCES_24	(1u << 2)	// value usage forces 24-bit usage
+#define NUMBER_FORCEBITS	(NUMBER_FORCES_8 | NUMBER_FORCES_16 | NUMBER_FORCES_24)
+#define NUMBER_FITS_BYTE	(1u << 3)	// value is guaranteed to fit in one byte
+#define NUMBER_EVER_UNDEFINED	(1u << 4)	// value once was related to
+	// undefined expression. Needed for producing the same addresses in all
+	// passes; because in the first pass there will almost for sure be
+	// labels that are undefined, we can't simply get the addressing mode
+	// from looking at the parameter's value.	FIXME - rename to TAINTED :)
 
+/*
 // FIXME - replace all the functions below with a single one using a "flags" arg!
-/* its return value would then be:
-enum expression_result {
-	EXRE_ERROR,	// error (has been reported, so skip remainder of statement)
-	EXRE_NOTHING,	// next char after space was comma or end-of-statement
-	EXRE_NUMBER,	// int or float (what is returned by the current functions)
-	EXRE_STRING,	// TODO
-	EXRE_RESERVED,	// reserved cpu constant (register names), TODO
-	EXRE_LIST	// TODO
-};
+// its return value would then be "error"/"ok".
 // input flags:
-#define ACCEPT_EMPTY		(1u << 0)	// if not given, throws error
-#define ACCEPT_UNDEFINED	(1u << 1)	// if not given, undefined throws serious error
-//#define ACCEPT_INT		(1u << )	needed when strings come along!
+#define ACCEPT_UNDEFINED	(1u << 0)	// if not given, undefined throws serious error
+#define ACCEPT_INT		(1u << 1)
 #define ACCEPT_FLOAT		(1u << 2)	// if not given, floats are converted to integer
 #define ACCEPT_OPENPARENTHESIS	(1u << 3)	// if not given, throws syntax error
-//#define ACCEPT_STRING
-// do I need ACCEPT_INT and/or ACCEPT_ADDRESS?
+#define ACCEPT_STRING		(1u << 4)	// if not given, convert 1-char strings to int?
+#define ACCEPT_LIST		(1u << 5)
+// do I need ACCEPT_NONADDR and/or ACCEPT_ADDRESS?
 */
 
-// stores int value if given. Returns whether stored. Throws error if undefined.
-extern int ALU_optional_defined_int(intval_t *target);
-// returns int value (0 if result was undefined)
-extern intval_t ALU_any_int(void);
-// stores int value and flags (floats are transformed to int)
-extern void ALU_int_result(struct result *intresult);
+// stores int value (0 if result was undefined)
+extern void ALU_any_int(intval_t *target);
 // stores int value and flags (floats are transformed to int)
 // if result was undefined, serious error is thrown
-extern void ALU_defined_int(struct result *intresult);
-// stores int value and flags, allowing for one '(' too many (x-indirect addr).
-// returns number of additional '(' (1 or 0).
-extern int ALU_liberal_int(struct result *intresult);
-// stores value and flags (result may be either int or float)
-extern void ALU_any_result(struct result *result);
+extern void ALU_defined_int(struct number *intresult);
+// stores int value and flags, allowing for "paren" '(' too many (x-indirect addr).
+extern void ALU_addrmode_int(struct expression *expression, int paren);
+// stores resulting object
+extern void ALU_any_result(struct object *result);
 
 
 #endif

src/config.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Configuration
@@ -7,21 +7,64 @@
 #define config_H
 
 
+// make sure the enum below works with strange compilers, too:
+#ifdef FALSE
+#undef FALSE
+#endif
+#ifdef TRUE
+#undef TRUE
+#endif
 // types
+typedef enum { FALSE = 0, TRUE }	boolean;	// yes, I could include <stdbool.h>, but this source should work with ancient compilers as well...
+typedef unsigned int	bits;
 typedef unsigned int	scope_t;
 typedef signed long	intval_t;	// at least 32 bits
 typedef unsigned long	uintval_t;	// just for logical shift right
-// result structure type definition with support for floating point
-// future result types: EMPTY, UNDEFINED, INT, FLOAT (, STRING)
-struct result {	// either int or float
-	int	flags;	// expression flags
+enum numtype {
+	NUMTYPE_UNDEFINED,
+	NUMTYPE_INT,
+	NUMTYPE_FLOAT,
+};
+
+// structure for ints/floats
+struct number {
+	enum numtype	ntype;
+	bits		flags;	// FITS_IN_BYTE etc. (see alu.h)
 	union {
 		intval_t	intval;	// integer value
 		double		fpval;	// floating point value
-	} val;	// Expression value
+	} val;
 	int	addr_refs;	// address reference count (only look at this if value is DEFINED)
 };
 
+struct type;
+struct string;
+struct listitem;
+// structure for ints/floats/lists/strings (anything that can be assigned to symbol)
+struct object {
+	struct type	*type;
+	union {
+		struct number	number;
+		struct string	*string;
+		struct listitem	*listhead;
+	} u;
+};
+struct string {
+	int 	length;
+	int 	refs;
+	char	payload[1];	// real structs are malloc'd to correct size
+};
+struct listitem {
+	struct listitem	*next;
+	struct listitem	*prev;
+	union {
+		struct {
+			int	length;	// this does not include the head element
+			int	refs;
+		} listinfo;			// if item is list head
+		struct object	payload;	// if item is not list head
+	} u;
+};
 
 // debugging flag, should be undefined in release version
 // #define FDEBUG
@@ -38,11 +81,6 @@ struct result {	// either int or float
 #ifndef NULL
 #define NULL	((void *) 0)
 #endif
-// Boolean values
-#ifndef FALSE
-#define FALSE	0
-#define TRUE	(!FALSE)
-#endif
 
 
 #endif

src/cpu.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // CPU type stuff
@@ -7,7 +7,7 @@
 #include "config.h"
 #include "alu.h"
 #include "dynabuf.h"
-#include "global.h"	// FIXME - remove when no longer needed
+#include "global.h"
 #include "input.h"
 #include "mnemo.h"
 #include "output.h"
@@ -17,36 +17,37 @@
 // constants
 static struct cpu_type	cpu_type_6502	= {
 	keyword_is_6502_mnemo,
-	CPUFLAG_INDIRECTJMPBUGGY,	// JMP ($xxFF) is buggy
+	CPUFLAG_WARN_ABOUT_FF_PTR | CPUFLAG_INDIRECTJMPBUGGY,	// warn about "XYZ ($ff),y" and "jmp ($XYff)"
 	234	// !align fills with "NOP"
 };
-static struct cpu_type	cpu_type_6510	= {
-	keyword_is_6510_mnemo,
-	CPUFLAG_INDIRECTJMPBUGGY | CPUFLAG_8B_AND_AB_NEED_0_ARG,	// JMP ($xxFF) is buggy, ANE/LXA #$xx are unstable unless arg is $00
+static struct cpu_type	cpu_type_nmos6502	= {
+	keyword_is_nmos6502_mnemo,
+	CPUFLAG_WARN_ABOUT_FF_PTR | CPUFLAG_INDIRECTJMPBUGGY | CPUFLAG_8B_AND_AB_NEED_0_ARG,	// ANE/LXA #$xx are unstable unless arg is $00
 	234	// !align fills with "NOP"
 };
 static struct cpu_type	cpu_type_c64dtv2	= {
 	keyword_is_c64dtv2_mnemo,
-	CPUFLAG_INDIRECTJMPBUGGY | CPUFLAG_8B_AND_AB_NEED_0_ARG,	// JMP ($xxFF) is buggy, ANE/LXA #$xx are unstable unless arg is $00
+	CPUFLAG_WARN_ABOUT_FF_PTR | CPUFLAG_INDIRECTJMPBUGGY | CPUFLAG_8B_AND_AB_NEED_0_ARG,
 	234	// !align fills with "NOP"
 };
 static struct cpu_type	cpu_type_65c02	= {
 	keyword_is_65c02_mnemo,
-	0,	// no flags
+	CPUFLAG_WARN_ABOUT_FF_PTR,	// from WDC docs
 	234	// !align fills with "NOP"
 };
 static struct cpu_type	cpu_type_r65c02	= {
 	keyword_is_r65c02_mnemo,
-	0,	// no flags
+	CPUFLAG_WARN_ABOUT_FF_PTR,	// from WDC docs
 	234	// !align fills with "NOP"
 };
 static struct cpu_type	cpu_type_w65c02	= {
 	keyword_is_w65c02_mnemo,
-	0,	// no flags
+	CPUFLAG_WARN_ABOUT_FF_PTR,	// from WDC docs
 	234	// !align fills with "NOP"
 };
 static struct cpu_type	cpu_type_65816	= {
 	keyword_is_65816_mnemo,
+	// TODO - what about CPUFLAG_WARN_ABOUT_FF_PTR? only needed for old opcodes in emulation mode!
 	CPUFLAG_SUPPORTSLONGREGS,	// allows A and XY to be 16bits wide
 	234	// !align fills with "NOP"
 };
@@ -60,24 +61,31 @@ static struct cpu_type	cpu_type_4502	= {
 	CPUFLAG_DECIMALSUBTRACTBUGGY,	// SBC does not work reliably in decimal mode
 	234	// !align fills with "NOP"
 };
+static struct cpu_type	cpu_type_m65	= {
+	keyword_is_m65_mnemo,
+	CPUFLAG_WARN_ABOUT_FF_PTR,	// TODO - remove this? check datasheets/realhw!
+	234	// !align fills with "NOP"
+};
 
 
 // variables
 
 // predefined stuff
-static struct ronode	*cputype_tree	= NULL;
-static struct ronode	cputype_list[]	= {
-#define KNOWN_TYPES	"'6502', '6510', '65c02', 'r65c02', 'w65c02', '65816', '65ce02', '4502', 'c64dtv2'"	// shown in CLI error message for unknown types
+static struct ronode	cputype_tree[]	= {
+	PREDEF_START,
+#define KNOWN_TYPES	"'6502', 'nmos6502', '6510', '65c02', 'r65c02', 'w65c02', '65816', '65ce02', '4502', 'm65', 'c64dtv2'"	// shown in CLI error message for unknown types
 //	PREDEFNODE("z80",		&cpu_type_Z80),
 	PREDEFNODE("6502",		&cpu_type_6502),
-	PREDEFNODE("6510",		&cpu_type_6510),
+	PREDEFNODE("nmos6502",		&cpu_type_nmos6502),
+	PREDEFNODE("6510",		&cpu_type_nmos6502),
 	PREDEFNODE("65c02",		&cpu_type_65c02),
 	PREDEFNODE("r65c02",		&cpu_type_r65c02),
 	PREDEFNODE("w65c02",		&cpu_type_w65c02),
 	PREDEFNODE("65816",		&cpu_type_65816),
 	PREDEFNODE("65ce02",		&cpu_type_65ce02),
 	PREDEFNODE("4502",		&cpu_type_4502),
-	PREDEFLAST("c64dtv2",		&cpu_type_c64dtv2),
+	PREDEFNODE("m65",		&cpu_type_m65),
+	PREDEF_END("c64dtv2",		&cpu_type_c64dtv2),
 	//    ^^^^ this marks the last element
 };
 const char	cputype_names[]	= KNOWN_TYPES;	// string to show if cputype_find() returns NULL
@@ -87,9 +95,6 @@ const struct cpu_type *cputype_find(void)
 {
 	void	*node_body;
 
-	// make sure tree is initialised
-	if (cputype_tree == NULL)
-		Tree_add_table(&cputype_tree, cputype_list);
 	// perform lookup
 	if (!Tree_easy_scan(cputype_tree, &node_body, GlobalDynaBuf))
 		return NULL;
@@ -102,7 +107,7 @@ const struct cpu_type *cputype_find(void)
 // if cpu type and value don't match, complain instead.
 // FIXME - error message might be confusing if it is thrown not because of
 // initial change, but because of reverting back to old cpu type after "{}" block!
-void vcpu_check_and_set_reg_length(int *var, int make_long)
+void vcpu_check_and_set_reg_length(boolean *var, boolean make_long)
 {
 	if (((CPU_state.type->flags & CPUFLAG_SUPPORTSLONGREGS) == 0) && make_long)
 		Throw_error("Chosen CPU does not support long registers.");
@@ -116,4 +121,6 @@ void cputype_passinit(const struct cpu_type *cpu_type)
 {
 	// handle cpu type (default is 6502)
 	CPU_state.type = cpu_type ? cpu_type : &cpu_type_6502;
+	CPU_state.a_is_long = FALSE;	// short accu
+	CPU_state.xy_are_long = FALSE;	// short index regs
 }

src/cpu.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // CPU type stuff
@@ -14,20 +14,20 @@
 struct cpu_type {
 	// This function is not allowed to change GlobalDynaBuf
 	// because that's where the mnemonic is stored!
-	int	(*keyword_is_mnemonic)(int);
-	int	flags;	// see below for bit meanings
-	char	default_align_value;
+	boolean		(*keyword_is_mnemonic)(int);
+	bits		flags;	// see below for bit meanings
+	unsigned char	default_align_value;
 };
 #define	CPUFLAG_INDIRECTJMPBUGGY	(1u << 0)	// warn if "jmp ($xxff)" is assembled
 #define CPUFLAG_SUPPORTSLONGREGS	(1u << 1)	// allow "!al" and "!rl" pseudo opcodes
 #define CPUFLAG_8B_AND_AB_NEED_0_ARG	(1u << 2)	// warn if "ane/lxa #$xx" uses non-zero arg
 #define CPUFLAG_ISBIGENDIAN		(1u << 3)	// for 16/24/32-bit values, output msb first
 #define CPUFLAG_DECIMALSUBTRACTBUGGY	(1u << 4)	// warn if "sed" is assembled
-
+#define CPUFLAG_WARN_ABOUT_FF_PTR	(1u << 5)	// warn if MNEMO($ff) is assembled
 
 // if cpu type and value match, set register length variable to value.
 // if cpu type and value don't match, complain instead.
-extern void vcpu_check_and_set_reg_length(int *var, int make_long);
+extern void vcpu_check_and_set_reg_length(boolean *var, boolean make_long);
 // set default value for pass
 extern void cputype_passinit(const struct cpu_type *cpu_type);
 // lookup cpu type held in DynaBuf and return its struct pointer (or NULL on failure)

src/dynabuf.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Dynamic buffer stuff
@@ -21,10 +21,12 @@
 // initial size for global dynabuf
 // (as it holds macros, loop bodies, etc., make it large to begin with)
 #define GLOBALDYNABUF_INITIALSIZE	1024	// should be >0 (see above)
+// TODO - get rid of this, or move to global.c
 
 
 // Variables
-struct dynabuf	*GlobalDynaBuf;	// global dynamic buffer
+STRUCT_DYNABUF_REF(GlobalDynaBuf, GLOBALDYNABUF_INITIALSIZE);	// global dynamic buffer
+// TODO - get rid of this, or move to global.c
 
 
 // Functions
@@ -34,37 +36,42 @@ static void resize(struct dynabuf *db, size_t new_size)
 {
 	char	*new_buf;
 
+	//printf("Growing dynabuf to size %d.\n", new_size);
 	new_buf = realloc(db->buffer, new_size);
 	if (new_buf == NULL)
 		Throw_serious_error(exception_no_memory_left);
 	db->reserved = new_size;
 	db->buffer = new_buf;
 }
+// get buffer mem and fill in struct
+static void initstruct(struct dynabuf *db, size_t initial_size)
+{
+	//printf("dynabuf-init: %d.\n", initial_size);
+	if (initial_size < DYNABUF_MINIMUM_INITIALSIZE)
+		initial_size = DYNABUF_MINIMUM_INITIALSIZE;
+	db->size = 0;
+	db->reserved = initial_size;
+	db->buffer = malloc(initial_size);
+	if (db->buffer == NULL) {
+		// scream and die because there is not enough memory 
+		fputs("Error: No memory for dynamic buffer.\n", stderr);
+		exit(EXIT_FAILURE);
+	}
+}
 
 
 // Exported functions
 
-// Create and init a dynamic buffer and return pointer
-struct dynabuf *DynaBuf_create(int initial_size)
+// (ensure buffer is ready to use, then) clear dynamic buffer
+void dynabuf_clear(struct dynabuf *db)
 {
-	struct dynabuf	*db;
-
-	if (initial_size < DYNABUF_MINIMUM_INITIALSIZE)
-		initial_size = DYNABUF_MINIMUM_INITIALSIZE;
-	if ((db = malloc(sizeof(*db)))) {
-		db->size = 0;
-		db->reserved = initial_size;
-		db->buffer = malloc(initial_size);
-		if (db->buffer)
-			return db;	// if both pointers are != NULL, no error
-	}
-	// otherwise, complain
-	fputs("Error: No memory for dynamic buffer.\n", stderr);
-	exit(EXIT_FAILURE);
+	if (db->buffer == NULL)
+		initstruct(db, db->reserved);	// get initial buffer
+	db->size = 0;	// clear buffer
 }
 
 // Enlarge buffer
-void DynaBuf_enlarge(struct dynabuf *db)
+void dynabuf_enlarge(struct dynabuf *db)
 {
 	resize(db, MAKE_LARGER_THAN(db->reserved));
 }
@@ -109,7 +116,13 @@ static char *ensure_free_space(struct dynabuf *db, int size)
 void DynaBuf_to_lower(struct dynabuf *target, struct dynabuf *source)
 {
 	char	*read,
-		*write;
+		*write,
+		byte;
+
+	// if target has not been initialised yet, do it now
+	// (do not clear it unconditionally, because it may equal source!)
+	if (target->buffer == NULL)
+		initstruct(target, target->reserved);	// get initial buffer
 
 	// make sure target can take it
 	if (source->size > target->reserved)
@@ -117,17 +130,16 @@ void DynaBuf_to_lower(struct dynabuf *target, struct dynabuf *source)
 	// convert to lower case
 	read = source->buffer;	// CAUTION - ptr may change when buf grows!
 	write = target->buffer;	// CAUTION - ptr may change when buf grows!
-	while (*read)
-		*write++ = (*read++) | 32;
+	while ((byte = *read++)) {
+		// we want to keep underscore, so this check restricts:
+		if (byte <= 'Z')
+			byte |= 32;
+		*write++ = byte;
+	}
 	// Okay, so this method of converting to lowercase is lousy.
 	// But actually it doesn't matter, because only pre-defined
 	// keywords are converted, and all of those are plain
 	// old-fashioned 7-bit ASCII anyway. So I guess it'll do.
+	// FIXME - use BYTE_ macro from global.h
 	*write = '\0';	// terminate
 }
-
-// Initialisation - allocate global dynamic buffer
-void DynaBuf_init(void)
-{
-	GlobalDynaBuf = DynaBuf_create(GLOBALDYNABUF_INITIALSIZE);
-}

src/dynabuf.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Dynamic buffer stuff
@@ -8,14 +8,14 @@
 
 
 #include "config.h"
+#include <stdlib.h>	// for size_t
 
 
 // macros
-#define DYNABUF_CLEAR(db)		do {db->size = 0;} while (0)
 #define DYNABUF_APPEND(db, byte)	\
 do {					\
 	if (db->size == db->reserved)	\
-		DynaBuf_enlarge(db);	\
+		dynabuf_enlarge(db);	\
 	db->buffer[(db->size)++] = byte;\
 } while (0)
 // the next one is dangerous - the buffer location can change when a character
@@ -27,21 +27,28 @@ do {					\
 // dynamic buffer structure
 struct dynabuf {
 	char	*buffer;	// pointer to buffer
-	int	size;		// size of buffer's used portion
-	int	reserved;	// total size of buffer
+	size_t	size;		// size of buffer's used portion
+	size_t	reserved;	// total size of buffer
 };
+// new way of declaration/definition:
+// the small struct above is static, only the buffer itself gets malloc'd (on
+// first "clear").
+#define STRUCT_DYNABUF_REF(name, size)	struct dynabuf name[1]	= {{NULL, 0, size}}
+// the "[1]" makes sure the name refers to the address and not the struct
+// itself, so existing code where the name referred to a pointer does not need
+// to be changed.
 
