what do you want

1. float 9.801E+07 into long 98010000;
2. string "9.801E+07" to string "98010000"
2. string "9.801E+07" to string long 98010000

I guess he wants float to string for display purposes, which can be done in a very small string handling routine. Dunno if there is a standard C function, but there is in my Pascal compiler ! And I have written similar routines in Pascal in around 8 lines of code.

This code is ancient hack to get round real-string issues in another compiler. Its more than 20 years old !

The STR procedure is a standard function to turn a number into a string, but it can't handle scientific notation

The VAL function turns strings into numbers.

Ureal is the number to be converted
Ustr is the resulting string
dp is the number of significant figures.

<pre>
PROCEDURE using( Ureal : REAL; VAR Ustr : STRING[15];dp:BYTE);

{procedure to strip normalised REALS and spit them out in a sensible}
{format}

VAR
dum,pos_ctr : BYTE;
eb : integer;
count : WORD;
num:STRING [15];
T_str2:STRING [5];


BEGIN
str (Ureal,num);
If dp>0 THEN
BEGIN
Ustr:='';
FOR count:=11 to length (num) DO
Ustr:=Ustr+NUM[count];
val (Ustr,eb,dum); {Eb is the exponent}
pos_ctr:=3;
Ustr:='';
FOR count:=0 to (DP-1) DO
BEGIN
Ustr:=Ustr+num[pos_ctr];
inc (pos_ctr);
END;
Ustr:=Ustr+'.';
FOR count:=(DP-1) to 3 DO
BEGIN
Ustr:=Ustr+num[pos_ctr];
inc (pos_ctr);
END;
Ustr:=Ustr+'E';
eb:=eb-dp;
STR (eb,t_str2);
Ustr:=Ustr+t_str2;
END
ELSE
Ustr:=num;

END; {of procedure USING}

Steve