Forums before death by AOL, social media and spammers... "We can't have nice things"
|    comp.lang.forth    |    Forth programmers eat a lot of Bratwurst    |    117,927 messages    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
|    Message 117,131 of 117,927    |
|    dxf to dxf    |
|    Re: (FG.) FG.R (was Re: Bring your Forth    |
|    08 Mar 25 12:57:50    |
      From: dxforth@gmail.com              On 7/03/2025 12:43 pm, dxf wrote:       > ...       > Turns out I needed a new function to duplicate the output. I must have       > run into the same issue before as a decade ago I defined FG.R etc which       > simulates Fortran's 'G' format output. The original had some quirks so       > I've taken the opportunity to update it.       > ...              A slightly improved version that avoids a calculation. The latter       was always superfluous but I couldn't see a way of removing it without       increasing code elsewhere ... until now. Also removed is the '1 MAX'       since 'zero significant digits' represents an ambiguous condition.              \ Purpose: derive a floating-point output function with       \ characteristics similar to Fortran's 'G' format. Useful       \ for displaying tables of formatted results.       \       \ Assumes the function:       \ (FS.) ( r n -- a u )       \ Convert r to a string a u in scientific notation to n       \ decimal places. Both '.' and 'E' must be present in the       \ returned string (NAN/INFs excepted).       \       \ Public domain (no warranty)              \ Misc tools       \ SCAN ( a u char -- a2 u2 ) common usage       : (NUMBER) ( a u -- ud a' u' ) 0 0 2swap >number ;       : /SIGN ( a u -- a' u' f ) \ skip leading sign if exists        dup if over c@ dup [char] + = swap [char] - =        dup >r or negate /string r> exit then 0 ;       : /NUMBER ( a u -- a' u' d|ud )        /sign >r (number) 2swap r> if dnegate then ;       : CSKIP 1 /string ;       : 2NIP 2swap 2drop ;       : S.R ( a u wid -- ) over - spaces type ;              \ Main              0 value d 0 value e \ location of '.' 'E'              \ Convert real number r to string with n digits of precision.       \ Use fixed-point if exponent -1 to n or scientific otherwise.       : (FG.) ( r n -- c-addr u )        dup >r 1- (fs.) 2dup [char] . scan ?dup if ( not nan/inf)        over to d [char] E scan over to e cskip        /number 2nip d>s dup -1 r@ within if ( fixed-point)        >r [char] . d dup r@ 0< 2* 1+ + over r@ abs move        r@ + c! ( a u) drop e over - r>        then        then r> 2drop ;              : FG.R ( r n u -- ) >r (fg.) r> s.r ; \ print right-justified              \ behead d e              --- SoupGate-DOS v1.05        * Origin: you cannot sedate... all the things you hate (1:229/2)    |
[   << oldest   |   < older   |   list   |   newer >   |   newest >>   ]
(c) 1994, bbs@darkrealms.ca