 
 // variables
-extern struct dynabuf *GlobalDynaBuf;	// global dynamic buffer
+extern struct dynabuf	GlobalDynaBuf[1];	// global dynamic buffer
+// TODO - get rid of this, or move to global.c
 
 
-// create global DynaBuf (call once on program startup)
-extern void DynaBuf_init(void);
-// create (private) DynaBuf
-extern struct dynabuf *DynaBuf_create(int initial_size);
+// (ensure buffer is ready to use, then) clear dynamic buffer
+#define DYNABUF_CLEAR(db)	dynabuf_clear(db)	// TODO - remove old macro
+extern void dynabuf_clear(struct dynabuf *db);
 // call whenever buffer is too small
-extern void DynaBuf_enlarge(struct dynabuf *db);
+extern void dynabuf_enlarge(struct dynabuf *db);
 // return malloc'd copy of buffer contents
 extern char *DynaBuf_get_copy(struct dynabuf *db);
 // copy string to buffer (without terminator)

src/encoding.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Character encoding stuff
@@ -9,7 +9,7 @@
 #include "alu.h"
 #include "acme.h"
 #include "dynabuf.h"
-#include "global.h"	// FIXME - remove when no longer needed
+#include "global.h"
 #include "output.h"
 #include "input.h"
 #include "tree.h"
@@ -17,41 +17,41 @@
 
 // struct definition
 struct encoder {
-	char	(*fn)(char);
+	unsigned char	(*fn)(unsigned char);
 	// maybe add table pointer?
 };
 
 
 // variables
-static char		outermost_table[256];	// space for encoding table...
+static unsigned char	outermost_table[256];	// space for encoding table...
 const struct encoder	*encoder_current;	// gets set before each pass
-char			*encoding_loaded_table	= outermost_table;	// ...loaded from file
+unsigned char		*encoding_loaded_table	= outermost_table;	// ...loaded from file
 
 
 // encoder functions:
 
 
 // convert raw to raw (do not convert at all)
-static char encoderfn_raw(char byte)
+static unsigned char encoderfn_raw(unsigned char byte)
 {
 	return byte;
 }
 // convert raw to petscii
-static char encoderfn_pet(char byte)
+static unsigned char encoderfn_pet(unsigned char byte)
 {
-	if ((byte >= 'A') && (byte <= 'Z'))
-		return (char) (byte | 0x80);	// FIXME - check why SAS-C
-	if ((byte >= 'a') && (byte <= 'z'))	//	wants these casts.
-		return (char) (byte - 32);	//	There are more below.
+	if ((byte >= (unsigned char) 'A') && (byte <= (unsigned char) 'Z'))
+		return byte | 0x80;
+	if ((byte >= (unsigned char) 'a') && (byte <= (unsigned char) 'z'))
+		return byte - 32;
 	return byte;
 }
 // convert raw to C64 screencode
-static char encoderfn_scr(char byte)
+static unsigned char encoderfn_scr(unsigned char byte)
 {
-	if ((byte >= 'a') && (byte <= 'z'))
-		return (char) (byte - 96);	// shift uppercase down
-	if ((byte >= '[') && (byte <= '_'))
-		return (char) (byte - 64);	// shift [\]^_ down
+	if ((byte >= (unsigned char) 'a') && (byte <= (unsigned char) 'z'))
+		return byte - 96;	// shift uppercase down
+	if ((byte >= (unsigned char) '[') && (byte <= (unsigned char) '_'))
+		return byte - 64;	// shift [\]^_ down
 	if (byte == '`')
 		return 64;	// shift ` down
 	if (byte == '@')
@@ -59,9 +59,9 @@ static char encoderfn_scr(char byte)
 	return byte;
 }
 // convert raw to whatever is defined in table
-static char encoderfn_file(char byte)
+static unsigned char encoderfn_file(unsigned char byte)
 {
-	return encoding_loaded_table[(unsigned char) byte];
+	return encoding_loaded_table[byte];
 }
 
 
@@ -83,12 +83,12 @@ const struct encoder	encoder_file	= {
 
 
 // keywords for "!convtab" pseudo opcode
-static struct ronode	*encoder_tree	= NULL;	// tree to hold encoders
-static struct ronode	encoder_list[]	= {
+static struct ronode	encoder_tree[]	= {
+	PREDEF_START,
 //no!	PREDEFNODE("file",	&encoder_file),	"!ct file" is not needed; just use {} after initial loading of table!
-	PREDEFNODE(s_pet,	&encoder_pet),
-	PREDEFNODE(s_raw,	&encoder_raw),
-	PREDEFLAST(s_scr,	&encoder_scr),
+	PREDEFNODE("pet",	&encoder_pet),
+	PREDEFNODE("raw",	&encoder_raw),
+	PREDEF_END("scr",	&encoder_scr),
 	//    ^^^^ this marks the last element
 };
 
@@ -97,7 +97,7 @@ static struct ronode	encoder_list[]	= {
 
 
 // convert character using current encoding (exported for use by alu.c and pseudoopcodes.c)
-char encoding_encode_char(char byte)
+unsigned char encoding_encode_char(unsigned char byte)
 {
 	return encoder_current->fn(byte);
 }
@@ -109,17 +109,10 @@ void encoding_passinit(void)
 }
 
 // try to load encoding table from given file
-void encoding_load(char target[256], const char *filename)
+void encoding_load_from_file(unsigned char target[256], FILE *stream)
 {
-	FILE	*fd	= fopen(filename, FILE_READBINARY);
-
-	if (fd) {
-		if (fread(target, sizeof(char), 256, fd) != 256)
-			Throw_error("Conversion table incomplete.");
-		fclose(fd);
-	} else {
-		Throw_error(exception_cannot_open_input_file);
-	}
+	if (fread(target, sizeof(char), 256, stream) != 256)
+		Throw_error("Conversion table incomplete.");
 }
 
 // lookup encoder held in DynaBuf and return its struct pointer (or NULL on failure)
@@ -127,9 +120,6 @@ const struct encoder *encoding_find(void)
 {
 	void	*node_body;
 
-	// make sure tree is initialised
-	if (encoder_tree == NULL)
-		Tree_add_table(&encoder_tree, encoder_list);
 	// perform lookup
 	if (!Tree_easy_scan(encoder_tree, &node_body, GlobalDynaBuf)) {
 		Throw_error("Unknown encoding.");

src/encoding.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Character encoding stuff
@@ -7,23 +7,26 @@
 #define encoding_H
 
 
+#include <stdio.h>	// for FILE*
+
+
 //struct encoder;
-extern const struct encoder	*encoder_current;	// gets set before each pass
+extern const struct encoder	*encoder_current;	// gets set before each pass	TODO - set for each part
 extern const struct encoder	encoder_raw;
 extern const struct encoder	encoder_pet;
 extern const struct encoder	encoder_scr;
 extern const struct encoder	encoder_file;
-extern char			*encoding_loaded_table;	// ...loaded from file
+extern unsigned char		*encoding_loaded_table;	// ...loaded from file
 
 
 // prototypes
 
 // convert character using current encoding
-extern char encoding_encode_char(char byte);
+extern unsigned char encoding_encode_char(unsigned char byte);
 // set "raw" as default encoding
 extern void encoding_passinit(void);
 // try to load encoding table from given file
-extern void encoding_load(char target[256], const char *filename);
+extern void encoding_load_from_file(unsigned char target[256], FILE *stream);
 // lookup encoder held in DynaBuf and return its struct pointer (or NULL on failure)
 extern const struct encoder *encoding_find(void);
 

src/flow.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Flow control stuff (loops, conditional assembly etc.)
@@ -17,14 +17,39 @@
 #include "alu.h"
 #include "config.h"
 #include "dynabuf.h"
-#include "global.h"	// FIXME - remove when no longer needed
+#include "global.h"
 #include "input.h"
 #include "mnemo.h"
 #include "symbol.h"
 #include "tree.h"
 
 
-// helper functions for "!for" and "!do"
+// helper functions for if/ifdef/ifndef/else/for/do/while
+
+
+// parse symbol name and return if symbol has defined value (called by ifdef/ifndef)
+boolean check_ifdef_condition(void)
+{
+	scope_t		scope;
+	struct rwnode	*node;
+	struct symbol	*symbol;
+
+	// read symbol name
+	if (Input_read_scope_and_keyword(&scope) == 0)	// skips spaces before
+		return FALSE;	// there was an error, it has been reported, so return value is more or less meaningless anway
+
+	// look for it
+	Tree_hard_scan(&node, symbols_forest, scope, FALSE);
+	if (!node)
+		return FALSE;	// not found -> no, not defined
+
+	symbol = (struct symbol *) node->body;
+	symbol->has_been_read = TRUE;	// we did not really read the symbol's value, but checking for its existence still counts as "used it"
+	if (symbol->object.type == NULL)
+		Bug_found("ObjectHasNullType", 0);
+	return symbol->object.type->is_defined(&symbol->object);
+}
+
 
 // parse a loop body (TODO - also use for macro body?)
 static void parse_ram_block(struct block *block)
@@ -43,55 +68,96 @@ void flow_forloop(struct for_loop *loop)
 {
 	struct input	loop_input,
 			*outer_input;
-	struct result	loop_counter;
+	struct object	loop_counter;
 
 	// switching input makes us lose GotByte. But we know it's '}' anyway!
 	// set up new input
 	loop_input = *Input_now;	// copy current input structure into new
-	loop_input.source_is_ram = TRUE;	// set new byte source
+	loop_input.source = INPUTSRC_RAM;	// set new byte source
 	// remember old input
 	outer_input = Input_now;
 	// activate new input
 	// (not yet useable; pointer and line number are still missing)
 	Input_now = &loop_input;
+	// fix line number (not for block, but in case symbol handling throws errors)
+	Input_now->line_number = loop->block.start;
 	// init counter
-	loop_counter.flags = MVALUE_DEFINED | MVALUE_EXISTS;
-	loop_counter.val.intval = loop->counter.first;
-	loop_counter.addr_refs = loop->counter.addr_refs;
-	symbol_set_value(loop->symbol, &loop_counter, TRUE);
-	if (loop->old_algo) {
+	loop_counter.type = &type_number;
+	loop_counter.u.number.ntype = NUMTYPE_INT;
+	loop_counter.u.number.flags = 0;
+	loop_counter.u.number.val.intval = loop->counter.first;
+	loop_counter.u.number.addr_refs = loop->counter.addr_refs;
+	// CAUTION: next line does not have power to change symbol type, but if
+	// "symbol already defined" error is thrown, the type will still have
+	// been changed. this was done so the code below has a counter var.
+	symbol_set_object(loop->symbol, &loop_counter, POWER_CHANGE_VALUE);
+	// TODO: in versions before 0.97, force bit handling was broken
+	// in both "!set" and "!for":
+	// trying to change a force bit correctly raised an error, but
+	// in any case, ALL FORCE BITS WERE CLEARED in symbol. only
+	// cases like !set N=N+1 worked, because the force bit was
+	// taken from result.
+	// maybe support this behaviour via --dialect?
+	if (loop->force_bit)
+		symbol_set_force_bit(loop->symbol, loop->force_bit);
+	loop_counter = loop->symbol->object;	// update local copy with force bit
+	loop->symbol->has_been_read = TRUE;	// lock force bit
+	if (loop->use_old_algo) {
 		// old algo for old syntax:
 		// if count == 0, skip loop
 		if (loop->counter.last) {
 			do {
-				loop_counter.val.intval += loop->counter.increment;
-				symbol_set_value(loop->symbol, &loop_counter, TRUE);
+				loop_counter.u.number.val.intval += loop->counter.increment;
+				loop->symbol->object = loop_counter;	// overwrite whole struct, in case some joker has re-assigned loop counter var
 				parse_ram_block(&loop->block);
-			} while (loop_counter.val.intval < loop->counter.last);
+			} while (loop_counter.u.number.val.intval < loop->counter.last);
 		}
 	} else {
 		// new algo for new syntax:
 		do {
 			parse_ram_block(&loop->block);
-			loop_counter.val.intval += loop->counter.increment;
-			symbol_set_value(loop->symbol, &loop_counter, TRUE);
-		} while (loop_counter.val.intval != (loop->counter.last + loop->counter.increment));
+			loop_counter.u.number.val.intval += loop->counter.increment;
+			loop->symbol->object = loop_counter;	// overwrite whole struct, in case some joker has re-assigned loop counter var
+		} while (loop_counter.u.number.val.intval != (loop->counter.last + loop->counter.increment));
 	}
 	// restore previous input:
 	Input_now = outer_input;
 }
 
 
-// try to read a condition into DynaBuf and store copy pointer in
+// read condition, make copy, link to struct
+static void copy_condition(struct condition *condition, char terminator)
+{
+	int	err;
+
+	SKIPSPACE();
+	DYNABUF_CLEAR(GlobalDynaBuf);
+	while ((GotByte != terminator) && (GotByte != CHAR_EOS)) {
+		// append to GlobalDynaBuf and check for quotes
+		DYNABUF_APPEND(GlobalDynaBuf, GotByte);
+		if ((GotByte == '"') || (GotByte == '\'')) {
+			err = Input_quoted_to_dynabuf(GotByte);
+			// here GotByte changes, it might become CHAR_EOS
+			DYNABUF_APPEND(GlobalDynaBuf, GotByte);	// add closing quotes (or CHAR_EOS) as well
+			if (err)
+				break;	// on error, exit before eating CHAR_EOS via GetByte()
+		}
+		GetByte();
+	}
+	DynaBuf_append(GlobalDynaBuf, CHAR_EOS);	// ensure terminator
+	condition->body = DynaBuf_get_copy(GlobalDynaBuf);
+}
+
+// try to read a condition into DynaBuf and store pointer to copy in
 // given loop_condition structure.
 // if no condition given, NULL is written to structure.
 // call with GotByte = first interesting character
-void flow_store_doloop_condition(struct loop_condition *condition, char terminator)
+void flow_store_doloop_condition(struct condition *condition, char terminator)
 {
 	// write line number
 	condition->line = Input_now->line_number;
 	// set defaults
-	condition->is_until = FALSE;
+	condition->invert = FALSE;
 	condition->body = NULL;
 	// check for empty condition
 	if (GotByte == terminator)
@@ -100,27 +166,34 @@ void flow_store_doloop_condition(struct loop_condition *condition, char terminat
 	// seems as if there really *is* a condition, so check for until/while
 	if (Input_read_and_lower_keyword()) {
 		if (strcmp(GlobalDynaBuf->buffer, "while") == 0) {
-			//condition.is_until = FALSE;
+			//condition.invert = FALSE;
 		} else if (strcmp(GlobalDynaBuf->buffer, "until") == 0) {
-			condition->is_until = TRUE;
+			condition->invert = TRUE;
 		} else {
 			Throw_error(exception_syntax);
 			return;
 		}
 		// write given condition into buffer
-		SKIPSPACE();
-		DYNABUF_CLEAR(GlobalDynaBuf);
-		Input_until_terminator(terminator);
-		DynaBuf_append(GlobalDynaBuf, CHAR_EOS);	// ensure terminator
-		condition->body = DynaBuf_get_copy(GlobalDynaBuf);
+		copy_condition(condition, terminator);
 	}
 }
 
 
-// check a condition expression
-static int check_condition(struct loop_condition *condition)
+// read a condition into DynaBuf and store pointer to copy in
+// given loop_condition structure.
+// call with GotByte = first interesting character
+void flow_store_while_condition(struct condition *condition)
 {
-	struct result	intresult;
+	condition->line = Input_now->line_number;
+	condition->invert = FALSE;
+	copy_condition(condition, CHAR_SOB);
+}
+
+
+// check a condition expression
+static boolean check_condition(struct condition *condition)
+{
+	struct number	intresult;
 
 	// first, check whether there actually *is* a condition
 	if (condition->body == NULL)
@@ -133,19 +206,19 @@ static int check_condition(struct loop_condition *condition)
 	ALU_defined_int(&intresult);
 	if (GotByte)
 		Throw_serious_error(exception_syntax);
-	return condition->is_until ? !intresult.val.intval : !!intresult.val.intval;
+	return condition->invert ? !intresult.val.intval : !!intresult.val.intval;
 }
 
 
-// back end function for "!do" pseudo opcode
-void flow_doloop(struct do_loop *loop)
+// back end function for "!do" and "!while" pseudo opcodes
+void flow_do_while(struct do_while *loop)
 {
 	struct input	loop_input;
 	struct input	*outer_input;
 
 	// set up new input
 	loop_input = *Input_now;	// copy current input structure into new
-	loop_input.source_is_ram = TRUE;	// set new byte source
+	loop_input.source = INPUTSRC_RAM;	// set new byte source
 	// remember old input
 	outer_input = Input_now;
 	// activate new input (not useable yet, as pointer and
@@ -170,62 +243,11 @@ void flow_doloop(struct do_loop *loop)
 }
 
 
-// helper functions for "!if", "!ifdef" and "!ifndef"
-
-// parse or skip a block. Returns whether block's '}' terminator was missing.
-// afterwards: GotByte = '}'
-static int skip_or_parse_block(int parse)
-{
-	if (!parse) {
-		Input_skip_or_store_block(FALSE);
-		return 0;
-	}
-	// if block was correctly terminated, return FALSE
-	Parse_until_eob_or_eof();
-	// if block isn't correctly terminated, complain and exit
-	if (GotByte != CHAR_EOB)
-		Throw_serious_error(exception_no_right_brace);
-	return 0;
-}
-
-
-// parse {block} [else {block}]
-void flow_parse_block_else_block(int parse_first)
-{
-	// Parse first block.
-	// If it's not correctly terminated, return immediately (because
-	// in that case, there's no use in checking for an "else" part).
-	if (skip_or_parse_block(parse_first))
-		return;
-
-	// now GotByte = '}'. Check for "else" part.
-	// If end of statement, return immediately.
-	NEXTANDSKIPSPACE();
-	if (GotByte == CHAR_EOS)
-		return;
-
-	// read keyword and check whether really "else"
-	if (Input_read_and_lower_keyword()) {
-		if (strcmp(GlobalDynaBuf->buffer, "else")) {
-			Throw_error(exception_syntax);
-		} else {
-			SKIPSPACE();
-			if (GotByte != CHAR_SOB)
-				Throw_serious_error(exception_no_left_brace);
-			skip_or_parse_block(!parse_first);
-			// now GotByte = '}'
-			GetByte();
-		}
-	}
-	Input_ensure_EOS();
-}
-
-
 // parse a whole source code file
 void flow_parse_and_close_file(FILE *fd, const char *filename)
 {
 	// be verbose
-	if (Process_verbosity > 2)
+	if (config.process_verbosity > 2)
 		printf("Parsing source file '%s'\n", filename);
 	// set up new input
 	Input_new_file(filename, fd);

src/flow.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // flow control stuff (loops, conditional assembly etc.)
@@ -19,7 +19,8 @@ struct block {
 // struct to pass "!for" loop stuff from pseudoopcodes.c to flow.c
 struct for_loop {
 	struct symbol	*symbol;
-	int		old_algo;	// actually bool
+	bits		force_bit;
+	boolean		use_old_algo;
 	struct {
 		intval_t	first,
 				last,
@@ -29,32 +30,36 @@ struct for_loop {
 	struct block	block;
 };
 
-// structs to pass "!do" loop stuff from pseudoopcodes.c to flow.c
-struct loop_condition {
+// structs to pass "!do"/"!while" stuff from pseudoopcodes.c to flow.c
+struct condition {
 	int	line;	// original line number
-	int	is_until;	// actually bool (0 for WHILE, 1 for UNTIL)
+	boolean	invert;	// only set for UNTIL conditions
 	char	*body;	// pointer to actual expression
 };
-struct do_loop {
-	struct loop_condition	head_cond;
+struct do_while {
+	struct condition	head_cond;
 	struct block		block;
-	struct loop_condition	tail_cond;
+	struct condition	tail_cond;
 };
 
 
+// parse symbol name and return if symbol has defined value (called by ifdef/ifndef)
+extern boolean check_ifdef_condition(void);
 // back end function for "!for" pseudo opcode
 extern void flow_forloop(struct for_loop *loop);
-// try to read a condition into DynaBuf and store copy pointer in
-// given loop_condition structure.
+// try to read a condition into DynaBuf and store pointer to copy in
+// given condition structure.
 // if no condition given, NULL is written to structure.
 // call with GotByte = first interesting character
-extern void flow_store_doloop_condition(struct loop_condition *condition, char terminator);
+extern void flow_store_doloop_condition(struct condition *condition, char terminator);
+// read a condition into DynaBuf and store pointer to copy in
+// given condition structure.
+// call with GotByte = first interesting character
+extern void flow_store_while_condition(struct condition *condition);
 // back end function for "!do" pseudo opcode
-extern void flow_doloop(struct do_loop *loop);
+extern void flow_do_while(struct do_while *loop);
 // parse a whole source code file
 extern void flow_parse_and_close_file(FILE *fd, const char *filename);
-// parse {block} [else {block}]
-extern void flow_parse_block_else_block(int parse_first);
 
 
 #endif

src/global.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Global stuff - things that are needed by several modules
@@ -8,12 +8,16 @@
 //  2 Jun 2014	Added warn_on_old_for and warn_on_type_mismatch
 // 19 Nov 2014	Merged Johann Klasek's report listing generator patch
 // 23 Nov 2014	Merged Martin Piper's "--msvc" error output patch
+//  9 Jan 2018	Made '/' a syntax char to allow for "//" comments
+// 14 Apr 2020	Added config vars for "ignore zeroes" and "segment warnings to errors"
 #include "global.h"
 #include <stdio.h>
 #include "platform.h"
 #include "acme.h"
+#include "alu.h"
 #include "cpu.h"
 #include "dynabuf.h"
+#include "encoding.h"
 #include "input.h"
 #include "macro.h"
 #include "output.h"
@@ -25,34 +29,23 @@
 
 
 // constants
-
-const char	s_and[]		= "and";
-const char	s_asl[]		= "asl";
-const char	s_asr[]		= "asr";
-const char	s_bra[]		= "bra";
-const char	s_brl[]		= "brl";
-const char	s_cbm[]		= "cbm";
-const char	s_eor[]		= "eor";
-const char	s_error[]	= "error";
-const char	s_lsr[]		= "lsr";
-const char	s_scrxor[]	= "scrxor";
 char		s_untitled[]	= "<untitled>";	// FIXME - this is actually const
-const char	s_Zone[]	= "Zone";
-const char	s_subzone[]	= "subzone";
-const char	s_pet[]		= "pet";
-const char	s_raw[]		= "raw";
-const char	s_scr[]		= "scr";
 
 
 // Exception messages during assembly
-const char	exception_cannot_open_input_file[] = "Cannot open input file.";
 const char	exception_missing_string[]	= "No string given.";
+const char	exception_negative_size[]	= "Negative size argument.";
 const char	exception_no_left_brace[]	= "Missing '{'.";
 const char	exception_no_memory_left[]	= "Out of memory.";
 const char	exception_no_right_brace[]	= "Found end-of-file instead of '}'.";
 //const char	exception_not_yet[]	= "Sorry, feature not yet implemented.";
+// TODO - show actual value in error message
 const char	exception_number_out_of_range[]	= "Number out of range.";
+const char	exception_number_out_of_8b_range[]	= "Number does not fit in 8 bits.";
+static const char	exception_number_out_of_16b_range[]	= "Number does not fit in 16 bits.";
+static const char	exception_number_out_of_24b_range[]	= "Number does not fit in 24 bits.";
 const char	exception_pc_undefined[]	= "Program counter undefined.";
+const char	exception_symbol_defined[]	= "Symbol already defined.";
 const char	exception_syntax[]		= "Syntax error.";
 // default value for number of errors before exiting
 #define MAXERRORS	10
@@ -65,17 +58,17 @@ const char	exception_syntax[]		= "Syntax error.";
 // 7.......	Byte allowed to start keyword
 // .6......	Byte allowed in keyword
 // ..5.....	Byte is upper case, can be lowercased by OR-ing this bit(!)
-// ...4....	special character for input syntax: 0x00 TAB LF CR SPC : ; }
+// ...4....	special character for input syntax: 0x00 TAB LF CR SPC / : ; }
 // ....3...	preceding sequence of '-' characters is anonymous backward
 //		label. Currently only set for ')', ',' and CHAR_EOS.
 // .....210	currently unused
-const char	Byte_flags[256]	= {
+const char	global_byte_flags[256]	= {
 /*$00*/	0x18, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,// control characters
 	0x00, 0x10, 0x10, 0x00, 0x00, 0x10, 0x00, 0x00,
 	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
 	0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,
 /*$20*/	0x10, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00, 0x00,// " !"#$%&'"
-	0x00, 0x08, 0x00, 0x00, 0x08, 0x00, 0x00, 0x00,// "()*+,-./"
+	0x00, 0x08, 0x00, 0x00, 0x08, 0x00, 0x00, 0x10,// "()*+,-./"
 	0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40, 0x40,// "01234567"
 	0x40, 0x40, 0x10, 0x10, 0x00, 0x00, 0x00, 0x00,// "89:;<=>?"
 /*$40*/	0x00, 0xe0, 0xe0, 0xe0, 0xe0, 0xe0, 0xe0, 0xe0,// "@ABCDEFG"
@@ -106,21 +99,28 @@ const char	Byte_flags[256]	= {
 
