[pforth] / fth / numberio.fth

\ @(#) numberio.fth 98/01/26 1.2
\ numberio.fth
\
\ numeric conversion
\
\ Author: Phil Burk
\ Copyright 1994 3DO, Phil Burk, Larry Polansky, David Rosenboom
\
\ The pForth software code is dedicated to the public domain,
\ and any third party may reproduce, distribute and modify
\ the pForth software code or any derivative works thereof
\ without any compensation or license.  The pForth software
\ code is provided on an "as is" basis without any warranty
\ of any kind, including, without limitation, the implied
\ warranties of merchantability and fitness for a particular
\ purpose and their equivalents under the laws of any jurisdiction.

anew task-numberio.fth
decimal

\ ------------------------ INPUT -------------------------------
\ Convert a single character to a number in the given base.
: DIGIT   ( char base -- n true | char false )
   >r
\ convert lower to upper
   dup ascii a < not
   IF
       ascii a - ascii A +
   THEN
\
   dup dup ascii A 1- >
   IF ascii A - ascii 9 + 1+
   ELSE ( char char )
       dup ascii 9 >
       IF
           ( between 9 and A is bad )
           drop 0 ( trigger error below )
       THEN
   THEN
   ascii 0 -
   dup r> <
   IF dup 1+ 0>
       IF nip true
       ELSE drop FALSE
       THEN
   ELSE drop FALSE
   THEN
;

: >NUMBER ( ud1 c-addr1 u1 -- ud2 c-addr2 u2 , convert till bad char , CORE )
   >r
   BEGIN
       r@ 0>    \ any characters left?
       IF
           dup c@ base @
           digit ( ud1 c-addr , n true | char false )
           IF
               TRUE
           ELSE
               drop FALSE
           THEN
       ELSE
           false
       THEN
   WHILE ( -- ud1 c-addr n  )
       swap >r  ( -- ud1lo ud1hi n  )
       swap  base @ ( -- ud1lo n ud1hi base  )
       um* drop ( -- ud1lo n ud1hi*baselo  )
       rot  base @ ( -- n ud1hi*baselo ud1lo base )
       um* ( -- n ud1hi*baselo ud1lo*basello ud1lo*baselhi )
       d+  ( -- ud2 )
       r> 1+     \ increment char*
       r> 1- >r  \ decrement count
   REPEAT
   r>
;

\ obsolete
: CONVERT  ( ud1 c-addr1 -- ud2 c-addr2 , convert till bad char , CORE EXT )
   256 >NUMBER DROP
;

0 constant NUM_TYPE_BAD
1 constant NUM_TYPE_SINGLE
2 constant NUM_TYPE_DOUBLE

\ Like >number, but temporarily switch BASE.
: (>number-with-base) ( ud c-addr u base -- ud' c-addr' u' )
       base @ >r base ! >number r> base !
;

\ This is similar to the F83 NUMBER? except that it returns a number type
\ and then either a single or double precision number.
: ((NUMBER?))   ( c-addr u -- 0 | n 1 | d 2 , convert string to number )
   dup 0= IF 2drop NUM_TYPE_BAD exit THEN   \ any chars?

   base @ -rot                 ( base c-addr u )

   \ Recognize prefixes and change base if needed
   over c@ >r                  ( base c-addr u  ) ( r: char )
   r@ [char] # = if rot drop 10 -rot 1 /string then
   r@ [char] $ = if rot drop 16 -rot 1 /string then
   r@ [char] % = if rot drop  2 -rot 1 /string then
   r@ [char] ' = if
           \ Recognize '<char>'
           dup 3 = if
                   over 2 chars + c@ [char] ' = if
                           drop nip rdrop
                           char+ c@ NUM_TYPE_SINGLE exit
                   then
           then
   then
   r> drop

\ check for '-' at beginning, skip if present
   over c@ ascii - = \ is it a '-'
   dup >r            \ save flag
   IF 1 /string  ( -- base c-addr+1 cnt-1 , skip past minus sign )
   THEN

   ( base c-addr cnt ) ( r: minus-flag )
   rot >r 0 0 2swap r>
   (>number-with-base) dup 0=   \ convert as much as we can
   IF
       2drop    \ drop addr cnt
       drop     \ drop hi part of num
       r@       \ check flag to see if '-' sign used
       IF  negate
       THEN
       NUM_TYPE_SINGLE
   ELSE  ( -- d addr cnt )
       1 = swap             \ if final character is '.' then double
       c@ ascii . =  AND
       IF
           r@      \ check flag to see if '-' sign used
           IF  dnegate
           THEN
           NUM_TYPE_DOUBLE
       ELSE
           2drop
           NUM_TYPE_BAD
       THEN
   THEN
   rdrop
;

: (NUMBER?)   ( $addr -- 0 | n 1 | d 2 , convert string to number )
   count ((number?))
;

' (number?) is number?
\ hex
\ 0sp c" xyz" (number?) .s
\ 0sp c" 234" (number?) .s
\ 0sp c" -234" (number?) .s
\ 0sp c" 234." (number?) .s
\ 0sp c" -234." (number?) .s
\ 0sp c" 1234567855554444." (number?) .s


\ ------------------------ OUTPUT ------------------------------
\ Number output based on F83
variable HLD    \ points to last character added

: hold   ( char -- , add character to text representation)
   -1 hld  +!
   hld @  c!
;
: <#     ( -- , setup conversion )
   pad hld !
;
: #>     ( d -- addr len , finish conversion )
   2drop  hld @  pad  over -
;
: sign   ( n -- , add '-' if negative )
   0<  if  ascii - hold  then
;
: #      ( d -- d , convert one digit )
  base @  mu/mod rot 9 over <
  IF  7 +
  THEN
  ascii 0 + hold
;
: #s     ( d -- d , convert remaining digits )
   BEGIN  #  2dup or 0=
   UNTIL
;


: (UD.) ( ud -- c-addr cnt )
   <# #s #>
;
: UD.   ( ud -- , print unsigned double number )
   (ud.)  type space
;
: UD.R  ( ud n -- )
   >r  (ud.)  r> over - spaces type
;
: (D.)  ( d -- c-addr cnt )
   tuck dabs <# #s rot sign #>
;
: D.    ( d -- )
   (d.)  type space
;
: D.R   ( d n -- , right justified )
   >r  (d.)  r>  over - spaces  type
;

: (U.)  ( u -- c-addr cnt )
   0 (ud.)
;
: U.    ( u -- , print unsigned number )
   0 ud.
;
: U.R   ( u n -- , print right justified )
   >r  (u.)  r> over - spaces  type
;
: (.)   ( n -- c-addr cnt )
   dup abs 0 <# #s rot sign #>
;
: .     ( n -- , print signed number)
  (.)  type space
;
: .R    ( n l -- , print right justified)
   >r  (.)  r> over - spaces type
;