Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
42 Character Printing
#1
The following two subroutines will support 42 character printing. Colour is allowed, but be aware that the characters are NOT guaranteed to fit into an attribute square, and as such, may cause colour clash.

Set colour with the permanent colour statements (ink/paper not in a print line).

The 42 character printing routine maintains its own X,Y coordinates for printing. Printing on all 24 lines is supported.

X values can range from 0 to 41 (0 <= X <= 42) and Y values can range from 0 to 24 (0 <= Y <= 24). Use the printAt42(y,x) call to change the current position of the printing, otherwise successive prints carry on where the last one left off (with runover to the next line, correctly after the 42nd character).

Size is kept to a minimum, because the routine uses the ROM character set and cuts out lines from it in order to reduce to 6 pixels wide; with a couple of exceptions that are improved by being thus defined here. A redesigned character set is possible, but only for about 31 characters, owing to space limitations in the data design.

Characters ARE vertically aligned to the 42 character grid, however. This is not proportional printing. Therefore this routine can be used safely for tables and other such gridded arrangements.

Original routine this is based on was written by Paul Wardle.

printat42 routine
Code:
SUB printat42 (y as uByte, x as uByte)
    POKE @printAt42Coords,x
    POKE @printAt42Coords+1,y
END sub

The main print42 routine, with assembly and documentation
Code:
SUB print42 (characters$ as string)


asm

        POP BC ; Grab this
        POP DE ; Grab the return address                    
        POP HL ; grab our string address
        PUSH DE ; put the return address back where we found it
      
        
        LD C,(HL)
        INC HL
        LD B,(HL)       ; all told, LD BC with the length of the string.

        INC HL          ;Puts HL to the first real character in the string.

        LD A,C
        OR B
        RET Z           ; Is the length of the string 0? If so, quit.

examineChar:
        LD A,(HL)       ; Grab the character at our pointer position
        CP 128          ;Too high to print?
        JR NC, nextChar ; Then we go to the next

        CP 22           ; Is this an AT?
        JR NZ, isNewline ; If not jump over the AT routine to isNewline

isAt:
        EX DE,HL        ; Get DE to hold HL for a moment
        AND A           ; Plays with the flags. One of the things it does is reset Carry.
        LD HL,00002
        SBC HL,BC       ; Subtract length of string from HL.
        EX DE,HL        ; Get HL back from DE
        RET NC          ; If the result WASN'T negative, return. (We need AT to have parameters to make sense)

        INC HL          ; Onto our Y co-ordinate
        LD D,(HL)       ; Put it in D
        DEC BC          ; and move our string remaining counter down one                
        INC HL          ; Onto our X co-ordinate
        LD E,(HL)       ; Put the next one in E
        DEC BC          ; and move our string remaining counter down one
        CALL nxtchar      ; Call routine to shuffle right a char
        JR newline      ; Hop over to

isNewline:
        CP 13           ; Is this character a newline?
        JR NZ,checkvalid     ; If not, jump forward
        

newline:
        LD DE,(63536)
        CALL nxtline       ; move to next line

        LD (63536),DE     ; and go on to next character
        JR nextChar
        
checkvalid:
        CP 31           ; Is character <31?
        JR C, nextChar  ; If not go to next character

prn:    PUSH HL          ; Save our position
        PUSH BC          ; Save our countdown of chars left
        CALL printachar       ; Go print a character
        POP BC           ; Recover our count
        POP HL           ; Recover our position                              
nextChar:
        INC HL           ; Move to the next position
        DEC BC           ; count off a character
        LD A,B
        OR C            ; Did we hit the end of our string? (BC=0?)
        JR NZ, examineChar    ; If not, we need to go look at the next character.
        RET               ; End the print routine        
basicVariableName:
         defb "z$"        ; The name of the variable we are looking at.


; This routine forms the new 6-bit wide characters and
;alters the colours to match the text. The y,x co-ordinates and eight
;bytes of workspace are located at the end of this chunk.
; it starts with the character ascii code in the accumulator

printachar:
        EXX
        PUSH HL ; Store H'L' where we can get it.
        EXX              

       ld c, a    ; Put a copy of the character in C
       ld h, 0    
       ld l, a    ; Put the Character in HL
      
       ld de, whichcolumn-32 ; the character is at least 32, so space = 0th entry.      
       add hl, de         ; HL -> table entry for char.
       ld a, (hl)         ; Load our column slice data from the table.
       cp 32             ; Is it less than 32?
       jr nc, calcChar   ; If so, go to the calculated character subroutine


; This is the special case 'we defined the character in the table' option    
       ld de, characters ; Point DE at our table
       ld l, a             ; Put our character number from our table lookup that's in HL in a
       call mult8         ; multiplies L by 8 and adds in DE [so HL points at our table entry]
       ld b, h            
       ld c, l             ; Copy our character data address into BC
       jr printdata         ; We have our data source, so we print it.

