1ECVT(3) Linux Programmer's Manual ECVT(3)
2
3
4
6 ecvt, fcvt - convert a floating-point number to a string
7
9 #include <stdlib.h>
10
11 char *ecvt(double number, int ndigits, int *decpt, int *sign);
12
13 char *fcvt(double number, int ndigits, int *decpt, int *sign);
14
16 The ecvt() function converts number to a null-terminated string of
17 ndigits digits (where ndigits is reduced to an system-specific limit
18 determined by the precision of a double), and returns a pointer to the
19 string. The high-order digit is non-zero, unless number is zero. The
20 low order digit is rounded. The string itself does not contain a deci‐
21 mal point; however, the position of the decimal point relative to the
22 start of the string is stored in *decpt. A negative value for *decpt
23 means that the decimal point is to the left of the start of the string.
24 If the sign of number is negative, *sign is set to a non-zero value,
25 otherwise it's set to 0. If number is zero, it is unspecified whether
26 *decpt is 0 or 1.
27
28 The fcvt() function is identical to ecvt(), except that ndigits speci‐
29 fies the number of digits after the decimal point.
30
32 Both the ecvt() and fcvt() functions return a pointer to a static
33 string containing the ASCII representation of number. The static
34 string is overwritten by each call to ecvt() or fcvt().
35
37 These functions are obsolete. Instead, sprintf() is recommended. Linux
38 libc4 and libc5 specified the type of ndigits as size_t. Not all
39 locales use a point as the radix character (`decimal point').
40
42 SVr2; marked as LEGACY in POSIX.1-2001.
43
45 ecvt_r(3), gcvt(3), qecvt(3), setlocale(3), sprintf(3)
46
47
48
49 1999-06-25 ECVT(3)