Topic: rtos and rounding

[DanB]

Can someone shed some light on the apparent rounding or display of the following code. Ideally I would like the reported information not to round and display the number of decimal places as shown.

Code: [Select]

(defun c:test1 (/ calcarea)           
 (setq calcarea 43559.999999)
 (prompt (strcat "\n Variable is:    "  (rtos calcarea); this displays "43559.999999"
                  "\n "
                  "\n Area:    "  (rtos calcarea 2 2); this displays "43560", why not 43559.99
                  "\n "
                  "\n Area:    "  (rtos calcarea 2 4); this displays "43560", why not 43559.9999
                  "\n "
                  "\n Area:    "  (rtos calcarea 2 6))) ; this displays "43559.999999"
 (princ)
)


[tombu]

My results were as expected:
Command: (rtos calcarea 2 2)
"43560.00"
Command: (rtos calcarea 2 4)
"43560.0000"
Command: (rtos calcarea 2 6)
"43559.999999"
Command: (rtos calcarea 2 5)
"43560.00000"

[Crank]

The DIMZIN variable affects real-to-string conversions performed by the AutoLISP rtos  and angtos  functions.

[DanB]

DIMZIN is a variable for suppressing zeros from a given value.

To clarify my problem/request: if the input value is "43559.999999" - I would like to be able to return the value "43559.99" by using (rtos calcarea 2 2). Where am I going wrong?

[Lee Mac]

To clarify my problem/request: if the input value is "43559.999999" - I would like to be able to return the value "43559.99" by using (rtos calcarea 2 2). Where am I going wrong?

The value 43559.999999 rounded to 2 decimal places is 43560

It sounds like you actually want to truncate to 2 decimal places?

Code - Auto/Visual Lisp: [Select]

1. _$ (setq x 43559.999999)
2. 43560.0 ;; Console automatically rounds double output
3. _$ (rtos (- x (rem x 0.01)) 2 2)
4. "43559.99"

[DanB]

Yes, truncating the value should achieve my goals. Thanks again.

Not sure I fully understand what Lee provided (the dash or negative proceeding the "x" in line 3??) but going to see what I can learn from it, thank you.

Code: [Select]

(rtos (- x (rem x 0.01)) 2 2)

Edit - Nevermind, with some minor testing I'm starting to see what's happening in line #3 - thanks.

[d2010]

I fully understand what Lee provided (the dash or negative proceeding the "x" in line 3??) but going to see what I can learn from it, thank you.

;;bellow. We use this code . You can translate manually to autolisp..
;;This solution is for zero-trailing from autocadR14.

Code: [Select]

(defun dfn_real_rtos (x151 / rr aa bb un od dz xx)
/*c2s:
       rr=nil,
       xx=x151,
       un=getvar(_T("LUNITS")),
       dz=_T("DIMZIN");
       if ((un==4)||(un==5)) rr=rtos(xx);
            else { od=getvar(dz);
                   setvar(dz,(od&(~8)));
                   aa=rtos(xx);
                   setvar(dz,(od|8));
                   bb=rtos(xx,un,15);
                   setvar(dz,od);
                   rr=equal(distof(aa),distof(bb),0.000001)?aa:bb;
                 };
*/
rr)