calcChar: ; this is the calculate from the ROM data option
            ; a holds the column kill data
       ld de, 15360         ; Character set-256. We could use CHARS here, maybe; but might not work with a redefiend character set.
       ld l, c             ; Get our character back from C
       call mult8         ; Multiply l by 8 and add to DE. (HL points at the ROM data for our character now)
      
       ld de, workspace  ; Point DE at our 8 byte workspace.
       push de             ; Save it
       exx                 ;
       ld c, a             ; Put our kill column in C'
       cpl                 ; Invert
       ld b, a             ; Put the inverse in B'
       exx                 ;
       ld b, 8             ; 8 bytes to a character loop counter

loop1:
       ld a, (hl)         ; Load a byte of character data
       inc hl             ; point at the next byte
       exx                 ;
       ld e, a             ; Put it in e'
       and c             ; keep the left column block we're using
       ld d, a             ; and put it in d'
       ld a, e             ; grab our original back
       rla                 ; shift it left (which pushes out our unwanted column)
       and b             ; keep just the right block
       or d                 ; mix with the left block
       exx                 ;
       ld (de), a         ; put it into our workspace
       inc de             ; next workspace byte
       djnz loop1         ; go round for our other bytes
    
       pop bc             ; Recover a pointer to our workspace.

printdata:
       call testcoords     ; check our position, and wrap around if necessary. [returns with d=y,e=x]
       inc e             ; Bump along to next co-ordinate
       ld (xycoords), de ; Store our coordinates for the next character
       dec e             ; Bump back to our current one
       ld a, e             ; get x
       sla a             ;  Shift Left Arithmetic - *2
       ld l, a             ; put x*2 into L
       sla a             ; make it x*4
       add a, l             ; (x*2)+(x*4)=6x
       ld l, a             ; put 6x into L [Since we're in a 6 pixel font, L now contains the # of first pixel we're interested in]
       srl a             ; divide by 2
       srl a             ; divide by another 2 (/4)
       srl a             ; divide by another 2 (/8)
       ld e, a             ; Put the result in e (Since the screen has 8 pixel bytes, pixel/8 = which char pos along our first pixel is in)
       ld a, l             ; Grab our pixel number again
       and 7             ; And do mod 8 [So now we have how many pixels into the character square we're starting at]
       push af             ; Save A
       ex af, af'          
       ld a, d             ; Put y Coord into A'
       sra a             ; Divide by 2
       sra a             ; Divide by another 2 (/4 total)
       sra a             ; Divide by another 2 (/8) [Gives us a 1/3 of screen number]
       add a, 88         ; Add in start of screen attributes high byte
       ld h, a             ; And put the result in H
       ld a, d             ; grab our Y co-ord again
       and 7             ; Mod 8 (why? *I thought to give a line in this 1/3 of screen, but we're in attrs here)
       rrca                 ;
       rrca                  
       rrca                 ; Bring the bottom 3 bits to the top - Multiply by 32(since there are 32 bytes across the screen), here, in other words. [Faster than 5 SLA instructions]
       add a, e             ; add in our x coordinate byte to give us a low screen byte
       ld l, a                ; Put the result in L. So now HL -> screen byte at the top of the character
      
       ld a, (23693)     ; ATTR P      Permanent current colours, etc (as set up by colour statements).
       ld e, a             ; Copy ATTR into e
       ld (hl), e         ; Drop ATTR value into screen
       inc hl             ; Go to next position along
       pop af             ; Pull how many pixels into this square we are
       cp 3                 ; It more than 2?
       jr c, hop1         ; No? It all fits in this square - jump changing the next attribute
    
       ld (hl), e         ; 63446 Must be yes - we're setting the attributes in the next square too.
hop1:
       dec hl             ; Back up to last position
       ld a, d             ; Y Coord into A'
       and 248             ; Turn it into 0,8 or 16. (y=0-23)
       add a, 64         ; Turn it into 64,72,80  [40,48,50 Hex] for high byte of screen pos
       ld h, a             ; Stick it in H
       push hl             ; Save it
       exx                 ; Swap registers
       pop hl             ; Put it into H'L'
       exx                 ; Swap Back
       ld a, 8              
hop4:
       push af             ; Save Accumulator
       ld a, (bc)         ; Grab a byte of workspace
       exx                 ; Swap registers
       push hl             ; Save h'l'
       ld c, 0             ; put 0 into c'
       ld de, 1023         ; Put 1023 into D'E'
       ex af, af'         ; Swap AF
       and a             ; Flags on A
       jr z, hop3         ; If a is zero jump forward