 
 // variables
-int		pass_count;			// number of current pass (starts 0)
 char		GotByte;			// Last byte read (processed)
-int		Process_verbosity	= 0;	// Level of additional output
-int		warn_on_indented_labels	= TRUE;	// warn if indented label is encountered
-int		warn_on_old_for		= TRUE;	// warn if "!for" with old syntax is found
-int		warn_on_type_mismatch	= FALSE;	// use type-checking system
-// global counters
-int		pass_undefined_count;	// "NeedValue" type errors
-int		pass_real_errors;	// Errors yet
-signed long	max_errors		= MAXERRORS;	// errors before giving up
-FILE		*msg_stream		= NULL;	// set to stdout by --use-stdout
-int		format_msvc		= FALSE;	// actually bool, enabled by --msvc
-int		format_color		= FALSE;	// actually bool, enabled by --color
 struct report 	*report			= NULL;
+struct config	config;
+struct pass	pass;
 
+// set configuration to default values
+void config_default(struct config *conf)
+{
+	conf->pseudoop_prefix		= '!';	// can be changed to '.' by CLI switch
+	conf->process_verbosity		= 0;	// level of additional output
+	conf->warn_on_indented_labels	= TRUE;	// warn if indented label is encountered
+	conf->warn_on_type_mismatch	= FALSE;	// use type-checking system
+	conf->warn_bin_mask		= 3;  // %11 -> warn if not divisible by four
+	conf->max_errors		= MAXERRORS;	// errors before giving up
+	conf->format_msvc		= FALSE;	// enabled by --msvc
+	conf->format_color		= FALSE;	// enabled by --color
+	conf->msg_stream		= stderr;	// set to stdout by --use-stdout
+	conf->honor_leading_zeroes	= TRUE;		// disabled by --ignore-zeroes
+	conf->segment_warning_is_error	= FALSE;	// enabled by --strict-segments		TODO - toggle default?
+	conf->test_new_features		= FALSE;	// enabled by --test
+	conf->wanted_version		= VER_CURRENT;	// changed by --dialect
+}
 
 // memory allocation stuff
 
@@ -137,24 +137,9 @@ void *safe_malloc(size_t size)
 
 // Parser stuff
 
-// Parse (re-)definitions of program counter
-static void parse_pc_def(void)	// Now GotByte = "*"
-{
-	NEXTANDSKIPSPACE();	// proceed with next char
-	// re-definitions of program counter change segment
-	if (GotByte == '=') {
-		GetByte();	// proceed with next char
-		notreallypo_setpc();
-		Input_ensure_EOS();
-	} else {
-		Throw_error(exception_syntax);
-		Input_skip_remainder();
-	}
-}
-
 
 // Check and return whether first label of statement. Complain if not.
-static int first_label_of_statement(int *statement_flags)
+static int first_label_of_statement(bits *statement_flags)
 {
 	if ((*statement_flags) & SF_IMPLIED_LABEL) {
 		Throw_error(exception_syntax);
@@ -166,11 +151,83 @@ static int first_label_of_statement(int *statement_flags)
 }
 
 
-// Parse global symbol definition or assembler mnemonic
-static void parse_mnemo_or_global_symbol_def(int *statement_flags)
+// parse label definition (can be either global or local).
+// name must be held in GlobalDynaBuf.
+// called by parse_symbol_definition, parse_backward_anon_def, parse_forward_anon_def
+// "powers" is used by backward anons to allow changes
+static void set_label(scope_t scope, bits stat_flags, bits force_bit, bits powers)
 {
+	struct symbol	*symbol;
+	struct number	pc;
+	struct object	result;
+
+	if ((stat_flags & SF_FOUND_BLANK) && config.warn_on_indented_labels)
+		Throw_first_pass_warning("Label name not in leftmost column.");
+	symbol = symbol_find(scope);
+	vcpu_read_pc(&pc);	// FIXME - if undefined, check pass.complain_about_undefined and maybe throw "value not defined"!
+	result.type = &type_number;
+	result.u.number.ntype = NUMTYPE_INT;	// FIXME - if undefined, use NUMTYPE_UNDEFINED!
+	result.u.number.flags = 0;
+	result.u.number.val.intval = pc.val.intval;
+	result.u.number.addr_refs = pc.addr_refs;
+	symbol_set_object(symbol, &result, powers);
+	if (force_bit)
+		symbol_set_force_bit(symbol, force_bit);
+	symbol->pseudopc = pseudopc_get_context();
+	// global labels must open new scope for cheap locals
+	if (scope == SCOPE_GLOBAL)
+		section_new_cheap_scope(section_now);
+}
+
+
+// call with symbol name in GlobalDynaBuf and GotByte == '='
+// "powers" is for "!set" pseudo opcode so changes are allowed (see symbol.h for powers)
+void parse_assignment(scope_t scope, bits force_bit, bits powers)
+{
+	struct symbol	*symbol;
+	struct object	result;
+
+	GetByte();	// eat '='
+	symbol = symbol_find(scope);
+	ALU_any_result(&result);
+	// if wanted, mark as address reference
+	if (typesystem_says_address()) {
+		// FIXME - checking types explicitly is ugly...
+		if (result.type == &type_number)
+			result.u.number.addr_refs = 1;
+	}
+	symbol_set_object(symbol, &result, powers);
+	if (force_bit)
+		symbol_set_force_bit(symbol, force_bit);
+}
+
+
+// parse symbol definition (can be either global or local, may turn out to be a label).
+// name must be held in GlobalDynaBuf.
+static void parse_symbol_definition(scope_t scope, bits stat_flags)
+{
+	bits	force_bit;
+
+	force_bit = Input_get_force_bit();	// skips spaces after	(yes, force bit is allowed for label definitions)
+	if (GotByte == '=') {
+		// explicit symbol definition (symbol = <something>)
+		parse_assignment(scope, force_bit, POWER_NONE);
+		Input_ensure_EOS();
+	} else {
+		// implicit symbol definition (label)
+		set_label(scope, stat_flags, force_bit, POWER_NONE);
+	}
+}
+
+
+// Parse global symbol definition or assembler mnemonic
+static void parse_mnemo_or_global_symbol_def(bits *statement_flags)
+{
+	boolean	is_mnemonic;
+
+	is_mnemonic = CPU_state.type->keyword_is_mnemonic(Input_read_keyword());
 	// It is only a label if it isn't a mnemonic
-	if ((CPU_state.type->keyword_is_mnemonic(Input_read_keyword()) == FALSE)
+	if ((!is_mnemonic)
 	&& first_label_of_statement(statement_flags)) {
 		// Now GotByte = illegal char
 		// 04 Jun 2005: this fix should help to explain "strange" error messages.
@@ -178,41 +235,45 @@ static void parse_mnemo_or_global_symbol_def(int *statement_flags)
 		if ((*GLOBALDYNABUF_CURRENT == (char) 0xa0)
 		|| ((GlobalDynaBuf->size >= 2) && (GLOBALDYNABUF_CURRENT[0] == (char) 0xc2) && (GLOBALDYNABUF_CURRENT[1] == (char) 0xa0)))
 			Throw_first_pass_warning("Label name starts with a shift-space character.");
-		symbol_parse_definition(SCOPE_GLOBAL, *statement_flags);
+		parse_symbol_definition(SCOPE_GLOBAL, *statement_flags);
 	}
 }
 
 
-// parse local symbol definition
-static void parse_local_symbol_def(int *statement_flags)
+// parse (cheap) local symbol definition
+static void parse_local_symbol_def(bits *statement_flags, scope_t scope)
 {
 	if (!first_label_of_statement(statement_flags))
 		return;
-	GetByte();	// start after '.'
+
+	GetByte();	// start after '.'/'@'
 	if (Input_read_keyword())
-		symbol_parse_definition(section_now->scope, *statement_flags);
+		parse_symbol_definition(scope, *statement_flags);
 }
 
 
 // parse anonymous backward label definition. Called with GotByte == '-'
-static void parse_backward_anon_def(int *statement_flags)
+static void parse_backward_anon_def(bits *statement_flags)
 {
 	if (!first_label_of_statement(statement_flags))
 		return;
+
 	DYNABUF_CLEAR(GlobalDynaBuf);
 	do
 		DYNABUF_APPEND(GlobalDynaBuf, '-');
 	while (GetByte() == '-');
 	DynaBuf_append(GlobalDynaBuf, '\0');
-	symbol_set_label(section_now->scope, *statement_flags, 0, TRUE);	// this "TRUE" is the whole secret
+	// backward anons change their value!
+	set_label(section_now->local_scope, *statement_flags, NO_FORCE_BIT, POWER_CHANGE_VALUE);
 }
 
 
 // parse anonymous forward label definition. called with GotByte == ?
-static void parse_forward_anon_def(int *statement_flags)
+static void parse_forward_anon_def(bits *statement_flags)
 {
 	if (!first_label_of_statement(statement_flags))
 		return;
+
 	DYNABUF_CLEAR(GlobalDynaBuf);
 	DynaBuf_append(GlobalDynaBuf, '+');
 	while (GotByte == '+') {
@@ -221,8 +282,8 @@ static void parse_forward_anon_def(int *statement_flags)
 	}
 	symbol_fix_forward_anon_name(TRUE);	// TRUE: increment counter
 	DynaBuf_append(GlobalDynaBuf, '\0');
-	//printf("[%d, %s]\n", section_now->scope, GlobalDynaBuf->buffer);
-	symbol_set_label(section_now->scope, *statement_flags, 0, FALSE);
+	//printf("[%d, %s]\n", section_now->local_scope, GlobalDynaBuf->buffer);
+	set_label(section_now->local_scope, *statement_flags, NO_FORCE_BIT, POWER_NONE);
 }
 
 
@@ -231,7 +292,7 @@ static void parse_forward_anon_def(int *statement_flags)
 // Has to be re-entrant.
 void Parse_until_eob_or_eof(void)
 {
-	int	statement_flags;
+	bits	statement_flags;
 
 //	// start with next byte, don't care about spaces
 //	NEXTANDSKIPSPACE();
@@ -243,46 +304,52 @@ void Parse_until_eob_or_eof(void)
 		statement_flags = 0;	// no "label = pc" definition yet
 		typesystem_force_address_statement(FALSE);
 		// Parse until end of statement. Only loops if statement
-		// contains "label = pc" definition and something else; or
-		// if "!ifdef" is true, or if "!addr" is used without block.
+		// contains implicit label definition (=pc) and something else; or
+		// if "!ifdef/ifndef" is true/false, or if "!addr" is used without block.
 		do {
-			switch (GotByte) {
-			case CHAR_EOS:	// end of statement
-				// Ignore now, act later
-				// (stops from being "default")
-				break;
-			case ' ':	// space
-				statement_flags |= SF_FOUND_BLANK;
-				/*FALLTHROUGH*/
-			case CHAR_SOL:	// start of line
-				GetByte();	// skip
-				break;
-			case '-':
-				parse_backward_anon_def(&statement_flags);
-				break;
-			case '+':
-				GetByte();
-				if ((GotByte == LOCAL_PREFIX)
-				|| (BYTEFLAGS(GotByte) & CONTS_KEYWORD))
-					Macro_parse_call();
-				else
-					parse_forward_anon_def(&statement_flags);
-				break;
-			case PSEUDO_OPCODE_PREFIX:
+			// check for pseudo opcodes was moved out of switch,
+			// because prefix character is now configurable.
+			if (GotByte == config.pseudoop_prefix) {
 				pseudoopcode_parse();
-				break;
-			case '*':
-				parse_pc_def();
-				break;
-			case LOCAL_PREFIX:
-				parse_local_symbol_def(&statement_flags);
-				break;
-			default:
-				if (BYTEFLAGS(GotByte) & STARTS_KEYWORD) {
-					parse_mnemo_or_global_symbol_def(&statement_flags);
-				} else {
-					Throw_error(exception_syntax);
-					Input_skip_remainder();
+			} else {
+				switch (GotByte) {
+				case CHAR_EOS:	// end of statement
+					// Ignore now, act later
+					// (stops from being "default")
+					break;
+				case ' ':	// space
+					statement_flags |= SF_FOUND_BLANK;
+					/*FALLTHROUGH*/
+				case CHAR_SOL:	// start of line
+					GetByte();	// skip
+					break;
+				case '-':
+					parse_backward_anon_def(&statement_flags);
+					break;
+				case '+':
+					GetByte();
+					if ((GotByte == LOCAL_PREFIX)	// TODO - allow "cheap macros"?!
+					|| (BYTE_CONTINUES_KEYWORD(GotByte)))
+						Macro_parse_call();
+					else
+						parse_forward_anon_def(&statement_flags);
+					break;
+				case '*':
+					notreallypo_setpc();	// define program counter (fn is in pseudoopcodes.c)
+					break;
+				case LOCAL_PREFIX:
+					parse_local_symbol_def(&statement_flags, section_now->local_scope);
+					break;
+				case CHEAP_PREFIX:
+					parse_local_symbol_def(&statement_flags, section_now->cheap_scope);
+					break;
+				default:
+					if (BYTE_STARTS_KEYWORD(GotByte)) {
+						parse_mnemo_or_global_symbol_def(&statement_flags);
+					} else {
+						Throw_error(exception_syntax);
+						Input_skip_remainder();
+					}
 				}
 			}
 		} while (GotByte != CHAR_EOS);	// until end-of-statement
@@ -321,15 +388,16 @@ int Throw_get_counter(void)
 // This function will do the actual output for warnings, errors and serious
 // errors. It shows the given message string, as well as the current
 // context: file name, line number, source type and source title.
+// TODO: make un-static so !info and !debug can use this.
 static void throw_message(const char *message, const char *type)
 {
 	++throw_counter;
-	if (format_msvc)
-		fprintf(msg_stream, "%s(%d) : %s (%s %s): %s\n",
+	if (config.format_msvc)
+		fprintf(config.msg_stream, "%s(%d) : %s (%s %s): %s\n",
 			Input_now->original_filename, Input_now->line_number,
 			type, section_now->type, section_now->title, message);
 	else
-		fprintf(msg_stream, "%s - File %s, line %d (%s %s): %s\n",
+		fprintf(config.msg_stream, "%s - File %s, line %d (%s %s): %s\n",
 			type, Input_now->original_filename, Input_now->line_number,
 			section_now->type, section_now->title, message);
 }
@@ -342,7 +410,7 @@ static void throw_message(const char *message, const char *type)
 void Throw_warning(const char *message)
 {
 	PLATFORM_WARNING(message);
-	if (format_color)
+	if (config.format_color)
 		throw_message(message, "\033[33mWarning\033[0m");
 	else
 		throw_message(message, "Warning");
@@ -350,7 +418,7 @@ void Throw_warning(const char *message)
 // Output a warning if in first pass. See above.
 void Throw_first_pass_warning(const char *message)
 {
-	if (pass_count == 0)
+	if (FIRST_PASS)
 		Throw_warning(message);
 }
 
@@ -363,12 +431,12 @@ void Throw_first_pass_warning(const char *message)
 void Throw_error(const char *message)
 {
 	PLATFORM_ERROR(message);
-	if (format_color)
+	if (config.format_color)
 		throw_message(message, "\033[31mError\033[0m");
 	else
 		throw_message(message, "Error");
-	++pass_real_errors;
-	if (pass_real_errors >= max_errors)
+	++pass.error_count;
+	if (pass.error_count >= config.max_errors)
 		exit(ACME_finalize(EXIT_FAILURE));
 }
 
@@ -380,7 +448,7 @@ void Throw_error(const char *message)
 void Throw_serious_error(const char *message)
 {
 	PLATFORM_SERIOUS(message);
-	if (format_color)
+	if (config.format_color)
 		throw_message(message, "\033[1m\033[31mSerious error\033[0m");
 	else
 		throw_message(message, "Serious error");
@@ -396,3 +464,125 @@ void Bug_found(const char *message, int code)
 	fprintf(stderr, "(0x%x:)", code);
 	Throw_serious_error(message);
 }
+
+
+// insert object (in case of list, will iterate/recurse until done)
+void output_object(struct object *object, struct iter_context *iter)
+{
+	struct listitem	*item;
+	int		length;
+	char		*read;
+
+	if (object->type == &type_number) {
+		if (object->u.number.ntype == NUMTYPE_UNDEFINED)
+			iter->fn(0);
+		else if (object->u.number.ntype == NUMTYPE_INT)
+			iter->fn(object->u.number.val.intval);
+		else if (object->u.number.ntype == NUMTYPE_FLOAT)
+			iter->fn(object->u.number.val.fpval);
+		else
+			Bug_found("IllegalNumberType0", object->u.number.ntype);
+	} else if (object->type == &type_list) {
+		// iterate over list
+		item = object->u.listhead->next;
+		while (item != object->u.listhead) {
+			output_object(&item->u.payload, iter);
+			item = item->next;
+		}
+	} else if (object->type == &type_string) {
+		// iterate over string
+		read = object->u.string->payload;
+		length = object->u.string->length;
+		// single-char strings are accepted, to be more compatible with
+		// versions before 0.97 (and empty strings are not really a problem...)
+		if (iter->accept_long_strings || (length < 2)) {
+			while (length--)
+				iter->fn(iter->stringxor ^ encoding_encode_char(*(read++)));
+		} else {
+			Throw_error("There's more than one character.");	// see alu.c for the original of this error
+		}
+	} else {
+		Bug_found("IllegalObjectType", 0);
+	}
+}
+
+
+// output 8-bit value with range check
+void output_8(intval_t value)
+{
+	if ((value < -0x80) || (value > 0xff))
+		Throw_error(exception_number_out_of_8b_range);
+	Output_byte(value);
+}
+
+
+// output 16-bit value with range check big-endian
+void output_be16(intval_t value)
+{
+	if ((value < -0x8000) || (value > 0xffff))
+		Throw_error(exception_number_out_of_16b_range);
+	Output_byte(value >> 8);
+	Output_byte(value);
+}
+
+
+// output 16-bit value with range check little-endian
+void output_le16(intval_t value)
+{
+	if ((value < -0x8000) || (value > 0xffff))
+		Throw_error(exception_number_out_of_16b_range);
+	Output_byte(value);
+	Output_byte(value >> 8);
+}
+
+
+// output 24-bit value with range check big-endian
+void output_be24(intval_t value)
+{
+	if ((value < -0x800000) || (value > 0xffffff))
+		Throw_error(exception_number_out_of_24b_range);
+	Output_byte(value >> 16);
+	Output_byte(value >> 8);
+	Output_byte(value);
+}
+
+
+// output 24-bit value with range check little-endian
+void output_le24(intval_t value)
+{
+	if ((value < -0x800000) || (value > 0xffffff))
+		Throw_error(exception_number_out_of_24b_range);
+	Output_byte(value);
+	Output_byte(value >> 8);
+	Output_byte(value >> 16);
+}
+
+
+// FIXME - the range checks below are commented out because 32-bit
+// signed integers cannot exceed the range of 32-bit signed integers.
+// But now that 64-bit machines are the norm, "intval_t" might be a
+// 64-bit int. I need to address this problem one way or another.
+
+
+// output 32-bit value (without range check) big-endian
+void output_be32(intval_t value)
+{
+//	if ((value < -0x80000000) || (value > 0xffffffff))
+//		Throw_error(exception_number_out_of_32b_range);
+	Output_byte(value >> 24);
+	Output_byte(value >> 16);
+	Output_byte(value >> 8);
+	Output_byte(value);
+}
+
+
+// output 32-bit value (without range check) little-endian
+void output_le32(intval_t value)
+{
+//	if ((value < -0x80000000) || (value > 0xffffffff))
+//		Throw_error(exception_number_out_of_32b_range);
+	Output_byte(value);
+	Output_byte(value >> 8);
+	Output_byte(value >> 16);
+	Output_byte(value >> 24);
+}

src/global.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Global stuff - things that are needed by several modules
@@ -14,65 +14,81 @@
 #include <string.h>
 #include "config.h"
 
-#define PSEUDO_OPCODE_PREFIX	'!'	// FIXME - this is not yet used consistently!
 #define LOCAL_PREFIX		'.'	// FIXME - this is not yet used consistently!
+#define CHEAP_PREFIX		'@'	// prefix character for cheap locals
 
