From: dxforth@gmail.com   
      
   On 30/10/2024 7:37 pm, Hans Bezemer wrote:   
   > On 30-10-2024 01:31, dxf wrote:   
   >> On 30/10/2024 12:45 am, Hans Bezemer wrote:   
   >>> On 29-10-2024 10:25, dxf wrote:   
   >>>> On 29/10/2024 4:07 am, Hans Bezemer wrote:   
   >>>>> ...   
   >>>>> I have put the complication elsewhere. If we assume we're working in   
   decimal, you don't even need >NUMBER:   
   >>>>>   
   >>>>> char 0 negate +constant 0-   
   >>>>>   
   >>>>> : /int    ( a1 n1 -- a2 n2 n3)   
   >>>>>     0 >r 1 >r 1- chars over +   
   >>>>>     begin   
   >>>>>      over 1- over <   
   >>>>>     while   
   >>>>>       dup c@ is-digit   
   >>>>>     while   
   >>>>>       dup c@ 0- r> tuck * r> + >r 10 * >r 1-   
   >>>>>     repeat over - rdrop r> -rot   
   >>>>> ;   
   >>>>> ...   
   >>>>   
   >>>> So that's how to convert a numeric string from the other end.   
   >>>> Nice!  Easier than I imagined it would be.   
   >>>   
   >>> TORS is the multiplier, 2ORS is the accumulator. Normally, the multiplier   
   is multiplied after each run with BASE @ - but that makes little sense when   
   IS-DIGIT isn't properly adjusted. Would be a nice exercise, though.   
   >>   
   >> Something like this...   
   >>   
   >> \  \CHAR ( a u -- a u-1 c )   
   >> \  >DIGIT ( c base -- u -1 | c 0 )   
   >>   
   >> : /INT ( a1 n1 -- a2 n2 u )   
   >>    0 >r  1  begin  >r  dup while   
   >>      \char  base @ >digit  while   
   >>      r> tuck  *  r> + >r  base @ *   
   >>    repeat  drop ( -1 /string)  then  rdrop  r> ;   
   >>   
   >> OTOH the necessity to convert R-L is probably rare.   
   >   
   > Yes, R-L number parsing is very rare. This time, it felt like the easiest   
   solution, though. Equally when you're parsing a non-decimal date. That's why I   
   left it out here. But it doesn't hurt to contemplate such things. ;-)   
   >   
   > I got the words you describe as "STRING/C" (rings a bell?) and "DIGIT?".   
   Seem like natural candidates to tackle this problem.   
      
   Char extraction occurs often that it made sense to make them bona-fide words.   
   Having them in the kernel meant I could code them in asm for extra speed.   
      
   It raises the question of whether it's better to test then extract - or extract   
   then test.  The latter is simpler but it can mean one has to undo the   
   operation.   
   That was the purpose of ( -1 /string) ... which BTW was incorrect; it should   
   have   
   been ( 1+) .   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)