Split words with mads

[ivop]

Hi,

 

Is there a way to easily "split" words into two separate tables? Here's what I mean:

 

table
	.word $0123, $4567, $89ab, $cdef

; becomes:

tablelo
	.byte $23, $67, $ab, $ef
tablehi
	.byte $01, $45, $89, $cd

With four words, it is not that much of a problem to do it by hand, but with eight or more, it becomes a burden. Especially if you are bound to change and tweak those values to match what you are trying to achieve

 

 

Edited February 25, 2021 by ivop

[TGB1718]

Not sure if this would do, it's MAC/65, but I assume MADS will do something similar

 

A1       =   $0123
A2       =   $4567
A3       =   $89AB
A4       =   $CDEF
TABLELO  .BYTE  <A1, <A2, <A3, <A4
TABLEHI  .BYTE  >A1, >A2, >A3, >A4

 

assembles into:-

 

    =0123       10 A1    =   $0123
    =4567       20 A2    =   $4567
    =89AB       30 A3    =   $89AB
    =CDEF       40 A4    =   $CDEF
0000 2367ABEF   50 TABLELO .BYTE  <A1, <A2, <A3, <A4
0004 014589CD   60 TABLEHI .BYTE  >A1, >A2, >A3, >A4

[tebe]

dta l(a,b,c,d)

dta h(a,b,c,d)

[ivop]

I had totally forgotten about l() and h(). That's a 50% solution IMHO

 

Would you consider implementing something like this:

table
	.startsplitword
		.splitword $0123, $4567, $89ab, $cdef
		.splitword $8976, $afe0, $dead, $beef
		; etc...
	.endsplitword
tableend

tablelo = table
tablehi = table + (tableend - table) / 2

Syntax is debatable of course

 

 

[ivop]

    opt h-
    .macro splitwords
  
;    :%%0 dta #              ; # works, counts from 0 to number of parameters-1

    :%%0 dta l(:#+1)        ; # in an expression does not work
    :%%0 dta h(:.R+1)       ; neither does .R

    .endm

    splitwords $dead, $beef, $0123, $4567, $89ab, $cdef

 

Edit: this happens to be my 2048th post

Edited February 26, 2021 by ivop

[+Stephen]

17 minutes ago, ivop said:

Edit: this happens to be my 2048th post

Way to go Robotron!

[Wrathchild]

I could only get as far as this:

 

mads 2.1.1 build 17 (18 Nov 20)
Source: test.asm
     1                     org $2000, $600
     2                     
     3                     .macro splitwords
     4                 
     5                 :1_LO: 
     6                 :(:0-1)    .byte <(%%(# + 2))
     7                 :1_HI: 
     8                     .rept :0-1
     9                     .byte >(%%(# + 2))
    10                     .endr
    11                 
    12                     .endm
    13                     
    14                     org *
    15                     
    16                 .proc main
    17                 endless
    18 FFFF> 2000-200E> 4C 00 +     jmp endless
    19 2003                splitwords table $dead, $beef, $0123, $4567, $89ab, $cdef
Macro: SPLITWORDS [Source: test.asm]
     2 2003            TABLE_LO: 
     3 2003 02 03 04 05 06 07    :(7-1)    .byte <(%%(# + 2))
     4 2009            TABLE_HI: 
     6 2009 00                .BYTE >(%%(# + 2))
     6 200A 00                .BYTE >(%%(# + 2))
     6 200B 00                .BYTE >(%%(# + 2))
     6 200C 00                .BYTE >(%%(# + 2))
     6 200D 00                .BYTE >(%%(# + 2))
     6 200E 00                .BYTE >(%%(# + 2))
Source: test.asm
    20                 .endp
    21
    22 02E0-02E1> 00 20            run main

So it seems that the %% can't handle resolution of the brackets to give it the index.

 

Using ':' instead makes this worse:

    splitwords table $dead, $beef, $0123, $4567, $89ab, $cdef
        .BYTE >(:(# + 2))
test.asm (9) ERROR: Extra characters on line

The other thing I could think of that would be useful would be if the repeat could be passed all the arguments from the macro as ':*' with an option to start from a given position, e.g. something like this to start from the second parameter:

    .macro splitwords

:1_LO: 
    .rept :0-1 :*+2
    .byte <(:.R)
    .endr
:1_HI: 
    .rept :0-1 :*+2
    .byte >(:.R)
    .endr

    .endm

But this still needs ":.R" or ":#" or "%%#" to be understood... one for @tebe

Edited February 26, 2021 by Wrathchild
this->thing

[dmsc]

Hi!

1 hour ago, ivop said:

Is there a way to easily "split" words into two separate tables?

 

I don't know much about MADS syntax, but this is easy in CA65:

.define MyTable TableItem0, TableItem1, TableItem2, TableItem3

        TableLookupLo:   .lobytes MyTable
        TableLookupHi:   .hibytes MyTable

Perhaps there is something like the "define" keyword in MADS?

 

Have Fun!

[Rybags]

I like the l( h( method - it's recognisable for what you're trying to achieve.  And you just copy/paste and change the h to l in the copy - remembering we'd be using a modern PC based editor in most cases.