 // Constants
 
 #define SF_FOUND_BLANK		(1u << 0)	// statement had space or tab
 #define SF_IMPLIED_LABEL	(1u << 1)	// statement had implied label def
-extern const char	s_and[];
-extern const char	s_asl[];
-extern const char	s_asr[];
-extern const char	s_bra[];
-extern const char	s_brl[];
-extern const char	s_cbm[];
-extern const char	s_eor[];
-extern const char	s_error[];
-extern const char	s_lsr[];
-extern const char	s_scrxor[];
 extern char		s_untitled[];
-extern const char	s_Zone[];
-#define s_zone	(s_subzone + 3)	// Yes, I know I'm sick
-extern const char	s_subzone[];
-extern const char	s_pet[];
-extern const char	s_raw[];
-extern const char	s_scr[];
 // error messages during assembly
-extern const char	exception_cannot_open_input_file[];
 extern const char	exception_missing_string[];
+extern const char	exception_negative_size[];
 extern const char	exception_no_left_brace[];
 extern const char	exception_no_memory_left[];
 extern const char	exception_no_right_brace[];
 //extern const char	exception_not_yet[];
 extern const char	exception_number_out_of_range[];
+extern const char	exception_number_out_of_8b_range[];
 extern const char	exception_pc_undefined[];
+extern const char	exception_symbol_defined[];
 extern const char	exception_syntax[];
 // byte flags table
-extern const char	Byte_flags[];
-#define BYTEFLAGS(c)	(Byte_flags[(unsigned char) c])
-#define STARTS_KEYWORD	(1u << 7)	// Byte is allowed to start a keyword
-#define CONTS_KEYWORD	(1u << 6)	// Byte is allowed in a keyword
-#define BYTEIS_UPCASE	(1u << 5)	// Byte is upper case and can be
-			// converted to lower case by OR-ing this bit(!)
-#define BYTEIS_SYNTAX	(1u << 4)	// special character for input syntax
-#define FOLLOWS_ANON	(1u << 3)	// preceding '-' are backward label
+extern const char	global_byte_flags[];
+#define BYTE_STARTS_KEYWORD(b)		(global_byte_flags[(unsigned char) b] & (1u << 7))	// byte is allowed at start of keyword (a-z, A-Z, _, everything>127)
+#define BYTE_CONTINUES_KEYWORD(b)	(global_byte_flags[(unsigned char) b] & (1u << 6))	// byte is allowed in a keyword (as above, plus digits)
+//#define BYTE_TO_LOWER_CASE(b)	bit 5 means: "byte is upper case, and can be converted to lower case by ORing this bit" - but this is not used at the moment!
+#define BYTE_IS_SYNTAX_CHAR(b)		(global_byte_flags[(unsigned char) b] & (1u << 4))	// special character for input syntax
+#define BYTE_FOLLOWS_ANON(b)		(global_byte_flags[(unsigned char) b] & (1u << 3))	// preceding '-' are backward label
 // bits 2, 1 and 0 are currently unused
 
-
-extern int	pass_count;
-extern int	Process_verbosity;	// Level of additional output
-extern int	warn_on_indented_labels;	// warn if indented label is encountered
-extern int	warn_on_old_for;	// warn if "!for" with old syntax is found
-extern int	warn_on_type_mismatch;	// use type-checking system
+// TODO - put in runtime struct:
 extern char	GotByte;	// Last byte read (processed)
-// global counters
-extern int	pass_undefined_count;	// "NeedValue" type errors in current pass
-extern int	pass_real_errors;	// Errors yet
-extern signed long	max_errors;	// errors before giving up
-extern FILE	*msg_stream;		// set to stdout by --errors_to_stdout
-extern int	format_msvc;		// actually bool, enabled by --msvc
-extern int	format_color;		// actually bool, enabled by --color
+
+enum version {
+	VER_OLDEST_SUPPORTED,		// v0.85 looks like the oldest version it makes sense to actually support
+	VER_DEPRECATE_REALPC,		// v0.86 made !pseudopc/!realpc give a warning to use !pseudopc{} instead, and !to wants a file format
+	VER_SHORTER_SETPC_WARNING,	// v0.93 claimed to allow *= inside !pseudopc blocks, but didn't. It shortened the warning, but '}' or !realpc clobbered PC
+	VER_RIGHTASSOCIATIVEPOWEROF,	// v0.94.6 made "power of" operator right-associative
+					// v0.94.7 fixed a bug: empty code segments no longer included in output file
+	VER_DISABLED_OBSOLETE_STUFF,	// v0.94.8 made *= work inside !pseudopc, disabled !cbm/!realpc/!subzone
+	VER_NEWFORSYNTAX,		// v0.94.12 introduced the new "!for" syntax
+					// v0.95.2 changed ANC#8 from 0x2b to 0x0b
+	VER_BACKSLASHESCAPING,		// v0.97 introduced backslash escaping (and therefore strings)
+	VER_CURRENT,			// "RELEASE"
+					// possible changes in future versions:
+					//	paths should be relative to file, not start dir
+					//	ignore leading zeroes?
+	VER_FUTURE			// far future
+};
+// configuration
+struct config {
+	char		pseudoop_prefix;	// '!' or '.'
+	int		process_verbosity;	// level of additional output
+	boolean		warn_on_indented_labels;	// warn if indented label is encountered
+	boolean		warn_on_type_mismatch;	// use type-checking system
+	int		warn_bin_mask;	// bitmask for digit counter of binary literals
+	signed long	max_errors;	// errors before giving up
+	boolean		format_msvc;		// enabled by --msvc
+	boolean		format_color;		// enabled by --color
+	FILE		*msg_stream;		// defaults to stderr, changed to stdout by --use-stdout
+	boolean		honor_leading_zeroes;	// TRUE, disabled by --ignore-zeroes
+	boolean		segment_warning_is_error;	// FALSE, enabled by --strict-segments
+	boolean		test_new_features;	// FALSE, enabled by --test
+	enum version	wanted_version;	// set by --dialect (and --test --test)
+};
+extern struct config	config;
+
+struct pass {
+	int	number;	// counts up from zero
+	int	undefined_count;	// counts undefined expression results (if this stops decreasing, next pass must list them as errors)
+	//int	needvalue_count;	// counts undefined expression results actually needed for output (when this hits zero, we're done)	FIXME - use
+	int	error_count;
+	boolean	complain_about_undefined;	// will be FALSE until error pass is needed
+};
+extern struct pass	pass;
+#define FIRST_PASS	(pass.number == 0)
 
 // report stuff
 #define REPORT_ASCBUFSIZE	1024
@@ -86,7 +102,7 @@ struct report {
 	char		asc_buf[REPORT_ASCBUFSIZE];	// source bytes
 	char		bin_buf[REPORT_BINBUFSIZE];	// output bytes
 };
-extern struct report	*report;
+extern struct report	*report;	// TODO - put in "part" struct
 
 // Macros for skipping a single space character
 #define SKIPSPACE()		\
@@ -103,8 +119,13 @@ do {				\
 
 // Prototypes
 
+// set configuration to default values
+extern void config_default(struct config *conf);
 // allocate memory and die if not available
-extern void *safe_malloc(size_t);
+extern void *safe_malloc(size_t amount);
+// call with symbol name in GlobalDynaBuf and GotByte == '='
+// "powers" is for "!set" pseudo opcode so changes are allowed (see symbol.h for powers)
+extern void parse_assignment(scope_t scope, bits force_bit, bits powers);
 // Parse block, beginning with next byte.
 // End reason (either CHAR_EOB or CHAR_EOF) can be found in GotByte afterwards
 // Has to be re-entrant.
@@ -119,22 +140,44 @@ extern int Throw_get_counter(void);
 // This means the produced code looks as expected. But there has been a
 // situation that should be reported to the user, for example ACME may have
 // assembled a 16-bit parameter with an 8-bit value.
-extern void Throw_warning(const char *);
+extern void Throw_warning(const char *msg);
 // Output a warning if in first pass. See above.
-extern void Throw_first_pass_warning(const char *);
+extern void Throw_first_pass_warning(const char *msg);
 // Output an error.
 // This means something went wrong in a way that implies that the output
 // almost for sure won't look like expected, for example when there was a
 // syntax error. The assembler will try to go on with the assembly though, so
 // the user gets to know about more than one of his typos at a time.
-extern void Throw_error(const char *);
+extern void Throw_error(const char *msg);
 // Output a serious error, stopping assembly.
 // Serious errors are those that make it impossible to go on with the
 // assembly. Example: "!fill" without a parameter - the program counter cannot
 // be set correctly in this case, so proceeding would be of no use at all.
-extern void Throw_serious_error(const char *);
+extern void Throw_serious_error(const char *msg);
 // handle bugs
-extern void Bug_found(const char *, int);
+extern void Bug_found(const char *msg, int code);
+// insert object (in case of list, will iterate/recurse until done)
+struct iter_context {
+	void		(*fn)(intval_t);	// output function
+	boolean		accept_long_strings;	// if FALSE, only 1-char-strings work
+	unsigned char	stringxor;		// for !scrxor, 0 otherwise
+};
+extern void output_object(struct object *object, struct iter_context *iter);
+// output 8-bit value with range check
+extern void output_8(intval_t value);
+// output 16-bit value with range check big-endian
+extern void output_be16(intval_t value);
+// output 16-bit value with range check little-endian
+extern void output_le16(intval_t value);
+// output 24-bit value with range check big-endian
+extern void output_be24(intval_t value);
+// output 24-bit value with range check little-endian
+extern void output_le24(intval_t value);
+// output 32-bit value (without range check) big-endian
+extern void output_be32(intval_t value);
+// output 32-bit value (without range check) little-endian
+extern void output_le32(intval_t value);
+
 
 
 #endif

src/input.c

@@ -1,14 +1,15 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Input stuff
 // 19 Nov 2014	Merged Johann Klasek's report listing generator patch
+//  9 Jan 2018	Allowed "//" comments
 #include "input.h"
 #include "config.h"
 #include "alu.h"
 #include "dynabuf.h"
-#include "global.h"	// FIXME - remove when no longer needed
+#include "global.h"
 #include "platform.h"
 #include "section.h"
 #include "symbol.h"
@@ -17,20 +18,17 @@
 
 // Constants
 const char	FILE_READBINARY[]	= "rb";
-#define CHAR_TAB	(9)	// Tab character
 #define CHAR_LF		(10)	// line feed		(in file)
 		//	(10)	// start of line	(in high-level format)
 #define CHAR_CR		(13)	// carriage return	(in file)
 		//	(13)	// end of file		(in high-level format)
-#define CHAR_STATEMENT_DELIMITER	':'
-#define	CHAR_COMMENT_SEPARATOR		';'
-// if the characters above are changed, don't forget to adjust ByteFlags[]!
+// if the characters above are changed, don't forget to adjust byte_flags[]!
 
 // fake input structure (for error msgs before any real input is established)
 static struct input	outermost	= {
 	"<none>",	// file name
 	0,		// line number
-	FALSE,		// Faked file access, so no RAM read
+	INPUTSRC_FILE,	// fake file access, so no RAM read
 	INPUTSTATE_EOF,	// state of input
 	{
 		NULL	// RAM read pointer or file handle
@@ -49,8 +47,8 @@ void Input_new_file(const char *filename, FILE *fd)
 {
 	Input_now->original_filename	= filename;
 	Input_now->line_number		= 1;
-	Input_now->source_is_ram	= FALSE;
-	Input_now->state		= INPUTSTATE_NORMAL;
+	Input_now->source		= INPUTSRC_FILE;
+	Input_now->state		= INPUTSTATE_SOF;
 	Input_now->src.fd		= fd;
 }
 
@@ -78,14 +76,15 @@ static void report_srcchar(char new_char)
 		// show line number...
 		fprintf(report->fd, "%6d  ", Input_now->line_number - 1);
 		// prepare outbytes' start address
-		if (report->bin_used)
+		if (report->bin_used) {
 #if _BSD_SOURCE || _XOPEN_SOURCE >= 500 || _ISOC99_SOURCE || _POSIX_C_SOURCE >= 200112L
 			snprintf(hex_address, HEXBUFSIZE, "%04x", report->bin_address);
 #else
 			sprintf(hex_address, "%04x", report->bin_address);
 #endif
-		else
+		} else {
 			hex_address[0] = '\0';
+		}
 		// prepare outbytes
 		hexdump[0] = '\0';
 		for (ii = 0; ii < report->bin_used; ++ii)
@@ -115,10 +114,23 @@ static void report_srcchar(char new_char)
 // Deliver source code from current file (!) in shortened high-level format
 static char get_processed_from_file(void)
 {
-	int	from_file = 0;
+	static int	from_file	= 0;
 
 	for (;;) {
 		switch (Input_now->state) {
+		case INPUTSTATE_SOF:
+			// fetch first byte from the current source file
+			from_file = getc(Input_now->src.fd);
+			IF_WANTED_REPORT_SRCCHAR(from_file);
+			//TODO - check for bogus/malformed BOM and ignore?
+			// check for hashbang line and ignore
+			if (from_file == '#') {
+				// remember to skip remainder of line
+				Input_now->state = INPUTSTATE_COMMENT;
+				return CHAR_EOS;	// end of statement
+			}
+			Input_now->state = INPUTSTATE_AGAIN;
+			break;
 		case INPUTSTATE_NORMAL:
 			// fetch a fresh byte from the current source file
 			from_file = getc(Input_now->src.fd);
@@ -134,7 +146,7 @@ static char get_processed_from_file(void)
 			// defined "from_file", trouble may arise...
 			Input_now->state = INPUTSTATE_NORMAL;
 			// EOF must be checked first because it cannot be used
-			// as an index into Byte_flags[]
+			// as an index into global_byte_flags[]
 			if (from_file == EOF) {
 				// remember to send an end-of-file
 				Input_now->state = INPUTSTATE_EOF;
@@ -143,12 +155,12 @@ static char get_processed_from_file(void)
 
 			// check whether character is special one
 			// if not, everything's cool and froody, so return it
-			if ((BYTEFLAGS(from_file) & BYTEIS_SYNTAX) == 0)
+			if (BYTE_IS_SYNTAX_CHAR(from_file) == 0)
 				return (char) from_file;
 
-			// check special characters ("0x00 TAB LF CR SPC :;}")
+			// check special characters ("0x00 TAB LF CR SPC / : ; }")
 			switch (from_file) {
-			case CHAR_TAB:	// TAB character
+			case '\t':
 			case ' ':
 				// remember to skip all following blanks
 				Input_now->state = INPUTSTATE_SKIPBLANKS;
@@ -169,15 +181,26 @@ static char get_processed_from_file(void)
 				Input_now->state = INPUTSTATE_EOB;
 				return CHAR_EOS;	// end of statement
 
-			case CHAR_STATEMENT_DELIMITER:
-				// just deliver an EOS instead
-				return CHAR_EOS;	// end of statement
-
-			case CHAR_COMMENT_SEPARATOR:
+			case '/':
+				// to check for "//", get another byte:
+				from_file = getc(Input_now->src.fd);
+				IF_WANTED_REPORT_SRCCHAR(from_file);
+				if (from_file != '/') {
+					// not "//", so:
+					Input_now->state = INPUTSTATE_AGAIN;	// second byte must be parsed normally later on
+					return '/';	// first byte is returned normally right now
+				}
+				// it's really "//", so act as if ';'
+				/*FALLTHROUGH*/
+			case ';':
 				// remember to skip remainder of line
 				Input_now->state = INPUTSTATE_COMMENT;
 				return CHAR_EOS;	// end of statement
 
+			case ':':	// statement delimiter
+				// just deliver an EOS instead
+				return CHAR_EOS;	// end of statement
+
 			default:
 				// complain if byte is 0
 				Throw_error("Source file contains illegal character.");
@@ -188,7 +211,7 @@ static char get_processed_from_file(void)
 			do {
 				from_file = getc(Input_now->src.fd);
 				IF_WANTED_REPORT_SRCCHAR(from_file);
-			} while ((from_file == CHAR_TAB) || (from_file == ' '));
+			} while ((from_file == '\t') || (from_file == ' '));
 			// re-process last byte
 			Input_now->state = INPUTSTATE_AGAIN;
 			break;
@@ -239,7 +262,7 @@ static char get_processed_from_file(void)
 
 // This function delivers the next byte from the currently active byte source
 // in shortened high-level format. FIXME - use fn ptr?
-// When inside quotes, use GetQuotedByte() instead!
+// When inside quotes, use Input_quoted_to_dynabuf() instead!
 char GetByte(void)
 {
 //	for (;;) {
@@ -247,11 +270,17 @@ char GetByte(void)
 		// necessary, because in RAM the source already has
 		// high-level format
 		// Otherwise, the source is a file. This means we will call
-		// GetFormatted() which will do a shit load of conversions.
-		if (Input_now->source_is_ram)
+		// get_processed_from_file() which will do a shit load of conversions.
+		switch (Input_now->source) {
+		case INPUTSRC_RAM:
 			GotByte = *(Input_now->src.ram_ptr++);
-		else
+			break;
+		case INPUTSRC_FILE:
 			GotByte = get_processed_from_file();
+			break;
+		default:
+			Bug_found("IllegalInputSrc", Input_now->source);
+		}
 //		// if start-of-line was read, increment line counter and repeat
 //		if (GotByte != CHAR_SOL)
 //			return GotByte;
@@ -265,16 +294,18 @@ char GetByte(void)
 // This function delivers the next byte from the currently active byte source
 // in un-shortened high-level format.
 // This function complains if CHAR_EOS (end of statement) is read.
-char GetQuotedByte(void)
+// TODO - check if return value is actually used
+static char GetQuotedByte(void)
 {
 	int	from_file;	// must be an int to catch EOF
 
-	// if byte source is RAM, then no conversion is necessary,
-	// because in RAM the source already has high-level format
-	if (Input_now->source_is_ram) {
+	switch (Input_now->source) {
+	case INPUTSRC_RAM:
+		// if byte source is RAM, then no conversion is necessary,
+		// because in RAM the source already has high-level format
 		GotByte = *(Input_now->src.ram_ptr++);
-	// Otherwise, the source is a file.
-	} else {
+		break;
+	case INPUTSRC_FILE:
 		// fetch a fresh byte from the current source file
 		from_file = getc(Input_now->src.fd);
 		IF_WANTED_REPORT_SRCCHAR(from_file);
@@ -297,7 +328,9 @@ char GetQuotedByte(void)
 		default:
 			GotByte = from_file;
 		}
-
+		break;
+	default:
+		Bug_found("IllegalInputSrc", Input_now->source);
 	}
 	// now check for end of statement
 	if (GotByte == CHAR_EOS)
@@ -306,6 +339,7 @@ char GetQuotedByte(void)
 }
 
 // Skip remainder of statement, for example on error
+// FIXME - check for quotes, otherwise this might treat a quoted colon like EOS!
 void Input_skip_remainder(void)
 {
 	while (GotByte)
@@ -318,20 +352,111 @@ void Input_ensure_EOS(void)	// Now GotByte = first char to test
 {
 	SKIPSPACE();
 	if (GotByte) {
-		Throw_error("Garbage data at end of statement.");
+		char	buf[80];	// actually needed are 51
+		char	quote;		// character before and after
+
+		quote = (GotByte == '\'') ? '"' : '\'';	// use single quotes, unless byte is a single quote (then use double quotes)
+		sprintf(buf, "Garbage data at end of statement (unexpected %c%c%c).", quote, GotByte, quote);
+		Throw_error(buf);
 		Input_skip_remainder();
 	}
 }
 
+// read string to dynabuf until closing quote is found
+// returns 1 on errors (unterminated, escaping error)
+int Input_quoted_to_dynabuf(char closing_quote)
+{
+	boolean	escaped	= FALSE;
+
+	//DYNABUF_CLEAR(GlobalDynaBuf);	// do not clear, caller might want to append to existing contents (TODO - check!)
+	for (;;) {
+		GetQuotedByte();
+		if (GotByte == CHAR_EOS)
+			return 1;	// unterminated string constant; GetQuotedByte will have complained already
+
+		if (escaped) {
+			// previous byte was backslash, so do not check for terminator nor backslash
+			escaped = FALSE;
+			// do not actually _convert_ escape sequences to their target byte, that is done by Input_unescape_dynabuf() below!
+			// TODO - but maybe check for illegal escape sequences?
+			// at the moment checking is only done when the string
+			// gets used for something...
+		} else {
+			// non-escaped: only terminator and backslash are of interest
+			if (GotByte == closing_quote)
+				return 0;	// ok
+
+			if ((GotByte == '\\') && (config.wanted_version >= VER_BACKSLASHESCAPING))
+				escaped = TRUE;
+		}
+		DYNABUF_APPEND(GlobalDynaBuf, GotByte);
+	}
+}
+
+// process backslash escapes in GlobalDynaBuf (so size might shrink)
+// returns 1 on errors (escaping errors)
+// TODO - check: if this is only ever called directly after Input_quoted_to_dynabuf, integrate that call here?
+int Input_unescape_dynabuf(int read_index)
+{
+	int	write_index;
+	char	byte;
+	boolean	escaped;
+
+	if (config.wanted_version < VER_BACKSLASHESCAPING)
+		return 0;	// ok
+
+	write_index = read_index;
+	escaped = FALSE;
+	// CAUTION - contents of dynabuf are not terminated:
+	while (read_index < GlobalDynaBuf->size) {
+		byte = GLOBALDYNABUF_CURRENT[read_index++];
+		if (escaped) {
+			switch (byte) {
+			case '\\':
+			case '\'':
+			case '"':
+				break;
+			case '0':	// NUL
+				byte = 0;
+				break;
+			case 't':	// TAB
+				byte = 9;
+				break;
+			case 'n':	// LF
+				byte = 10;
+				break;
+			case 'r':	// CR
+				byte = 13;
+				break;
+			// TODO - 'a' to BEL? others?
+			default:
+				Throw_error("Unsupported backslash sequence.");	// TODO - add unexpected character to error message?
+			}
+			GLOBALDYNABUF_CURRENT[write_index++] = byte;
+			escaped = FALSE;
+		} else {
+			if (byte == '\\') {
+				escaped = TRUE;
+			} else {
+				GLOBALDYNABUF_CURRENT[write_index++] = byte;
+			}
+		}
+	}
+	if (escaped)
+		Bug_found("PartialEscapeSequence", 0);
+	GlobalDynaBuf->size = write_index;
+	return 0;	// ok
+}
+
 // Skip or store block (starting with next byte, so call directly after
 // reading opening brace).
-// If "Store" is TRUE, the block is read into GlobalDynaBuf, then a copy
-// is made and a pointer to that is returned.
+// the block is read into GlobalDynaBuf.
+// If "Store" is TRUE, then a copy is made and a pointer to that is returned.
 // If "Store" is FALSE, NULL is returned.
 // After calling this function, GotByte holds '}'. Unless EOF was found first,
 // but then a serious error would have been thrown.
 // FIXME - use a struct block *ptr argument!
-char *Input_skip_or_store_block(int store)
+char *Input_skip_or_store_block(boolean store)
 {
 	char	byte;
 	int	depth	= 1;	// to find matching block end
@@ -340,9 +465,8 @@ char *Input_skip_or_store_block(int store)
 	DYNABUF_CLEAR(GlobalDynaBuf);
 	do {
 		byte = GetByte();
-		// if wanted, store
-		if (store)
-			DYNABUF_APPEND(GlobalDynaBuf, byte);
+		// store
+		DYNABUF_APPEND(GlobalDynaBuf, byte);
 		// now check for some special characters
 		switch (byte) {
 		case CHAR_EOF:	// End-of-file in block? Sorry, no way.
@@ -350,12 +474,8 @@ char *Input_skip_or_store_block(int store)
 
 		case '"':	// Quotes? Okay, read quoted stuff.
 		case '\'':
-			do {
-				GetQuotedByte();
-				// if wanted, store
-				if (store)
-					DYNABUF_APPEND(GlobalDynaBuf, GotByte);
-			} while ((GotByte != CHAR_EOS) && (GotByte != byte));
+			Input_quoted_to_dynabuf(byte);
+			DYNABUF_APPEND(GlobalDynaBuf, GotByte);	// add closing quote
 			break;
 		case CHAR_SOB:
 			++depth;
@@ -368,6 +488,7 @@ char *Input_skip_or_store_block(int store)
 	// in case of skip, return now
 	if (!store)
 		return NULL;
+
 	// otherwise, prepare to return copy of block
 	// add EOF, just to make sure block is never read too far
 	DynaBuf_append(GlobalDynaBuf, CHAR_EOS);
@@ -376,33 +497,6 @@ char *Input_skip_or_store_block(int store)
 	return DynaBuf_get_copy(GlobalDynaBuf);
 }
 