       ld b, a             ; A -> B
       ex af, af'         ; Swap to A'F'
hop2:; Slides a byte right to the right position in the block (and puts leftover bits in the left side of c)
       and a             ; Clear Carry Flag
       rra                 ; Rotate Right A
       rr c                 ; Rotate right C (Rotates a carry flag off A and into C)
       scf                 ; Set Carry Flag
       rr d                 ; Rotate Right D
       rr e                 ; Rotate Right E (D flows into E, with help from the carry bit)
       djnz hop2         ; Decrement B and loop back
      
       ex af, af'        
hop3:
       ex af, af'        
       ld b, a            
       ld a, (hl)        
       and d            
       or b                
       ld (hl), a         ; Write out our byte
       inc hl             ; Go one byte right
       ld a, (hl)         ; Bring it in
       and e            
       or c                 ; mix those leftover bits into the next block
       ld (hl), a         ; Write it out again
       pop hl            
       inc h                ; Next line
       exx                
       inc bc              ; Next workspace byte
       pop af            
       dec a             
       jr nz, hop4         ; And go back!
    
       exx                 ; Tidy up
       pop hl             ; Clear stack leftovers
       exx                 ; And...
       ret                 ; Go home.

mult8: ; Multiplies L by 8 -> HL and adds it to DE. Used for 8 byte table vectors.
        ld h, 0            
        add hl, hl        
        add hl, hl        
        add hl, hl          
        add hl, de        
        ret
                    
testcoords:
        ld de, (xycoords)    ; get our current screen co-ordinates (d=y,e=x - little endian)
nxtchar:
        ld a, e             ;
        cp 42             ; Are we >42?
        jr c, ycoord     ; if not, hop forward
nxtline:
       inc d             ; if so, so bump us to the next line down
       ld e, 0             ; and reset x to left edge
ycoord:
        ld a, d             ;
       cp 24             ; are we >24 lines?
       ret c             ; if no, exit subroutine
       ld d, 0             ; if yes, wrap around to top line again.
       ret                 ; exit subroutine
end asm
printAt42Coords:
asm      
xycoords:
        defb 0      ; x coordinate      
        defb 0      ; y coordinate

workspace:
        defb 0      
        defb 0        
        defb 0
        defb 0        
        defb 0         
        defb 0
        defb 0        
        defb 0        
    
; The data below identifies a column in the character to remove. It consists of 1's
; from the left edge. First zero bit is the column we're removing.
; If the leftmost bit is NOT 1, then the byte represents a redefined character position
; in the lookup table.
    
whichcolumn:            
    defb 254         ; SPACE
    defb 254         ; !
    defb 128         ; "
    defb 224         ; #
    defb 128         ; $
    defb 0           ; % (Redefined below)
    defb 1           ; &  (Redefined below)
    defb 128         ; '
    defb 128         ; (
    defb 128         ; )
    defb 128         ; *
    defb 128         ; +
    defb 128         ; ,
    defb 128         ; -
    defb 128         ; .
    defb 128         ; /
    defb 2           ; 0 (Redefined below)
    defb 128       ; 1
    defb 224       ; 2
    defb 224       ; 3
    defb 252       ; 4
    defb 224        ; 5
    defb 224        ; 6
    defb 192       ; 7
    defb 240       ; 8
    defb 240       ; 9
    defb 240       ; :
    defb 240       ; ;
    defb 192       ; <
    defb 240       ; =
    defb 192       ; >
    defb 192       ; ?
    defb 248       ; @
    defb 240       ; A
    defb 240       ; B
    defb 240       ; C
    defb 240       ; D
    defb 240       ; E
    defb 240       ; F
    defb 240       ; G
    defb 240       ; H
    defb 128       ; I
    defb 240       ; J
    defb 192       ; K
    defb 240       ; L
    defb 240       ; M
    defb 248       ; N
    defb 240       ; O
    defb 240       ; P
    defb 248       ; Q
    defb 240       ; R
    defb 240       ; S
    defb 3         ; T
    defb 240       ; U
    defb 240       ; V
    defb 240       ; W
    defb 240       ; X
    defb 4         ; Y
    defb 252       ; Z
    defb 224       ; [
    defb 252       ; \
    defb 240        ; ]    
    defb 252        ; ^
    defb 240        ; _
    defb 240        ; UK Pound (Currency) Symbol
    defb 255        ; a
    defb 128        ; b
    defb 255        ; c    
    defb 255        ; d    
    defb 255        ; e    
    defb 255        ; f    
    defb 255        ; g    
    defb 255        ; h    
    defb 255        ; i    
    defb 255        ; j    
    defb 255        ; k    
    defb 255        ; l    
    defb 255        ; m    
    defb 255        ; n    
    defb 255        ; o    
    defb 255        ; p    
    defb 255        ; q    
    defb 255        ; r    
    defb 255        ; s    
    defb 255        ; t    
    defb 255        ; u    
    defb 255        ; v    
    defb 255        ; w    
    defb 255        ; x    
    defb 255        ; y    
    defb 255        ; z    
    defb 128        ; {
    defb 128        ; |
    defb 255        ; }    
    defb 128        ; ~
    defb 5            ; (c)  end column data
    