-// Read bytes and add to GlobalDynaBuf until the given terminator (or CHAR_EOS)
-// is found. Act upon single and double quotes by entering (and leaving) quote
-// mode as needed (So the terminator does not terminate when inside quotes).
-void Input_until_terminator(char terminator)
-{
-	char	byte	= GotByte;
-
-	for (;;) {
-		// Terminator? Exit. EndOfStatement? Exit.
-		if ((byte == terminator) || (byte == CHAR_EOS))
-			return;
-		// otherwise, append to GlobalDynaBuf and check for quotes
-		DYNABUF_APPEND(GlobalDynaBuf, byte);
-		if ((byte == '"') || (byte == '\'')) {
-			do {
-				// Okay, read quoted stuff.
-				GetQuotedByte();	// throws error on EOS
-				DYNABUF_APPEND(GlobalDynaBuf, GotByte);
-			} while ((GotByte != CHAR_EOS) && (GotByte != byte));
-			// on error, exit now, before calling GetByte()
-			if (GotByte != byte)
-				return;
-		}
-		byte = GetByte();
-	}
-}
-
 // Append to GlobalDynaBuf while characters are legal for keywords.
 // Throws "missing string" error if none.
 // Returns number of characters added.
@@ -411,7 +505,7 @@ int Input_append_keyword_to_global_dynabuf(void)
 	int	length	= 0;
 
 	// add characters to buffer until an illegal one comes along
-	while (BYTEFLAGS(GotByte) & CONTS_KEYWORD) {
+	while (BYTE_CONTINUES_KEYWORD(GotByte)) {
 		DYNABUF_APPEND(GlobalDynaBuf, GotByte);
 		++length;
 		GetByte();
@@ -421,19 +515,20 @@ int Input_append_keyword_to_global_dynabuf(void)
 	return length;
 }
 
-// Check whether GotByte is LOCAL_PREFIX (default '.').
-// If not, store global scope value.
-// If yes, store current local scope value and read next byte.
+// Check GotByte.
+// If LOCAL_PREFIX ('.'), store current local scope value and read next byte.
+// If CHEAP_PREFIX ('@'), store current cheap scope value and read next byte.
+// Otherwise, store global scope value.
 // Then jump to Input_read_keyword(), which returns length of keyword.
 int Input_read_scope_and_keyword(scope_t *scope)
 {
 	SKIPSPACE();
 	if (GotByte == LOCAL_PREFIX) {
 		GetByte();
-		*scope = section_now->scope;
-/* TODO	} else if (GotByte == CHEAP_PREFIX) {
+		*scope = section_now->local_scope;
+	} else if (GotByte == CHEAP_PREFIX) {
 		GetByte();
-		*scope = symbol_cheap_scope;	*/
+		*scope = section_now->cheap_scope;
 	} else {
 		*scope = SCOPE_GLOBAL;
 	}
@@ -471,25 +566,31 @@ int Input_read_and_lower_keyword(void)
 	return length;
 }
 
-// Try to read a file name. If "allow_library" is TRUE, library access by using
-// <...> quoting is possible as well. The file name given in the assembler
-// source code is converted from UNIX style to platform style.
-// Returns whether error occurred (TRUE on error). Filename in GlobalDynaBuf.
+// Try to read a file name.
+// If "allow_library" is TRUE, library access by using <...> quoting
+// is possible as well. If "uses_lib" is non-NULL, info about library
+// usage is stored there.
+// The file name given in the assembler source code is converted from
+// UNIX style to platform style.
+// Returns nonzero on error. Filename in GlobalDynaBuf.
 // Errors are handled and reported, but caller should call
 // Input_skip_remainder() then.
-int Input_read_filename(int allow_library)
+int Input_read_filename(boolean allow_library, boolean *uses_lib)
 {
+	int	start_of_string;
 	char	*lib_prefix,
-		end_quote;
+		terminator;
 
 	DYNABUF_CLEAR(GlobalDynaBuf);
 	SKIPSPACE();
-	// check for library access
-	if (GotByte == '<') {
+	switch (GotByte) {
+	case '<':	// library access
+		if (uses_lib)
+			*uses_lib = TRUE;
 		// if library access forbidden, complain
-		if (allow_library == FALSE) {
+		if (!allow_library) {
 			Throw_error("Writing to library not supported.");
-			return TRUE;
+			return 1;	// error
 		}
 
 		// read platform's lib prefix
@@ -498,40 +599,46 @@ int Input_read_filename(int allow_library)
 		// if lib prefix not set, complain
 		if (lib_prefix == NULL) {
 			Throw_error("\"ACME\" environment variable not found.");
-			return TRUE;
+			return 1;	// error
 		}
 #endif
 		// copy lib path and set quoting char
 		DynaBuf_add_string(GlobalDynaBuf, lib_prefix);
-		end_quote = '>';
-	} else {
-		if (GotByte == '"') {
-			end_quote = '"';
-		} else {
-			Throw_error("File name quotes not found (\"\" or <>).");
-			return TRUE;
-		}
+		terminator = '>';
+		break;
+	case '"':	// normal access
+		if (uses_lib)
+			*uses_lib = FALSE;
+		terminator = '"';
+		break;
+	default:	// none of the above
+		Throw_error("File name quotes not found (\"\" or <>).");
+		return 1;	// error
 	}
-	// read first character, complain if closing quote
-	if (GetQuotedByte() == end_quote) {
+	// remember border between optional library prefix and string from assembler source file
+	start_of_string = GlobalDynaBuf->size;
+	// read file name string
+	if (Input_quoted_to_dynabuf(terminator))
+		return 1;	// unterminated or escaping error
+
+	GetByte();	// eat terminator
+	// check length
+	if (GlobalDynaBuf->size == start_of_string) {
 		Throw_error("No file name given.");
-		return TRUE;
+		return 1;	// error
 	}
 
-	// read characters until closing quote (or EOS) is reached
-	// append platform-converted characters to current string
-	while ((GotByte != CHAR_EOS) && (GotByte != end_quote)) {
-		DYNABUF_APPEND(GlobalDynaBuf, PLATFORM_CONVERTPATHCHAR(GotByte));
-		GetQuotedByte();
-	}
-	// on error, return
-	if (GotByte == CHAR_EOS)
-		return TRUE;
+	// resolve backslash escapes
+	if (Input_unescape_dynabuf(start_of_string))
+		return 1;	// escaping error
 
-	GetByte();	// fetch next to forget closing quote
 	// terminate string
-	DynaBuf_append(GlobalDynaBuf, '\0');	// add terminator
-	return FALSE;	// no error
+	DynaBuf_append(GlobalDynaBuf, '\0');
+#ifdef PLATFORM_CONVERTPATH
+	// platform-specific path name conversion
+	PLATFORM_CONVERTPATH(GLOBALDYNABUF_CURRENT + start_of_string);
+#endif
+	return 0;	// ok
 }
 
 // Try to read a comma, skipping spaces before and after. Return TRUE if comma
@@ -547,19 +654,20 @@ int Input_accept_comma(void)
 }
 
 // read optional info about parameter length
-int Input_get_force_bit(void)
+// FIXME - move to different file!
+bits Input_get_force_bit(void)
 {
 	char	byte;
-	int	force_bit	= 0;
+	bits	force_bit	= 0;
 
 	if (GotByte == '+') {
 		byte = GetByte();
 		if (byte == '1')
-			force_bit = MVALUE_FORCE08;
+			force_bit = NUMBER_FORCES_8;
 		else if (byte == '2')
-			force_bit = MVALUE_FORCE16;
+			force_bit = NUMBER_FORCES_16;
 		else if (byte == '3')
-			force_bit = MVALUE_FORCE24;
+			force_bit = NUMBER_FORCES_24;
 		if (force_bit)
 			GetByte();
 		else
@@ -568,3 +676,76 @@ int Input_get_force_bit(void)
 	SKIPSPACE();
 	return force_bit;
 }
+
+
+// include path stuff - should be moved to its own file:
+
+// ring list struct
+struct ipi {
+	struct ipi	*next,
+			*prev;
+	const char	*path;
+};
+static struct ipi	ipi_head	= {&ipi_head, &ipi_head, NULL};	// head element
+static	STRUCT_DYNABUF_REF(pathbuf, 256);	// to combine search path and file spec
+
+// add entry
+void includepaths_add(const char *path)
+{
+	struct ipi	*ipi;
+
+	ipi = safe_malloc(sizeof(*ipi));
+	ipi->path = path;
+	ipi->next = &ipi_head;
+	ipi->prev = ipi_head.prev;
+	ipi->next->prev = ipi;
+	ipi->prev->next = ipi;
+}
+// open file for reading (trying list entries as prefixes)
+// "uses_lib" tells whether to access library or to make use of include paths
+// file name is expected in GlobalDynaBuf
+FILE *includepaths_open_ro(boolean uses_lib)
+{
+	FILE		*stream;
+	struct ipi	*ipi;
+
+	// first try directly, regardless of whether lib or not:
+	stream = fopen(GLOBALDYNABUF_CURRENT, FILE_READBINARY);
+	// if failed and not lib, try include paths:
+	if ((stream == NULL) && !uses_lib) {
+		for (ipi = ipi_head.next; ipi != &ipi_head; ipi = ipi->next) {
+			DYNABUF_CLEAR(pathbuf);
+			// add first part
+			DynaBuf_add_string(pathbuf, ipi->path);
+			// if wanted and possible, ensure last char is directory separator
+			if (DIRECTORY_SEPARATOR
+			&& pathbuf->size
+			&& (pathbuf->buffer[pathbuf->size - 1] != DIRECTORY_SEPARATOR))
+				DynaBuf_append(pathbuf, DIRECTORY_SEPARATOR);
+			// add second part
+			DynaBuf_add_string(pathbuf, GLOBALDYNABUF_CURRENT);
+			// terminate
+			DynaBuf_append(pathbuf, '\0');
+			// try
+			stream = fopen(pathbuf->buffer, FILE_READBINARY);
+			//printf("trying <<%s>> - ", pathbuf->buffer);
+			if (stream) {
+				//printf("ok\n");
+				break;
+			} else {
+				//printf("failed\n");
+			}
+		}
+	}
+	if (stream == NULL) {
+		// CAUTION, I'm re-using the path dynabuf to assemble the error message:
+		DYNABUF_CLEAR(pathbuf);
+		DynaBuf_add_string(pathbuf, "Cannot open input file \"");
+		DynaBuf_add_string(pathbuf, GLOBALDYNABUF_CURRENT);
+		DynaBuf_add_string(pathbuf, "\".");
+		DynaBuf_append(pathbuf, '\0');
+		Throw_error(pathbuf->buffer);
+	}
+	//fprintf(stderr, "File is [%s]\n", GLOBALDYNABUF_CURRENT);
+	return stream;
+}

src/input.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Input stuff
@@ -8,13 +8,14 @@
 
 
 #include <stdio.h>	// for FILE
-#include "config.h"	// for scope_t
+#include "config.h"	// for bits and scope_t
 
 
 // type definitions
 
 // values for input component "src.state"
 enum inputstate {
+	INPUTSTATE_SOF,		// start of file (check for hashbang)
 	INPUTSTATE_NORMAL,	// everything's fine
 	INPUTSTATE_AGAIN,	// re-process last byte
 	INPUTSTATE_SKIPBLANKS,	// shrink multiple spaces
@@ -25,10 +26,14 @@ enum inputstate {
 	INPUTSTATE_EOB,		// send end-of-block after end-of-statement
 	INPUTSTATE_EOF,		// send end-of-file after end-of-statement
 };
+enum inputsrc {
+	INPUTSRC_FILE,
+	INPUTSRC_RAM
+};
 struct input {
 	const char	*original_filename;	// during RAM reads, too
-	int		line_number,	// in file (on RAM reads, too)
-			source_is_ram;	// TRUE if RAM, FALSE if file
+	int		line_number;	// in file (on RAM reads, too)
+	enum inputsrc	source;
 	enum inputstate	state;	// state of input
 	union {
 		FILE	*fd;		// file descriptor
@@ -46,7 +51,7 @@ extern const char	FILE_READBINARY[];
 #define CHAR_EOB	'}'	// end of block
 #define CHAR_SOL	(10)	// start of line	(in high-level format)
 #define CHAR_EOF	(13)	// end of file		(in high-level format)
-// If the characters above are changed, don't forget to adjust Byte_flags[]!
+// if the characters above are changed, don't forget to adjust global_byte_flags[]!
 
 
 // Variables
@@ -58,34 +63,38 @@ extern struct input	*Input_now;	// current input structure
 // let current input point to start of file
 extern void Input_new_file(const char *filename, FILE *fd);
 // get next byte from currently active byte source in shortened high-level
-// format. When inside quotes, use GetQuotedByte() instead!
+// format. When inside quotes, use Input_quoted_to_dynabuf() instead!
 extern char GetByte(void);
-// get next byte from currently active byte source in un-shortened high-level
-// format. Complains if CHAR_EOS (end of statement) is read.
-extern char GetQuotedByte(void);
 // Skip remainder of statement, for example on error
 extern void Input_skip_remainder(void);
 // Ensure that the remainder of the current statement is empty, for example
 // after mnemonics using implied addressing.
 extern void Input_ensure_EOS(void);
+
+// read string to dynabuf until closing quote is found
+// returns 1 on errors (unterminated, escaping error)
+extern int Input_quoted_to_dynabuf(char closing_quote);
+
+// process backslash escapes in GlobalDynaBuf (so size might shrink)
+// returns 1 on errors (escaping errors)
+extern int Input_unescape_dynabuf(int start_index);
+
 // Skip or store block (starting with next byte, so call directly after
 // reading opening brace).
-// If "Store" is TRUE, the block is read into GlobalDynaBuf, then a copy
-// is made and a pointer to that is returned.
+// the block is read into GlobalDynaBuf.
+// If "Store" is TRUE, then a copy is made and a pointer to that is returned.
 // If "Store" is FALSE, NULL is returned.
 // After calling this function, GotByte holds '}'. Unless EOF was found first,
 // but then a serious error would have been thrown.
-extern char *Input_skip_or_store_block(int store);
-// Read bytes and add to GlobalDynaBuf until the given terminator (or CHAR_EOS)
-// is found. Act upon single and double quotes by entering (and leaving) quote
-// mode as needed (So the terminator does not terminate when inside quotes).
-extern void Input_until_terminator(char terminator);
+extern char *Input_skip_or_store_block(boolean store);
+
 // Append to GlobalDynaBuf while characters are legal for keywords.
 // Throws "missing string" error if none. Returns number of characters added.
 extern int Input_append_keyword_to_global_dynabuf(void);
-// Check whether GotByte is a dot.
-// If not, store global scope value.
-// If yes, store current scope value and read next byte.
+// Check GotByte.
+// If LOCAL_PREFIX ('.'), store current local scope value and read next byte.
+// If CHEAP_PREFIX ('@'), store current cheap scope value and read next byte.
+// Otherwise, store global scope value.
 // Then jump to Input_read_keyword(), which returns length of keyword.
 extern int Input_read_scope_and_keyword(scope_t *scope);
 // Clear dynamic buffer, then append to it until an illegal (for a keyword)
@@ -98,18 +107,31 @@ extern int Input_read_keyword(void);
 // Return its length (without terminator).
 // Zero lengths will produce a "missing string" error.
 extern int Input_read_and_lower_keyword(void);
-// Try to read a file name. If "allow_library" is TRUE, library access by using
-// <...> quoting is possible as well. The file name given in the assembler
-// source code is converted from UNIX style to platform style.
-// Returns whether error occurred (TRUE on error). Filename in GlobalDynaBuf.
+// Try to read a file name.
+// If "allow_library" is TRUE, library access by using <...> quoting
+// is possible as well. If "uses_lib" is non-NULL, info about library
+// usage is stored there.
+// The file name given in the assembler source code is converted from
+// UNIX style to platform style.
+// Returns nonzero on error. Filename in GlobalDynaBuf.
 // Errors are handled and reported, but caller should call
 // Input_skip_remainder() then.
-extern int Input_read_filename(int library_allowed);
+extern int Input_read_filename(boolean library_allowed, boolean *uses_lib);
 // Try to read a comma, skipping spaces before and after. Return TRUE if comma
 // found, otherwise FALSE.
 extern int Input_accept_comma(void);
 // read optional info about parameter length
-extern int Input_get_force_bit(void);
+extern bits Input_get_force_bit(void);
+
+
+// include path stuff - should be moved to its own file:
+
+// add entry
+extern void includepaths_add(const char *path);
+// open file for reading (trying list entries as prefixes)
+// "uses_lib" tells whether to access library or to make use of include paths
+// file name is expected in GlobalDynaBuf
+extern FILE *includepaths_open_ro(boolean uses_lib);
 
 
 #endif

src/macro.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Macro stuff
@@ -18,12 +18,10 @@
 
 
 // Constants
-#define MACRONAME_DYNABUF_INITIALSIZE	128
+#define NAME_INITIALSIZE	128
 #define ARG_SEPARATOR	' '	// separates macro title from arg types
-#define ARGTYPE_NUM_VAL	'v'
-#define ARGTYPE_NUM_REF	'V'
-//#define ARGTYPE_STR_VAL	's'
-//#define ARGTYPE_STR_REF	'S'
+#define ARGTYPE_VALUE	'v'
+#define ARGTYPE_REF	'r'
 #define REFERENCE_CHAR	'~'	// prefix for call-by-reference
 #define HALF_INITIAL_ARG_TABLE_SIZE	4
 static const char	exception_macro_twice[]	= "Macro already defined.";
@@ -42,14 +40,14 @@ struct macro {
 // gives us the possibility to find out which args are call-by-value and
 // which ones are call-by-reference.
 union macro_arg_t {
-	struct result	result;	// value and flags (call by value)
+	struct object	result;	// value and flags (call by value)
 	struct symbol	*symbol;	// pointer to symbol struct (call by reference)
 };
 
 
 // Variables
-static struct dynabuf	*user_macro_name;	// original macro title
-static struct dynabuf	*internal_name;		// plus param type chars
+static	STRUCT_DYNABUF_REF(user_macro_name, NAME_INITIALSIZE);	// original macro title
+static	STRUCT_DYNABUF_REF(internal_name, NAME_INITIALSIZE);	// plus param type chars
 static struct rwnode	*macro_forest[256];	// trees (because of 8b hash)
 // Dynamic argument table
 static union macro_arg_t	*arg_table	= NULL;
@@ -62,19 +60,12 @@ static int			argtable_size	= HALF_INITIAL_ARG_TABLE_SIZE;
 static void enlarge_arg_table(void)
 {
 	argtable_size *= 2;
+	//printf("Doubling arg table size to %d.\n", argtable_size);
 	arg_table = realloc(arg_table, argtable_size * sizeof(*arg_table));
 	if (arg_table == NULL)
 		Throw_serious_error(exception_no_memory_left);
 }
 
-// create dynamic buffers and arg table
-void Macro_init(void)
-{
-	user_macro_name = DynaBuf_create(MACRONAME_DYNABUF_INITIALSIZE);
-	internal_name = DynaBuf_create(MACRONAME_DYNABUF_INITIALSIZE);
-	enlarge_arg_table();
-}
-
 // Read macro scope and title. Title is read to GlobalDynaBuf and then copied
 // over to internal_name DynaBuf, where ARG_SEPARATOR is added.
 // In user_macro_name DynaBuf, the original name is reconstructed (even with
@@ -88,8 +79,10 @@ static scope_t get_scope_and_title(void)
 	// copy macro title to private dynabuf and add separator character
 	DYNABUF_CLEAR(user_macro_name);
 	DYNABUF_CLEAR(internal_name);
-	if (macro_scope != SCOPE_GLOBAL)
+	if (macro_scope != SCOPE_GLOBAL) {
+		// TODO - allow "cheap macros"?!
 		DynaBuf_append(user_macro_name, LOCAL_PREFIX);
+	}
 	DynaBuf_add_string(user_macro_name, GLOBALDYNABUF_CURRENT);
 	DynaBuf_add_string(internal_name, GLOBALDYNABUF_CURRENT);
 	DynaBuf_append(user_macro_name, '\0');
@@ -174,6 +167,8 @@ void Macro_parse_definition(void)	// Now GotByte = illegal char after "!macro"
 	// Valid argument formats are:
 	// .LOCAL_LABEL_BY_VALUE
 	// ~.LOCAL_LABEL_BY_REFERENCE
+	// @CHEAP_LOCAL_LABEL_BY_VALUE
+	// ~@CHEAP_LOCAL_LABEL_BY_REFERENCE
 	// GLOBAL_LABEL_BY_VALUE	global args are very uncommon,
 	// ~GLOBAL_LABEL_BY_REFERENCE	but not forbidden
 	// now GotByte = non-space
@@ -181,15 +176,16 @@ void Macro_parse_definition(void)	// Now GotByte = illegal char after "!macro"
 		do {
 			// handle call-by-reference character ('~')
 			if (GotByte != REFERENCE_CHAR) {
-				DynaBuf_append(internal_name, ARGTYPE_NUM_VAL);
+				DynaBuf_append(internal_name, ARGTYPE_VALUE);
 			} else {
-				DynaBuf_append(internal_name, ARGTYPE_NUM_REF);
+				DynaBuf_append(internal_name, ARGTYPE_REF);
 				DynaBuf_append(GlobalDynaBuf, REFERENCE_CHAR);
 				GetByte();
 			}
-			// handle prefix for local symbols (LOCAL_PREFIX, normally '.')
-			if (GotByte == LOCAL_PREFIX) {
-				DynaBuf_append(GlobalDynaBuf, LOCAL_PREFIX);
+			// handle prefix for (cheap) local symbols ('.'/'@')
+			if ((GotByte == LOCAL_PREFIX)
+			|| (GotByte == CHEAP_PREFIX)) {
+				DynaBuf_append(GlobalDynaBuf, GotByte);
 				GetByte();
 			}
 			// handle symbol name
@@ -238,6 +234,10 @@ void Macro_parse_call(void)	// Now GotByte = dot or first char of macro name
 	int		arg_count	= 0;
 	int		outer_err_count;
 
+	// make sure arg_table is ready (if not yet initialised, do it now)
+	if (arg_table == NULL)
+		enlarge_arg_table();
+
 	// Enter deeper nesting level
 	// Quit program if recursion too deep.
 	if (--macro_recursions_left < 0)
@@ -247,9 +247,8 @@ void Macro_parse_call(void)	// Now GotByte = dot or first char of macro name
 	// internal_name = MacroTitle ARG_SEPARATOR (grows to signature)
 	// Accept n>=0 comma-separated arguments before CHAR_EOS.
 	// Valid argument formats are:
-	// EXPRESSION (everything that does NOT start with '~'
-	// ~.LOCAL_LABEL_BY_REFERENCE
-	// ~GLOBAL_LABEL_BY_REFERENCE
+	//	~SYMBOL		call by ref
+	//	EXPRESSION	call by value (everything that does NOT start with '~')
 	// now GotByte = non-space
 	if (GotByte != CHAR_EOS) {	// any at all?
 		do {
@@ -260,14 +259,14 @@ void Macro_parse_call(void)	// Now GotByte = dot or first char of macro name
 			// In both cases, GlobalDynaBuf may be used.
 			if (GotByte == REFERENCE_CHAR) {
 				// read call-by-reference arg
-				DynaBuf_append(internal_name, ARGTYPE_NUM_REF);
-				GetByte();	// skip '~' character
+				DynaBuf_append(internal_name, ARGTYPE_REF);
+				GetByte();	// eat '~'
 				Input_read_scope_and_keyword(&symbol_scope);
 				// GotByte = illegal char
-				arg_table[arg_count].symbol = symbol_find(symbol_scope, 0);
+				arg_table[arg_count].symbol = symbol_find(symbol_scope);	// CAUTION, object type may be NULL!
 			} else {
 				// read call-by-value arg
-				DynaBuf_append(internal_name, ARGTYPE_NUM_VAL);
+				DynaBuf_append(internal_name, ARGTYPE_VALUE);
 				ALU_any_result(&(arg_table[arg_count].result));
 			}
 			++arg_count;
@@ -289,7 +288,7 @@ void Macro_parse_call(void)	// Now GotByte = dot or first char of macro name
 		// set up new input
 		new_input.original_filename = actual_macro->def_filename;
 		new_input.line_number = actual_macro->def_line_number;
-		new_input.source_is_ram = TRUE;
+		new_input.source = INPUTSRC_RAM;
 		new_input.state = INPUTSTATE_NORMAL;	// FIXME - fix others!
 		new_input.src.ram_ptr = actual_macro->parameter_list;
 		// remember old input
@@ -304,6 +303,7 @@ void Macro_parse_call(void)	// Now GotByte = dot or first char of macro name
 		// start new section (with new scope)
 		// FALSE = title mustn't be freed
 		section_new(&new_section, "Macro", actual_macro->original_name, FALSE);
+		section_new_cheap_scope(&new_section);
 		GetByte();	// fetch first byte of parameter list
 		// assign arguments
 		if (GotByte != CHAR_EOS) {	// any at all?
@@ -314,20 +314,21 @@ void Macro_parse_call(void)	// Now GotByte = dot or first char of macro name
 				// In both cases, GlobalDynaBuf may be used.
 				if (GotByte == REFERENCE_CHAR) {
 					// assign call-by-reference arg
-					GetByte();	// skip '~' character
+					GetByte();	// eat '~'
 					Input_read_scope_and_keyword(&symbol_scope);
+					// create new tree node and link existing symbol struct from arg list to it
 					if ((Tree_hard_scan(&symbol_node, symbols_forest, symbol_scope, TRUE) == FALSE)
-					&& (pass_count == 0))
+					&& (FIRST_PASS))
 						Throw_error("Macro parameter twice.");
-					symbol_node->body = arg_table[arg_count].symbol;
+					symbol_node->body = arg_table[arg_count].symbol;	// CAUTION, object type may be NULL
 				} else {
 					// assign call-by-value arg
 					Input_read_scope_and_keyword(&symbol_scope);
-					symbol = symbol_find(symbol_scope, 0);
-// FIXME - add a possibility to symbol_find to make it possible to find out
-// whether symbol was just created. Then check for the same error message here
-// as above ("Macro parameter twice.").
-					symbol->result = arg_table[arg_count].result;
+					symbol = symbol_find(symbol_scope);
+// FIXME - find out if symbol was just created.
+// Then check for the same error message here as above ("Macro parameter twice.").
+// TODO - on the other hand, this would rule out globals as args (stupid anyway, but not illegal yet!)
+					symbol->object = arg_table[arg_count].result;	// FIXME - this assignment redefines globals/whatever without throwing errors!
 				}
 				++arg_count;
 			} while (Input_accept_comma());

src/macro.h

@@ -12,8 +12,6 @@
 
 // Prototypes
 
-// create dynamic buffers and arg table
-extern void Macro_init(void);	// create private dynabuf
 // only call once (during first pass)
 extern void Macro_parse_definition(void);
 // Parse macro call ("+MACROTITLE"). Has to be re-entrant.

src/mnemo.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // mnemonic definitions
@@ -7,26 +7,29 @@
 #define mnemo_H
 
 
-// create dynamic buffer, build keyword trees
-extern void Mnemo_init(void);
-// check whether mnemonic in GlobalDynaBuf is supported by 6502 cpu.
-extern int keyword_is_6502_mnemo(int length);
-// check whether mnemonic in GlobalDynaBuf is supported by 6510 cpu.
-extern int keyword_is_6510_mnemo(int length);
+#include "config.h"
+
+
+// check whether mnemonic in GlobalDynaBuf is supported by standard 6502 cpu.
+extern boolean keyword_is_6502_mnemo(int length);
+// check whether mnemonic in GlobalDynaBuf is supported by NMOS 6502 cpu (includes undocumented opcodes).
+extern boolean keyword_is_nmos6502_mnemo(int length);
 // check whether mnemonic in GlobalDynaBuf is supported by C64DTV2 cpu.
-extern int keyword_is_c64dtv2_mnemo(int length);
-// check whether mnemonic in GlobalDynaBuf is supported by 65c02 cpu.
-extern int keyword_is_65c02_mnemo(int length);
-// check whether mnemonic in GlobalDynaBuf is supported by Rockwell 65c02 cpu.
-extern int keyword_is_r65c02_mnemo(int length);
-// check whether mnemonic in GlobalDynaBuf is supported by WDC 65c02 cpu.
-extern int keyword_is_w65c02_mnemo(int length);
+extern boolean keyword_is_c64dtv2_mnemo(int length);
+// check whether mnemonic in GlobalDynaBuf is supported by 65C02 cpu.
+extern boolean keyword_is_65c02_mnemo(int length);
+// check whether mnemonic in GlobalDynaBuf is supported by Rockwell 65C02 cpu.
+extern boolean keyword_is_r65c02_mnemo(int length);
+// check whether mnemonic in GlobalDynaBuf is supported by WDC 65C02 cpu.
+extern boolean keyword_is_w65c02_mnemo(int length);
 // check whether mnemonic in GlobalDynaBuf is supported by 65816 cpu.
-extern int keyword_is_65816_mnemo(int length);
-// check whether mnemonic in GlobalDynaBuf is supported by CSG 65ce02 cpu.
-extern int keyword_is_65ce02_mnemo(int length);
+extern boolean keyword_is_65816_mnemo(int length);
+// check whether mnemonic in GlobalDynaBuf is supported by CSG 65CE02 cpu.
+extern boolean keyword_is_65ce02_mnemo(int length);
 // check whether mnemonic in GlobalDynaBuf is supported by CSG 4502 cpu.
-extern int keyword_is_4502_mnemo(int length);
+extern boolean keyword_is_4502_mnemo(int length);
+// check whether mnemonic in GlobalDynaBuf is supported by MEGA65 cpu.
+extern boolean keyword_is_m65_mnemo(int length);
 
 
 #endif

src/output.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Output stuff
@@ -8,6 +8,7 @@
 //  5 Mar 2014	Fixed bug where setting *>0xffff resulted in hangups.
 // 19 Nov 2014	Merged Johann Klasek's report listing generator patch
 // 22 Sep 2015	Added big-endian output functions
+// 20 Apr 2019	Prepared for "make segment overlap warnings into errors" later on
 #include "output.h"
 #include <stdlib.h>
 #include <string.h>	// for memset()
@@ -23,7 +24,6 @@
 
 
 // constants
-#define OUTBUFFERSIZE		65536
 #define NO_SEGMENT_START	(-1)	// invalid value to signal "not in a segment"
 
 
@@ -38,19 +38,24 @@ struct segment {
 // structure for all output stuff:
 struct output {
 	// output buffer stuff
+	intval_t	bufsize;	// either 64 KiB or 16 MiB
 	char		*buffer;	// holds assembled code
 	intval_t	write_idx;	// index of next write
 	intval_t	lowest_written;		// smallest address used
 	intval_t	highest_written;	// largest address used
-	int		initvalue_set;	// actually bool
+	boolean		initvalue_set;
 	struct {
 		intval_t	start;	// start of current segment (or NO_SEGMENT_START)
 		intval_t	max;	// highest address segment may use
-		int		flags;	// segment flags ("overlay" and "invisible", see header file)
+		bits		flags;	// segment flags ("overlay" and "invisible", see header file)
 		struct segment	list_head;	// head element of doubly-linked ring list
 	} segment;
+	char		xor;		// output modifier
 };
 
+// for offset assembly:
+static struct pseudopc	*pseudopc_current_context;	// current struct (NULL when not in pseudopc block)
+
 
 // variables
 static struct output	default_output;
@@ -67,13 +72,14 @@ enum output_format {
 	OUTPUT_FORMAT_PLAIN		// code only
 };
 // predefined stuff
-static struct ronode	*file_format_tree	= NULL;	// tree to hold output formats (FIXME - a tree for three items, really?)
-static struct ronode	file_format_list[]	= {
+// tree to hold output formats (FIXME - a tree for three items, really?)
+static struct ronode	file_format_tree[]	= {
+	PREDEF_START,
 #define KNOWN_FORMATS	"'plain', 'cbm', 'apple'"	// shown in CLI error message for unknown formats
 	PREDEFNODE("apple",	OUTPUT_FORMAT_APPLE),
-	PREDEFNODE(s_cbm,	OUTPUT_FORMAT_CBM),
+	PREDEFNODE("cbm",	OUTPUT_FORMAT_CBM),
 //	PREDEFNODE("o65",	OUTPUT_FORMAT_O65),
-	PREDEFLAST("plain",	OUTPUT_FORMAT_PLAIN),
+	PREDEF_END("plain",	OUTPUT_FORMAT_PLAIN),
 	//    ^^^^ this marks the last element
 };
 // chosen file format
@@ -105,7 +111,7 @@ static void find_segment_max(intval_t new_pc)
 	while (test_segment->start <= new_pc)
 		test_segment = test_segment->next;
 	if (test_segment == &out->segment.list_head)
-		out->segment.max = OUTBUFFERSIZE - 1;
+		out->segment.max = out->bufsize - 1;
 	else
 		out->segment.max = test_segment->start - 1;	// last free address available
 }
@@ -114,10 +120,15 @@ static void find_segment_max(intval_t new_pc)
 //
 static void border_crossed(int current_offset)
 {
-	if (current_offset >= OUTBUFFERSIZE)
+	if (current_offset >= out->bufsize)
 		Throw_serious_error("Produced too much code.");
-	if (pass_count == 0) {
-		Throw_warning("Segment reached another one, overwriting it.");
+	// TODO - get rid of FIRST_PASS condition, because user can suppress these warnings if they want
+	if (FIRST_PASS) {
+		// TODO: make warn/err an arg for a general "Throw" function
+		if (config.segment_warning_is_error)
+			Throw_error("Segment reached another one, overwriting it.");
+		else
+			Throw_warning("Segment reached another one, overwriting it.");
 		find_segment_max(current_offset + 1);	// find new (next) limit
 	}
 }
@@ -130,7 +141,9 @@ void (*Output_byte)(intval_t byte);
 // send low byte to output buffer, automatically increasing program counter
 static void real_output(intval_t byte)
 {
-	// did we reach segment limit?
+	// CAUTION - there are two copies of these checks!
+	// TODO - add additional check for current segment's "limit" value
+	// did we reach next segment?
 	if (out->write_idx > out->segment.max)
 		border_crossed(out->write_idx);
 	// new minimum address?
@@ -142,7 +155,7 @@ static void real_output(intval_t byte)
 	// write byte and advance ptrs
 	if (report->fd)
 		report_binary(byte & 0xff);	// file for reporting, taking also CPU_2add
-	out->buffer[out->write_idx++] = byte & 0xff;
+	out->buffer[out->write_idx++] = (byte & 0xff) ^ out->xor;
 	++CPU_state.add_to_pc;
 }
 
@@ -157,20 +170,25 @@ static void no_output(intval_t byte)
 }
 
 
-// call this if really calling Output_byte would be a waste of time
-// FIXME - check all users of this, because future changes
-// ("several-projects-at-once") may be incompatible with this!
-void Output_fake(int size)
+// skip over some bytes in output buffer without starting a new segment
+// (used by "!skip", and also called by "!binary" if really calling
+// Output_byte would be a waste of time)
+void output_skip(int size)
 {
-	if (size < 1)
+	if (size < 1) {
+		// FIXME - ok for zero, but why is there no error message
+		// output for negative values?
 		return;
+	}
 
 	// check whether ptr undefined
 	if (Output_byte == no_output) {
 		Output_byte(0);	// trigger error with a dummy byte
 		--size;	// fix amount to cater for dummy byte
 	}
-	// did we reach segment limit?
+	// CAUTION - there are two copies of these checks!
+	// TODO - add additional check for current segment's "limit" value
+	// did we reach next segment?
 	if (out->write_idx + size - 1 > out->segment.max)
 		border_crossed(out->write_idx + size - 1);
 	// new minimum address?
@@ -185,98 +203,10 @@ void Output_fake(int size)
 }
 
 
-// output 8-bit value with range check
-void output_8(intval_t value)
-{
-	if ((value <= 0xff) && (value >= -0x80))
-		Output_byte(value);
-	else
-		Throw_error(exception_number_out_of_range);
-}
-
-
-// output 16-bit value with range check big-endian
-void output_be16(intval_t value)
-{
-	if ((value <= 0xffff) && (value >= -0x8000)) {
-		Output_byte(value >> 8);
-		Output_byte(value);
-	} else {
-		Throw_error(exception_number_out_of_range);
-	}
-}
-
-
-// output 16-bit value with range check little-endian
-void output_le16(intval_t value)
-{
-	if ((value <= 0xffff) && (value >= -0x8000)) {
-		Output_byte(value);
-		Output_byte(value >> 8);
-	} else {
-		Throw_error(exception_number_out_of_range);
-	}
-}
-
-
-// output 24-bit value with range check big-endian
-void output_be24(intval_t value)
-{
-	if ((value <= 0xffffff) && (value >= -0x800000)) {
-		Output_byte(value >> 16);
-		Output_byte(value >> 8);
-		Output_byte(value);
-	} else {
-		Throw_error(exception_number_out_of_range);
-	}
-}
-
-
-// output 24-bit value with range check little-endian
-void output_le24(intval_t value)
-{
-	if ((value <= 0xffffff) && (value >= -0x800000)) {
-		Output_byte(value);
-		Output_byte(value >> 8);
-		Output_byte(value >> 16);
-	} else {
-		Throw_error(exception_number_out_of_range);
-	}
-}
-
-
-// output 32-bit value (without range check) big-endian
-void output_be32(intval_t value)
-{
-//  if ((Value <= 0x7fffffff) && (Value >= -0x80000000)) {
-	Output_byte(value >> 24);
-	Output_byte(value >> 16);
-	Output_byte(value >> 8);
-	Output_byte(value);
-//  } else {
-//	Throw_error(exception_number_out_of_range);
-//  }
-}
-
-
-// output 32-bit value (without range check) little-endian
-void output_le32(intval_t value)
-{
-//  if ((Value <= 0x7fffffff) && (Value >= -0x80000000)) {
-	Output_byte(value);
-	Output_byte(value >> 8);
-	Output_byte(value >> 16);
-	Output_byte(value >> 24);
-//  } else {
-//	Throw_error(exception_number_out_of_range);
-//  }
-}
-
-
 // fill output buffer with given byte value
 static void fill_completely(char value)
 {
-	memset(out->buffer, value, OUTBUFFERSIZE);
+	memset(out->buffer, value, out->bufsize);
 }
 
 
@@ -294,8 +224,10 @@ int output_initmem(char content)
 	// init memory
 	fill_completely(content);
 	// enforce another pass
-	if (pass_undefined_count == 0)
-		pass_undefined_count = 1;
+	if (pass.undefined_count == 0)
+		pass.undefined_count = 1;
+	//if (pass.needvalue_count == 0)	FIXME - use? instead or additionally?
+	//	pass.needvalue_count = 1;
 // FIXME - enforcing another pass is not needed if there hasn't been any
 // output yet. But that's tricky to detect without too much overhead.
 // The old solution was to add &&(out->lowest_written < out->highest_written+1) to "if" above
@@ -308,9 +240,6 @@ int outputfile_set_format(void)
 {
 	void	*node_body;
 
-	// make sure tree is initialised
-	if (file_format_tree == NULL)
-		Tree_add_table(&file_format_tree, file_format_list);
 	// perform lookup
 	if (!Tree_easy_scan(file_format_tree, &node_body, GlobalDynaBuf))
 		return 1;
@@ -325,6 +254,7 @@ int outputfile_prefer_cbm_format(void)
 {
 	if (output_format != OUTPUT_FORMAT_UNSPECIFIED)
 		return 0;
+
 	output_format = OUTPUT_FORMAT_CBM;
 	return 1;
 }
@@ -346,9 +276,10 @@ int outputfile_set_filename(void)
 
 
 // init output struct (done later)
-void Output_init(signed long fill_value)
+void Output_init(signed long fill_value, boolean use_large_buf)
 {
-	out->buffer = safe_malloc(OUTBUFFERSIZE);
+	out->bufsize = use_large_buf ? 0x1000000 : 0x10000;
+	out->buffer = safe_malloc(out->bufsize);
 	if (fill_value == MEMINIT_USE_DEFAULT) {
 		fill_value = FILLVALUE_INITIAL;
 		out->initvalue_set = FALSE;
@@ -377,7 +308,7 @@ void Output_save_file(FILE *fd)
 		start = out->lowest_written;
 		amount = out->highest_written - start + 1;
 	}
-	if (Process_verbosity)
+	if (config.process_verbosity)
 		printf("Saving %ld (0x%lx) bytes (0x%lx - 0x%lx exclusive).\n",
 			amount, amount, start, start + amount);
 	// output file header according to file format
@@ -432,7 +363,7 @@ static void link_segment(intval_t start, intval_t length)
 
 
 // check whether given PC is inside segment.
-// only call in first pass, otherwise too many warnings might be thrown
+// only call in first pass, otherwise too many warnings might be thrown	(TODO - still?)
 static void check_segment(intval_t new_pc)
 {
 	struct segment	*test_segment	= out->segment.list_head.next;
@@ -443,7 +374,11 @@ static void check_segment(intval_t new_pc)
 	// search ring for matching entry
 	while (test_segment->start <= new_pc) {
 		if ((test_segment->start + test_segment->length) > new_pc) {
-			Throw_warning("Segment starts inside another one, overwriting it.");
+			// TODO - include overlap size in error message!
+			if (config.segment_warning_is_error)
+				Throw_error("Segment starts inside another one, overwriting it.");
+			else
+				Throw_warning("Segment starts inside another one, overwriting it.");
 			return;
 		}
 
@@ -459,7 +394,7 @@ void Output_passinit(void)
 
 //FIXME - why clear ring list in every pass?
 // Because later pass shouldn't complain about overwriting the same segment from earlier pass!
-// Currently this does not happen because segment checks are only done in first pass. FIXME!
+// Currently this does not happen because segment warnings are only generated in first pass. FIXME!
 	// delete segment list (and free blocks)
 //	while ((temp = segment_list)) {
 //		segment_list = segment_list->next;
@@ -467,21 +402,26 @@ void Output_passinit(void)
 //	}
 
 	// invalidate start and end (first byte actually written will fix them)
-	out->lowest_written = OUTBUFFERSIZE - 1;
+	out->lowest_written = out->bufsize - 1;
 	out->highest_written = 0;
 	// deactivate output - any byte written will trigger error:
 	Output_byte = no_output;
 	out->write_idx = 0;	// same as pc on pass init!
 	out->segment.start = NO_SEGMENT_START;	// TODO - "no active segment" could be made a segment flag!
-	out->segment.max = OUTBUFFERSIZE - 1;
+	out->segment.max = out->bufsize - 1;	// TODO - use end of bank?
 	out->segment.flags = 0;
+	out->xor = 0;
 