    
    
characters:    
    defb 0           ; %            
    defb 0            
    defb 100        
    defb 104        
    defb 16
    defb 44
    defb 76
    defb 0
    
    defb 0              ; &
    defb 32
    defb 80              
    defb 32
    defb 84              
    defb 72            
    defb 52        
    defb 0        
    
    defb 0             ; digit 0
    defb 56
    defb 76         
    defb 84            
    defb 84            
    defb 100        
    defb 56          
    defb 0
    
    defb 0             ; Letter T
    defb 124        
    defb 16
    defb 16          
    defb 16
    defb 16          
    defb 16
    defb 0          
    
    defb 0             ; Letter Y
    defb 68          
    defb 68            
    defb 40
    defb 16         
    defb 16
    defb 16         
    defb 0            
    
    defb 0             ; (c) symbol
    defb 48            
    defb 72            
    defb 180        
    defb 164        
    defb 180                                    
    defb 72            
    defb 48            
    
end asm    
END SUB

A little program to test and demonstrate:
Code:
DIM n as uByte
CLS
PRINT "01234567890123456789012345678901"
FOR n=1 to 6
    printat42(n+1,0)
    INK n
    print42("012345678901234567890123456789012345678901")
NEXT n
Reply
#2
Thanks. This will be useful...
I have somewhere a source for proportional printing too.
------------------------------------------------------------
http://lcd-one.da.ru redirector is dead
Visit my http://members.inode.at/838331/index.html home page!
Reply
#3
As you already know, this routine was already included.
I've cleaned it up today (and little bug removed). It will be updated in the next release, along with the new Britlion's print64 routine.
We must discuss how to document them in the Documentation sub-forum: I think it's not a good idea to include source code in the wiki. It's better to put an usage example instead. Idea
Reply
#4
Hi all

Is it possible to add/modify characters to the routine? I know that some can be modified (like %, T, Y or 0) but I need them; and there are others (like {, }. [ or ]) that I'm not using, so I was thinking to replace them.
The new chars I need are the spanish vocals á é í ó ú and the ¿ character.

Thanks and regards
Reply
#5
Yes, but not quite as simple as it looks - the routine works on minimal data. Rather than having a 6 bit wide character set, it has a list of columns to remove from the spectrum character set, turning an 8 bit wide font into a 6 bit wide font. Have a look at the source code! Each character, with a few exceptions, is "stored" as a single byte.

That said, if you pull it from the library, and edit it to make your own, you CAN change characters - it allows them to be redefined for the cases where "remove left column, plus one other column" didn't work well.

%, &, 0, T, Y and copyright symbol are all defined in the "characters" section. You could add characters here - needs 8 bytes. You'll have to change the marker in the main character information to note that it needs a redefined character set as well, however.
Reply
#6
Thanks, I'll play a bit with the library source code with the help of your hints Smile

Cheers
Reply
#7
Sorry to bump this after so long, but it seemed like the most relevant thread!

How do you include a newline with Print42? I have looked through the source and I thought adding CHR(13) to a string would do it, but it doesn't. For example:

Code:
#include <print42.bas>
dim myString as string
cls

printat42(0,0)
myString = "Hello"+chr(13)+"World"
print42(myString)

printat42(4,0)
print42("Hello"+chr(13)+"World")

' neither way works...?
Reply
#8
I don't think the routine allows for that specifically. I'd have to look - it's years since I played with it. It did in its original Aowen streams form, I'm sure.

for now, I think printat is the key.

Oddly, I'm working on fonts stuff. I think FZX should replace all the text drivers for the most part, because it's just more flexible; but an interface hasn't been added for Boriels' ZX Basic. Since I wanted to use it, I'm progressing on hacking a version together that works. We'll see if I make it! Smile

(FZX basically allows a set of font data, and fonts can be quite a startling array of sizes, and set to be proportional or fixed. By pointing the routine at a separate font data, you can have more than one font trivially)

<!-- m --><a class="postlink" href="http://www.worldofspectrum.org/infoseekid.cgi?id=0028171">http://www.worldofspectrum.org/infoseek ... id=0028171</a><!-- m -->
Reply
#9
Thanks for the reply britlion - I'll check out FZX. A version for easy use with Boriel would be great and much appreciated, if possible. Smile
Reply
#10
Not quite ready for publication, but up and should be working:

Listed quietly here as I was doing work in progress stuff:
<!-- l --><a class="postlink-local" href="http://www.boriel.com/forum/bug-reports/variable-corruption-t911.html">bug-reports/variable-corruption-t911.html</a><!-- l -->
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)