 	//vcpu stuff:
-	CPU_state.pc.flags = 0;	// not defined yet
+	CPU_state.pc.ntype = NUMTYPE_UNDEFINED;	// not defined yet
+	CPU_state.pc.flags = 0;
+	// FIXME - number type is "undefined", but still the intval 0 below will
+	// be used to calculate diff when pc is first set.
 	CPU_state.pc.val.intval = 0;	// same as output's write_idx on pass init
 	CPU_state.add_to_pc = 0;	// increase PC by this at end of statement
-	CPU_state.a_is_long = FALSE;	// short accu
-	CPU_state.xy_are_long = FALSE;	// short index regs
+
+	// pseudopc stuff:
+	pseudopc_current_context = NULL;
 }
 
 
@@ -492,7 +432,7 @@ void Output_end_segment(void)
 	intval_t	amount;
 
 	// in later passes, ignore completely
-	if (pass_count)
+	if (!FIRST_PASS)
 		return;
 
 	// if there is no segment, there is nothing to do
@@ -511,26 +451,30 @@ void Output_end_segment(void)
 	// link to segment list
 	link_segment(out->segment.start, amount);
 	// announce
-	if (Process_verbosity > 1)
+	if (config.process_verbosity > 1)
+		// TODO - change output to start, limit, size, name:
+		// TODO - output hex numbers as %04x? What about limit 0x10000?
 		printf("Segment size is %ld (0x%lx) bytes (0x%lx - 0x%lx exclusive).\n",
 			amount, amount, out->segment.start, out->write_idx);
 }
 
 
 // change output pointer and enable output
-void Output_start_segment(intval_t address_change, int segment_flags)
+// TODO - this only gets called from vcpu_set_pc so could be made static!
+void Output_start_segment(intval_t address_change, bits segment_flags)
 {
 	// properly finalize previous segment (link to list, announce)
 	Output_end_segment();
 
 	// calculate start of new segment
-	out->write_idx = (out->write_idx + address_change) & 0xffff;
+	out->write_idx = (out->write_idx + address_change) & (out->bufsize - 1);
 	out->segment.start = out->write_idx;
 	out->segment.flags = segment_flags;
 	// allow writing to output buffer
 	Output_byte = real_output;
 	// in first pass, check for other segments and maybe issue warning
-	if (pass_count == 0) {
+	// TODO - remove FIRST_PASS condition
+	if (FIRST_PASS) {
 		if (!(segment_flags & SEGMENT_FLAG_OVERLAY))
 			check_segment(out->segment.start);
 		find_segment_max(out->segment.start);
@@ -538,19 +482,43 @@ void Output_start_segment(intval_t address_change, int segment_flags)
 }
 
 
+char output_get_xor(void)
+{
+	return out->xor;
+}
+void output_set_xor(char xor)
+{
+	out->xor = xor;
+}
+
+
 // set program counter to defined value (FIXME - allow for undefined!)
 // if start address was given on command line, main loop will call this before each pass.
-// in addition to that, it will be called on each "* = VALUE".
-void vcpu_set_pc(intval_t new_pc, int segment_flags)
+// in addition to that, it will be called on each "*= VALUE".
+void vcpu_set_pc(intval_t new_pc, bits segment_flags)
 {
-	intval_t	new_offset;
+	intval_t	pc_change;
 
-	new_offset = (new_pc - CPU_state.pc.val.intval) & 0xffff;
-	CPU_state.pc.val.intval = new_pc;
-	CPU_state.pc.flags |= MVALUE_DEFINED;	// FIXME - remove when allowing undefined!
+	// support stupidly bad, old, ancient, deprecated, obsolete behaviour:
+	if (pseudopc_current_context != NULL) {
+		if (config.wanted_version < VER_SHORTER_SETPC_WARNING) {
+			Throw_warning("Offset assembly still active at end of segment. Switched it off.");
+			pseudopc_end_all();
+		} else if (config.wanted_version < VER_DISABLED_OBSOLETE_STUFF) {
+			Throw_warning("Offset assembly still active at end of segment.");
+			pseudopc_end_all();	// warning no longer said it
+			// would switch off, but still did. nevertheless, there
+			// is something different to older versions: when the
+			// closing '}' or !realpc is encountered, _really_ weird
+			// stuff happens! i see no reason to try to mimic that.
+		}
+	}
+	pc_change = new_pc - CPU_state.pc.val.intval;
+	CPU_state.pc.val.intval = new_pc;	// FIXME - oversized values are accepted without error and will be wrapped at end of statement!
+	CPU_state.pc.ntype = NUMTYPE_INT;	// FIXME - remove when allowing undefined!
 	CPU_state.pc.addr_refs = 1;	// yes, PC counts as address
 	// now tell output buffer to start a new segment
-	Output_start_segment(new_offset, segment_flags);
+	Output_start_segment(pc_change, segment_flags);
 }
 /*
 TODO - overhaul program counter and memory pointer stuff:
@@ -567,7 +535,7 @@ when encountering "!pseudopc VALUE { BLOCK }":
 	remember difference between current and new value
 	set PC to new value
 	after BLOCK, use remembered difference to change PC back
-when encountering "* = VALUE":
+when encountering "*= VALUE":
 	parse new value (NEW: might be undefined!)
 	calculate difference between current PC and new value
 	set PC to new value
@@ -576,14 +544,14 @@ when encountering "* = VALUE":
 Problem: always check for "undefined"; there are some problematic combinations.
 I need a way to return the size of a generated code block even if PC undefined.
 Maybe like this:
-	* = new_address [, invisible] [, overlay] [, &size_symbol_ref {]
+	*= new_address [, invisible] [, overlay] [, &size_symbol_ref {]
 		...code...
 	[} ; at end of block, size is written to size symbol given above!]
 */
 
 
 // get program counter
-void vcpu_read_pc(struct result *target)
+void vcpu_read_pc(struct number *target)
 {
 	*target = CPU_state.pc;
 }
@@ -599,6 +567,77 @@ int vcpu_get_statement_size(void)
 // adjust program counter (called at end of each statement)
 void vcpu_end_statement(void)
 {
-	CPU_state.pc.val.intval = (CPU_state.pc.val.intval + CPU_state.add_to_pc) & 0xffff;
+	CPU_state.pc.val.intval = (CPU_state.pc.val.intval + CPU_state.add_to_pc) & (out->bufsize - 1);
 	CPU_state.add_to_pc = 0;
 }
+
+
+// struct to describe a pseudopc context
+struct pseudopc {
+	struct pseudopc	*outer;	// next layer (to be able to "unpseudopc" labels by more than one level)
+	intval_t	offset;	// inner minus outer pc
+	enum numtype	ntype;	// type of outer pc (INT/UNDEFINED)
+};
+// start offset assembly
+void pseudopc_start(struct number *new_pc)
+{
+	struct pseudopc	*new_context;
+
+	new_context = safe_malloc(sizeof(*new_context));	// create new struct (this must never be freed, as it gets linked to labels!)
+	new_context->outer = pseudopc_current_context;	// let it point to previous one
+	pseudopc_current_context = new_context;	// make it the current one
+
+	new_context->ntype = CPU_state.pc.ntype;
+	new_context->offset = new_pc->val.intval - CPU_state.pc.val.intval;
+	CPU_state.pc.val.intval = new_pc->val.intval;
+	CPU_state.pc.ntype = NUMTYPE_INT;	// FIXME - remove when allowing undefined!
+	//new: CPU_state.pc.flags = new_pc->flags & (NUMBER_IS_DEFINED | NUMBER_EVER_UNDEFINED);
+}
+// end offset assembly
+void pseudopc_end(void)
+{
+	if (pseudopc_current_context == NULL) {
+		// trying to end offset assembly though it isn't active:
+		// in current versions this cannot happen and so must be a bug.
+		// but in versions older than 0.94.8 this was possible using
+		// !realpc, and offset assembly got automatically disabled when
+		// encountering "*=".
+		// so if wanted version is new enough, choke on bug!
+		if (config.wanted_version >= VER_DISABLED_OBSOLETE_STUFF)
+			Bug_found("ClosingUnopenedPseudopcBlock", 0);
+	} else {
+		CPU_state.pc.val.intval = (CPU_state.pc.val.intval - pseudopc_current_context->offset) & (out->bufsize - 1);	// pc might have wrapped around
+		CPU_state.pc.ntype = pseudopc_current_context->ntype;
+		pseudopc_current_context = pseudopc_current_context->outer;	// go back to outer block
+	}
+}
+// this is only for old, deprecated, obsolete, stupid "realpc":
+void pseudopc_end_all(void)
+{
+	while (pseudopc_current_context)
+		pseudopc_end();
+}
+// un-pseudopc a label value by given number of levels
+// returns nonzero on error (if level too high)
+int pseudopc_unpseudo(struct number *target, struct pseudopc *context, unsigned int levels)
+{
+	while (levels--) {
+		//if (target->ntype == NUMTYPE_UNDEFINED)
+		//	return 0;	// ok (no sense in trying to unpseudo this, and it might be an unresolved forward ref anyway)
+
+		if (context == NULL) {
+			Throw_error("Un-pseudopc operator '&' has no !pseudopc context.");
+			return 1;	// error
+		}
+		// FIXME - in future, check both target and context for NUMTYPE_UNDEFINED!
+		target->val.intval = (target->val.intval - context->offset) & (out->bufsize - 1);	// FIXME - is masking really needed?	TODO
+		context = context->outer;
+	}
+	return 0;	// ok
+}
+// return pointer to current "pseudopc" struct (may be NULL!)
+// this gets called when parsing label definitions
+struct pseudopc *pseudopc_get_context(void)
+{
+	return pseudopc_current_context;
+}

src/output.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // Output stuff (FIXME - split into outbuf, outfile/format and vcpu parts)
@@ -19,13 +19,13 @@
 
 
 // current CPU state
-// FIXME - move vcpu struct definition to .c file and change other .c files' accesses to fn calls
+// FIXME - move vcpu struct definition to .c file and change other .c files' accesses to fn calls. then replace "struct number" with minimized version.
 struct vcpu {
 	const struct cpu_type	*type;		// current CPU type (default 6502)	(FIXME - move out of struct again?)
-	struct result		pc;		// current program counter (pseudo value)
+	struct number		pc;		// current program counter (pseudo value)
 	int			add_to_pc;	// add to PC after statement
-	int			a_is_long;
-	int			xy_are_long;
+	boolean			a_is_long;
+	boolean			xy_are_long;
 };
 
 
@@ -35,31 +35,27 @@ extern struct vcpu	CPU_state;	// current CPU state	FIXME - restrict visibility t
 
 // Prototypes
 
-// alloc and init mem buffer (done later)
-extern void Output_init(signed long fill_value);
 // clear segment list and disable output
+//TODO - does this belong to outbuf stuff?
 extern void Output_passinit(void);
-// call this if really calling Output_byte would be a waste of time
-extern void Output_fake(int size);
+
+// outbuf stuff:
+
+// alloc and init mem buffer (done later)
+extern void Output_init(signed long fill_value, boolean use_large_buf);
+// skip over some bytes in output buffer without starting a new segment
+// (used by "!skip", and also called by "!binary" if really calling
+// Output_byte would be a waste of time)
+extern void output_skip(int size);
 // Send low byte of arg to output buffer and advance pointer
+// FIXME - replace by output_sequence(char *src, size_t size)
 extern void (*Output_byte)(intval_t);
-// Output 8-bit value with range check
-extern void output_8(intval_t);
-// Output 16-bit value with range check big-endian
-extern void output_be16(intval_t);
-// Output 16-bit value with range check little-endian
-extern void output_le16(intval_t);
-// Output 24-bit value with range check big-endian
-extern void output_be24(intval_t);
-// Output 24-bit value with range check little-endian
-extern void output_le24(intval_t);
-// Output 32-bit value (without range check) big-endian
-extern void output_be32(intval_t);
-// Output 32-bit value (without range check) little-endian
-extern void output_le32(intval_t);
 // define default value for empty memory ("!initmem" pseudo opcode)
 // returns zero if ok, nonzero if already set
 extern int output_initmem(char content);
+
+// outfile stuff:
+
 // try to set output format held in DynaBuf. Returns zero on success.
 extern int outputfile_set_format(void);
 extern const char	outputfile_formats[];	// string to show if outputfile_set_format() returns nonzero
@@ -71,18 +67,34 @@ extern int outputfile_set_filename(void);
 // write smallest-possible part of memory buffer to file
 extern void Output_save_file(FILE *fd);
 // change output pointer and enable output
-extern void Output_start_segment(intval_t address_change, int segment_flags);
+extern void Output_start_segment(intval_t address_change, bits segment_flags);
 // Show start and end of current segment
 extern void Output_end_segment(void);
+extern char output_get_xor(void);
+extern void output_set_xor(char xor);
 
 // set program counter to defined value (TODO - allow undefined!)
-extern void vcpu_set_pc(intval_t new_pc, int flags);
+extern void vcpu_set_pc(intval_t new_pc, bits flags);
 // get program counter
-extern void vcpu_read_pc(struct result *target);
+extern void vcpu_read_pc(struct number *target);
 // get size of current statement (until now) - needed for "!bin" verbose output
 extern int vcpu_get_statement_size(void);
 // adjust program counter (called at end of each statement)
 extern void vcpu_end_statement(void);
 
+struct pseudopc;
+// start offset assembly
+extern void pseudopc_start(struct number *new_pc);
+// end offset assembly
+extern void pseudopc_end(void);
+// this is only for old, deprecated, obsolete, stupid "realpc":
+extern void pseudopc_end_all(void);
+// un-pseudopc a label value by given number of levels
+// returns nonzero on error (if level too high)
+extern int pseudopc_unpseudo(struct number *target, struct pseudopc *context, unsigned int levels);
+// return pointer to current "pseudopc" struct (may be NULL!)
+// this gets called when parsing label definitions
+extern struct pseudopc *pseudopc_get_context(void);
+
 
 #endif

src/pseudoopcodes.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // pseudo opcode stuff
@@ -7,10 +7,8 @@
 #define pseudoopcodes_H
 
 
-// call when "* = EXPRESSION" is parsed
+// call when "*= EXPRESSION" is parsed
 extern void notreallypo_setpc(void);
-// register pseudo opcodes
-extern void pseudoopcodes_init(void);
 // parse pseudo opcode. has to be re-entrant.
 extern void pseudoopcode_parse(void);
 

src/section.c

@@ -1,20 +1,23 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // section stuff (move to symbol.h?)
 #include "section.h"
 #include "config.h"
 #include "dynabuf.h"
-#include "global.h"	// FIXME - remove when no longer needed
+#include "global.h"
 #include "input.h"
 #include "symbol.h"
 #include "tree.h"
 
 
+#define SCOPE_INCREMENT	2	// inc by 2 so locals are even and cheaps are odd
+
 // fake section structure (for error msgs before any real section is in use)
 static struct section	initial_section	= {
-	0,		// scope value
+	0,		// local scope value
+	1,		// cheap scope value
 	"during",	// "type"	=> normally "zone Title" or
 	"init",		// "title"	=>  "macro test", now "during init"
 	FALSE,		// no, title was not malloc'd
@@ -24,21 +27,37 @@ static struct section	initial_section	= {
 // variables
 struct section		*section_now	= &initial_section;	// current section
 static struct section	outer_section;	// outermost section struct
-static scope_t		scope_localcount;	// highest scope number yet
+static scope_t		local_scope_max;	// highest scope number yet
+static scope_t		cheap_scope_max;	// highest scope number yet
 
 
 // write given info into given structure and activate it
-void section_new(struct section *section, const char *type, char *title, int allocated)
+void section_new(struct section *section, const char *type, char *title, boolean allocated)
 {
-	section->scope = ++scope_localcount;
+	// new scope for locals
+	local_scope_max += SCOPE_INCREMENT;
+	section->local_scope = local_scope_max;
+	// keep scope for cheap locals
+	section->cheap_scope = section_now->cheap_scope;
+	// copy other data
 	section->type = type;
 	section->title = title;
 	section->allocated = allocated;
 	// activate new section
 	section_now = section;
-	//printf("[new zone %d: %s, %s]\n", section->scope, section->type, section->title);
+	//printf("[new section %d: %s, %s]\n", section->local_scope, section->type, section->title);
 }
 
+
+// change scope of cheap locals in given section
+void section_new_cheap_scope(struct section *section)
+{
+	// new scope for cheap locals
+	cheap_scope_max += SCOPE_INCREMENT;
+	section->cheap_scope = cheap_scope_max;
+}
+
+
 // Tidy up: If necessary, release section title.
 // Warning - the state of the component "Allocd" may have
 // changed in the meantime, so don't rely on a local variable.
@@ -48,9 +67,13 @@ void section_finalize(struct section *section)
 		free(section->title);
 }
 
+
 // setup outermost section
 void section_passinit(void)
 {
-	scope_localcount = SCOPE_GLOBAL;	// will be incremented by next line
-	section_new(&outer_section, s_Zone, s_untitled, FALSE);
+	//printf("[old maxima: locals=%d, cheap=%d]\n", local_scope_max, cheap_scope_max);
+	local_scope_max = 0;	// will be incremented by 2 by next line
+	section_new(&outer_section, "Zone", s_untitled, FALSE);
+	cheap_scope_max = -1;	// will be incremented by 2 by next line
+	section_new_cheap_scope(&outer_section);
 }

src/section.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // section stuff
@@ -12,10 +12,11 @@
 
 // "section" structure type definition
 struct section {
-	scope_t		scope;	// section's scope
+	scope_t		local_scope;	// section's scope for local symbols
+	scope_t		cheap_scope;	// section's scope for cheap locals
 	const char	*type;	// "Zone", "Subzone" or "Macro"
 	char		*title;	// zone title, subzone title or macro title
-	int		allocated;	// whether title was malloc()'d
+	boolean		allocated;	// whether title was malloc()'d
 };
 
 
@@ -24,7 +25,9 @@ extern struct section	*section_now;
 
 
 // write given info into given structure and activate it
-extern void section_new(struct section *section, const char *type, char *title, int allocated);
+extern void section_new(struct section *section, const char *type, char *title, boolean allocated);
+// change scope of cheap locals in given section
+extern void section_new_cheap_scope(struct section *section);
 // setup outermost section
 extern void section_passinit(void);
 // tidy up: if necessary, release section title.

src/symbol.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // symbol stuff
@@ -12,7 +12,7 @@
 #include "acme.h"
 #include "alu.h"
 #include "dynabuf.h"
-#include "global.h"	// FIXME - remove when no longer needed
+#include "global.h"
 #include "input.h"
 #include "output.h"
 #include "platform.h"
@@ -30,34 +30,40 @@ static void dump_one_symbol(struct rwnode *node, FILE *fd)
 {
 	struct symbol	*symbol	= node->body;
 
+	// if symbol is neither int nor float, skip
+	if (symbol->object.type != &type_number)
+		return;
+
+	// CAUTION: if more types are added, check for NULL before using type pointer!
+
 	// output name
-	if (warn_on_type_mismatch
-	&& symbol->result.addr_refs == 1)
+	if (config.warn_on_type_mismatch
+	&& symbol->object.u.number.addr_refs == 1)
 		fprintf(fd, "!addr");
 	fprintf(fd, "\t%s", node->id_string);
-	switch (symbol->result.flags & MVALUE_FORCEBITS) {
-	case MVALUE_FORCE16:
+	switch (symbol->object.u.number.flags & NUMBER_FORCEBITS) {
+	case NUMBER_FORCES_16:
 		fprintf(fd, "+2\t= ");
 		break;
-	case MVALUE_FORCE16 | MVALUE_FORCE24:
+	case NUMBER_FORCES_16 | NUMBER_FORCES_24:
 		/*FALLTHROUGH*/
-	case MVALUE_FORCE24:
+	case NUMBER_FORCES_24:
 		fprintf(fd, "+3\t= ");
 		break;
 	default:
 		fprintf(fd, "\t= ");
 	}
-	if (symbol->result.flags & MVALUE_DEFINED) {
-		if (symbol->result.flags & MVALUE_IS_FP)
-			fprintf(fd, "%.30f", symbol->result.val.fpval);	//FIXME %g
-		else
-			fprintf(fd, "$%x", (unsigned) symbol->result.val.intval);
-	} else {
-		fprintf(fd, " ?");
-	}
-	if (symbol->result.flags & MVALUE_UNSURE)
-		fprintf(fd, "\t; ?");
-	if (symbol->usage == 0)
+	if (symbol->object.u.number.ntype == NUMTYPE_UNDEFINED)
+		fprintf(fd, " ?");	// TODO - maybe write "UNDEFINED" instead? then the file could at least be parsed without errors
+	else if (symbol->object.u.number.ntype == NUMTYPE_INT)
+		fprintf(fd, "$%x", (unsigned) symbol->object.u.number.val.intval);
+	else if (symbol->object.u.number.ntype == NUMTYPE_FLOAT)
+		fprintf(fd, "%.30f", symbol->object.u.number.val.fpval);	//FIXME %g
+	else
+		Bug_found("IllegalNumberType4", symbol->object.u.number.ntype);
+	if (symbol->object.u.number.flags & NUMBER_EVER_UNDEFINED)
+		fprintf(fd, "\t; ?");	// TODO - write "forward" instead?
+	if (!symbol->has_been_read)
 		fprintf(fd, "\t; unused");
 	fprintf(fd, "\n");
 }
@@ -69,160 +75,141 @@ static void dump_vice_address(struct rwnode *node, FILE *fd)
 	struct symbol	*symbol	= node->body;
 
 	// dump address symbols even if they are not used
-	if ((symbol->result.flags & MVALUE_DEFINED)
-	&& !(symbol->result.flags & MVALUE_IS_FP)
-	&& (symbol->result.addr_refs == 1))
-		fprintf(fd, "al C:%04x .%s\n", (unsigned) symbol->result.val.intval, node->id_string);
+	if ((symbol->object.type == &type_number)
+	&& (symbol->object.u.number.ntype == NUMTYPE_INT)
+	&& (symbol->object.u.number.addr_refs == 1))
+		fprintf(fd, "al C:%04x .%s\n", (unsigned) symbol->object.u.number.val.intval, node->id_string);
 }
 static void dump_vice_usednonaddress(struct rwnode *node, FILE *fd)
 {
 	struct symbol	*symbol	= node->body;
 
 	// dump non-addresses that are used
-	if (symbol->usage
-	&& (symbol->result.flags & MVALUE_DEFINED)
-	&& !(symbol->result.flags & MVALUE_IS_FP)
-	&& (symbol->result.addr_refs != 1))
-		fprintf(fd, "al C:%04x .%s\n", (unsigned) symbol->result.val.intval, node->id_string);
+	if (symbol->has_been_read
+	&& (symbol->object.type == &type_number)
+	&& (symbol->object.u.number.ntype == NUMTYPE_INT)
+	&& (symbol->object.u.number.addr_refs != 1))
+		fprintf(fd, "al C:%04x .%s\n", (unsigned) symbol->object.u.number.val.intval, node->id_string);
 }
 static void dump_vice_unusednonaddress(struct rwnode *node, FILE *fd)
 {
 	struct symbol	*symbol	= node->body;
 
 	// dump non-addresses that are unused
-	if (!symbol->usage
-	&& (symbol->result.flags & MVALUE_DEFINED)
-	&& !(symbol->result.flags & MVALUE_IS_FP)
-	&& (symbol->result.addr_refs != 1))
-		fprintf(fd, "al C:%04x .%s\n", (unsigned) symbol->result.val.intval, node->id_string);
+	if (!symbol->has_been_read
+	&& (symbol->object.type == &type_number)
+	&& (symbol->object.u.number.ntype == NUMTYPE_INT)
+	&& (symbol->object.u.number.addr_refs != 1))
+		fprintf(fd, "al C:%04x .%s\n", (unsigned) symbol->object.u.number.val.intval, node->id_string);
 }
 
 
-// search for symbol. create if nonexistant. if created, give it flags "flags".
+// search for symbol. if it does not exist, create with NULL object (CAUTION!).
 // the symbol name must be held in GlobalDynaBuf.
-struct symbol *symbol_find(scope_t scope, int flags)
+struct symbol *symbol_find(scope_t scope)
 {
 	struct rwnode	*node;
 	struct symbol	*symbol;
-	int		node_created,
-			force_bits	= flags & MVALUE_FORCEBITS;
+	boolean		node_created;
 
 	node_created = Tree_hard_scan(&node, symbols_forest, scope, TRUE);
 	// if node has just been created, create symbol as well
 	if (node_created) {
 		// create new symbol structure
 		symbol = safe_malloc(sizeof(*symbol));
-		// finish empty symbol item
-		symbol->result.flags = flags;
-		symbol->result.addr_refs = 0;
-		if (flags & MVALUE_IS_FP)
-			symbol->result.val.fpval = 0;
-		else
-			symbol->result.val.intval = 0;
-		symbol->usage = 0;	// usage count
-		symbol->pass = pass_count;
 		node->body = symbol;
+		// finish empty symbol item
+		symbol->object.type = NULL;	// no object yet (CAUTION!)
+		symbol->pass = pass.number;
+		symbol->has_been_read = FALSE;
+		symbol->has_been_reported = FALSE;
+		symbol->pseudopc = NULL;
 	} else {
 		symbol = node->body;
 	}
-	// make sure the force bits don't clash
-	if ((node_created == FALSE) && force_bits)
-		if ((symbol->result.flags & MVALUE_FORCEBITS) != force_bits)
-			Throw_error("Too late for postfix.");
-	return symbol;
+	return symbol;	// now symbol->object.type can be tested to see if this was freshly created.
+	// CAUTION: this only works if caller always sets a type pointer after checking! if NULL is kept, the struct still looks new later on...
 }
 
 
-// assign value to symbol. the function acts upon the symbol's flag bits and
+// assign object to symbol. the function acts upon the symbol's flag bits and
 // produces an error if needed.
-void symbol_set_value(struct symbol *symbol, struct result *new_value, int change_allowed)
+// using "power" bits, caller can state which changes are ok.
+// called by:
+// implicit label definitions (including anons, backward anons have POWER_CHANGE_VALUE)
+// explicit symbol assignments
+// explicit symbol assignments via "!set" (has all powers)
+// loop counter var init via "!for" (has POWER_CHANGE_VALUE and POWER_CHANGE_NUMTYPE)
+//	CAUTION: actual incrementing of counter is then done directly without calls here!
+void symbol_set_object(struct symbol *symbol, struct object *new_value, bits powers)
 {
-	int	oldflags	= symbol->result.flags;
+	// if symbol has no object assigned to it yet, fine:
+	if (symbol->object.type == NULL) {
+		symbol->object = *new_value;	// copy whole struct including type
+		// as long as the symbol has not been read, the force bits can
+		// be changed, so the caller still has a chance to do that.
+		return;
+	}
 
-	// value stuff
-	if ((oldflags & MVALUE_DEFINED) && (change_allowed == FALSE)) {
-		// symbol is already defined, so compare new and old values
-		// if different type OR same type but different value, complain
-		if (((oldflags ^ new_value->flags) & MVALUE_IS_FP)
-		|| ((oldflags & MVALUE_IS_FP) ? (symbol->result.val.fpval != new_value->val.fpval) : (symbol->result.val.intval != new_value->val.intval)))
-			Throw_error("Symbol already defined.");
-	} else {
-		// symbol is not defined yet OR redefinitions are allowed
-		symbol->result = *new_value;
+	// now we know symbol already has a type
+
+	// compare types
+	// if too different, needs power (or complains)
+	if (symbol->object.type != new_value->type) {
+		if (!(powers & POWER_CHANGE_OBJTYPE))
+			Throw_error(exception_symbol_defined);
+		// CAUTION: if above line throws error, we still go ahead and change type!
+		// this is to keep "!for" working, where the counter var is accessed.
+		symbol->object = *new_value;	// copy whole struct including type
+		// clear flag so caller can adjust force bits:
+		symbol->has_been_read = FALSE;	// it's basically a new symbol now
+		return;
 	}
-	// flags stuff
-	// Ensure that "unsure" symbols without "isByte" state don't get that
-	if ((oldflags & (MVALUE_UNSURE | MVALUE_ISBYTE)) == MVALUE_UNSURE)
-		new_value->flags &= ~MVALUE_ISBYTE;
-	if (change_allowed) {
-		oldflags = (oldflags & MVALUE_UNSURE) | new_value->flags;
-	} else {
-		if ((oldflags & MVALUE_FORCEBITS) == 0)
-			if ((oldflags & (MVALUE_UNSURE | MVALUE_DEFINED)) == 0)
-				oldflags |= new_value->flags & MVALUE_FORCEBITS;
-		oldflags |= new_value->flags & ~MVALUE_FORCEBITS;
-	}
-	symbol->result.flags = oldflags;
+
+	// now we know symbol and new value have compatible types, so call handler:
+	symbol->object.type->assign(&symbol->object, new_value, !!(powers & POWER_CHANGE_VALUE));
 }
 
 
-// parse label definition (can be either global or local).
-// name must be held in GlobalDynaBuf.
-void symbol_set_label(scope_t scope, int stat_flags, int force_bit, int change_allowed)
+// set force bit of symbol. trying to change to a different one will raise error.
+void symbol_set_force_bit(struct symbol *symbol, bits force_bit)
 {
-	struct result	pc,
-			result;
-	struct symbol	*symbol;
+	if (!force_bit)
+		Bug_found("ForceBitZero", 0);
+	if (symbol->object.type == NULL)
+		Bug_found("NullTypeObject", 0);
 
-	symbol = symbol_find(scope, force_bit);
-	// label definition
-	if ((stat_flags & SF_FOUND_BLANK) && warn_on_indented_labels)
-		Throw_first_pass_warning("Label name not in leftmost column.");
-	vcpu_read_pc(&pc);
-	result.flags = pc.flags & MVALUE_DEFINED;
-	result.val.intval = pc.val.intval;
-	result.addr_refs = pc.addr_refs;
-	symbol_set_value(symbol, &result, change_allowed);
-}
-
-
-// parse symbol definition (can be either global or local, may turn out to be a label).
-// name must be held in GlobalDynaBuf.
-void symbol_parse_definition(scope_t scope, int stat_flags)
-{
-	struct result	result;
-	struct symbol	*symbol;
-	int		force_bit	= Input_get_force_bit();	// skips spaces after
-	// FIXME - force bit is allowed for label definitions?!
-
-	if (GotByte == '=') {
-		// explicit symbol definition (symbol = <something>)
-		symbol = symbol_find(scope, force_bit);
-		// symbol = parsed value
-		GetByte();	// skip '='
-		ALU_any_result(&result);
-		// if wanted, mark as address reference
-		if (typesystem_says_address())
-			result.addr_refs = 1;
-		symbol_set_value(symbol, &result, FALSE);
-		Input_ensure_EOS();
-	} else {
-		symbol_set_label(scope, stat_flags, force_bit, FALSE);
+	if (symbol->object.type != &type_number) {
+		Throw_error("Force bits can only be given to numbers.");
+		return;
 	}
+
+	// if change is ok, change
+	if (!symbol->has_been_read) {
+		symbol->object.u.number.flags &= ~NUMBER_FORCEBITS;
+		symbol->object.u.number.flags |= force_bit;
+		return;	// and be done with it
+	}
+
+	// it's too late to change, so check if the wanted bit is actually different
+	if ((symbol->object.u.number.flags & NUMBER_FORCEBITS) != force_bit)
+		Throw_error("Too late for postfix.");
 }
 
 
-// set global symbol to value, no questions asked (for "-D" switch)
+// set global symbol to integer value, no questions asked (for "-D" switch)
 // Name must be held in GlobalDynaBuf.
 void symbol_define(intval_t value)
 {
-	struct result	result;
+	struct object	result;
 	struct symbol	*symbol;
 
-	result.flags = MVALUE_GIVEN;
-	result.val.intval = value;
-	symbol = symbol_find(SCOPE_GLOBAL, 0);
-	symbol_set_value(symbol, &result, TRUE);
+	result.type = &type_number;
+	result.u.number.ntype = NUMTYPE_INT;
+	result.u.number.flags = 0;
+	result.u.number.val.intval = value;
+	symbol = symbol_find(SCOPE_GLOBAL);
+	symbol->object = result;
 }
 
 
@@ -245,6 +232,7 @@ void symbols_vicelabels(FILE *fd)
 	fputc('\n', fd);
 	// dump address symbols
 	Tree_dump_forest(symbols_forest, SCOPE_GLOBAL, dump_vice_address, fd);
+	// TODO - add trace points and watch points with load/store/exec args!
 }
 
 
@@ -252,20 +240,37 @@ void symbols_vicelabels(FILE *fd)
 // references the *next* anonymous forward label definition. The tricky bit is,
 // each name length would need its own counter. But hey, ACME's real quick in
 // finding symbols, so I'll just abuse the symbol system to store those counters.
-void symbol_fix_forward_anon_name(int increment)
+// example:
+//	forward anon name is "+++"
+//	we look up that symbol's value in the current local scope -> $12
+//	we attach hex digits to name -> "+++21"
+//	that's the name of the symbol that _actually_ contains the address
+// caller sets "increment" to TRUE for writing, FALSE for reading
+void symbol_fix_forward_anon_name(boolean increment)
 {
 	struct symbol	*counter_symbol;
 	unsigned long	number;
 
 	// terminate name, find "counter" symbol and read value
 	DynaBuf_append(GlobalDynaBuf, '\0');
-	counter_symbol = symbol_find(section_now->scope, 0);
-	// make sure it gets reset to zero in each new pass
-	if (counter_symbol->pass != pass_count) {
-		counter_symbol->pass = pass_count;
-		counter_symbol->result.val.intval = 0;
+	counter_symbol = symbol_find(section_now->local_scope);
+	if (counter_symbol->object.type == NULL) {
+		// finish freshly created symbol item
+		counter_symbol->object.type = &type_number;
+		counter_symbol->object.u.number.ntype = NUMTYPE_INT;
+		counter_symbol->object.u.number.flags = 0;
+		counter_symbol->object.u.number.addr_refs = 0;
+		counter_symbol->object.u.number.val.intval = 0;
+	} else if (counter_symbol->object.type != &type_number) {
+		// sanity check: it must be a number!
+		Bug_found("ForwardAnonCounterNotInt", 0);
 	}
-	number = (unsigned long) counter_symbol->result.val.intval;
+	// make sure it gets reset to zero in each new pass
+	if (counter_symbol->pass != pass.number) {
+		counter_symbol->pass = pass.number;
+		counter_symbol->object.u.number.val.intval = 0;
+	}
+	number = (unsigned long) counter_symbol->object.u.number.val.intval;
 	// now append to the name to make it unique
 	GlobalDynaBuf->size--;	// forget terminator, we want to append
 	do {
@@ -274,5 +279,5 @@ void symbol_fix_forward_anon_name(int increment)
 	} while (number);
 	DynaBuf_append(GlobalDynaBuf, '\0');
 	if (increment)
-		counter_symbol->result.val.intval++;
+		counter_symbol->object.u.number.val.intval++;
 }

src/symbol.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // symbol stuff
@@ -12,16 +12,16 @@
 
 
 struct symbol {
-	struct result	result;	// expression flags and value
-	int		usage;	// usage count
+	struct object	object;	// number/list/string
 	int		pass;	// pass of creation (for anon counters)
-	// add flag to indicate "has already been reported as undefined"
+	boolean		has_been_read;	// to find out if actually used
+	boolean		has_been_reported;	// indicates "has been reported as undefined"
+	struct pseudopc	*pseudopc;	// NULL when defined outside of !pseudopc block
 	// add file ref + line num of last definition
 };
 
 
 // Constants
-// TODO: add cheap locals (so there's SCOPE_GLOBAL, scope_zone and scope_cheap)
 #define SCOPE_GLOBAL	0	// number of "global zone"
 
 
@@ -29,17 +29,18 @@ struct symbol {
 extern struct rwnode	*symbols_forest[];	// trees (because of 8-bit hash)
 
 
-// function acts upon the symbol's flag bits and produces an error if needed.
-extern void symbol_set_value(struct symbol *symbol, struct result *new_value, int change_allowed);
-// parse label definition (can be either global or local).
-// name must be held in GlobalDynaBuf.
-extern void symbol_set_label(scope_t scope, int stat_flags, int force_bit, int change_allowed);
-// parse symbol definition (can be either global or local, may turn out to be a label).
-// name must be held in GlobalDynaBuf.
-extern void symbol_parse_definition(scope_t scope, int stat_flags);
-// search for symbol. create if nonexistant. if created, assign flags.
-// name must be held in GlobalDynaBuf.
-extern struct symbol *symbol_find(scope_t, int flags);
+// search for symbol. if it does not exist, create with NULL type object (CAUTION!).
+// the symbol name must be held in GlobalDynaBuf.
+extern struct symbol *symbol_find(scope_t scope);
+// assign object to symbol. function acts upon the symbol's flag bits and
+// produces an error if needed.
+// using "power" bits, caller can state which changes are ok.
+#define POWER_NONE		0
+#define POWER_CHANGE_VALUE	(1u << 0)	// e.g. change 3 to 5 or 2.71
+#define POWER_CHANGE_OBJTYPE	(1u << 1)	// e.g. change 3 to "somestring"
+extern void symbol_set_object(struct symbol *symbol, struct object *new_obj, bits powers);
+// set force bit of symbol. trying to change to a different one will raise error.
+extern void symbol_set_force_bit(struct symbol *symbol, bits force_bit);
 // set global symbol to value, no questions asked (for "-D" switch)
 // name must be held in GlobalDynaBuf.
 extern void symbol_define(intval_t value);
@@ -49,7 +50,7 @@ extern void symbols_list(FILE *fd);
 extern void symbols_vicelabels(FILE *fd);
 // fix name of anonymous forward label (held in GlobalDynaBuf, NOT TERMINATED!)
 // so it references the *next* anonymous forward label definition.
-extern void symbol_fix_forward_anon_name(int increment);
+extern void symbol_fix_forward_anon_name(boolean increment);
 
 
 #endif

src/tree.c

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // tree stuff
@@ -28,14 +28,15 @@ hash_t make_hash(struct ronode *node) {
 }
 
 // Link a predefined data set to a tree
-void add_node_to_tree(struct ronode **tree, struct ronode *node_to_add)
+static void add_node_to_tree(struct ronode **tree, struct ronode *node_to_add)
 {
 	hash_t	hash;
 
 	// compute hash value
 	hash = make_hash(node_to_add);
+	// search for NULL pointer to replace
 	while (*tree) {
-		// compare HashValue
+		// decide which way to go
 		if (hash > (*tree)->hash_value)
 			tree = &((*tree)->greater_than);
 		else
@@ -48,10 +49,11 @@ void add_node_to_tree(struct ronode **tree, struct ronode *node_to_add)
 // fields.
 }
 
-// Add predefined tree items to given tree. The PREDEF* macros set HashValue
-// to 1 in all entries but the last. The last entry contains 0.
-void Tree_add_table(struct ronode **tree, struct ronode *table_to_add)
+// Add predefined tree items to given tree. The PREDEF* macros set the hash
+// to 1 in all entries but the last, and to 0 in the last entry.
+static void tree_from_list(struct ronode **tree, struct ronode *table_to_add)
 {
+	//printf("Building tree from list.\n");
 	// Caution when trying to optimise this. :)
 	while (table_to_add->hash_value)
 		add_node_to_tree(tree, table_to_add++);
@@ -72,6 +74,15 @@ int Tree_easy_scan(struct ronode *tree, void **node_body, struct dynabuf *dyna_b
 			b2;
 	hash_t		hash;
 
+	// check if tree is actually ready to use. if not, build it from list.
+	// (list's first item does not contain real data, so "greater_than" is
+	// used to hold pointer to tree root)
+	if (tree->greater_than == NULL)
+		tree_from_list(&tree->greater_than, tree + 1);	// real data starts at next list item
+	tree = tree->greater_than;	// go from list head to tree root
+	// ok, we're done with this setup stuff.
+	// from now on, "greater_than" really means "greater_than"!
+
 	wanted.id_string = dyna_buf->buffer;
 	hash = make_hash(&wanted);
 	while (tree) {
@@ -110,7 +121,7 @@ int Tree_easy_scan(struct ronode *tree, void **node_body, struct dynabuf *dyna_b
 // a new tree item, link to tree, fill with data and store its pointer. If the
 // "create" flag is zero, store NULL as result.
 // Returns whether item was created.
-int Tree_hard_scan(struct rwnode **result, struct rwnode **forest, int id_number, int create)
+int Tree_hard_scan(struct rwnode **result, struct rwnode **forest, int id_number, boolean create)
 {
 	struct ronode	wanted;	// temporary storage
 	struct rwnode	**current_node;
@@ -158,7 +169,7 @@ int Tree_hard_scan(struct rwnode **result, struct rwnode **forest, int id_number
 		current_node = &((*current_node)->less_than_or_equal);
 	}
 	// node wasn't found. Check whether to create it
-	if (create == FALSE) {
+	if (!create) {
 		*result = NULL;	// indicate failure
 		return FALSE;	// return FALSE because node was not created
 	}
@@ -178,7 +189,7 @@ int Tree_hard_scan(struct rwnode **result, struct rwnode **forest, int id_number
 
 // Call given function for each object of matching type in the given tree.
 // Calls itself recursively.
-void dump_tree(struct rwnode *node, int id_number, void (*fn)(struct rwnode *, FILE *), FILE *env)
+static void dump_tree(struct rwnode *node, int id_number, void (*fn)(struct rwnode *, FILE *), FILE *env)
 {
 
 	if (node->id_number == id_number)

src/tree.h

@@ -1,5 +1,5 @@
 // ACME - a crossassembler for producing 6502/65c02/65816/65ce02 code.
-// Copyright (C) 1998-2016 Marco Baye
+// Copyright (C) 1998-2020 Marco Baye
 // Have a look at "acme.c" for further info
 //
 // tree stuff
@@ -8,12 +8,13 @@
 
 
 #include <stdio.h>	// for FILE
+#include "config.h"
 
 
 // macros for pre-defining tree node tables
+#define PREDEF_START		{NULL, NULL, 0, NULL, NULL}	// this is used to determine if list has been made into tree yet
 #define PREDEFNODE(s, v)	{NULL, NULL, 1, s, (void *) (v)}
-#define PREDEFLAST(s, v)	{NULL, NULL, 0, s, (void *) (v)}
-
+#define PREDEF_END(s, v)	{NULL, NULL, 0, s, (void *) (v)}
 
 // type definitions
 
@@ -35,14 +36,12 @@ struct rwnode {
 	hash_t		hash_value;
 	char		*id_string;	// name, zero-terminated
 	void		*body;		// macro/symbol body
-	unsigned int	id_number;	// scope number
+	int		id_number;	// scope number
 };
 
 
 // prototypes
 
-// Add predefined tree items to given tree.
-extern void Tree_add_table(struct ronode **tree, struct ronode *table_to_add);
 // Search for a given ID string in a given tree. Store "body" component in
 // node_body and return TRUE. Return FALSE if no matching item found.
 struct dynabuf;
@@ -50,10 +49,10 @@ extern int Tree_easy_scan(struct ronode *tree, void **node_body, struct dynabuf
 // Search for a "RAM tree" item. Save pointer to found tree item in given
 // location. If no matching item is found, check the "create" flag: If set,
 // create new tree item, link to tree, fill with data and store its pointer.
-// If "create" is zero, store NULL. Returns whether item was created.
-extern int Tree_hard_scan(struct rwnode **result, struct rwnode **forest, int id_number, int create);
+// If "create" is FALSE, store NULL. Returns whether item was created.
+extern int Tree_hard_scan(struct rwnode **result, struct rwnode **forest, int id_number, boolean create);
 // Calls given function for each node of each tree of given forest.
-extern void Tree_dump_forest(struct rwnode **, int, void (*)(struct rwnode *, FILE *), FILE *);
+extern void Tree_dump_forest(struct rwnode **, int id_number, void (*)(struct rwnode *, FILE *), FILE *);
 
 
 #endif