| 1 | .th ECVT III 4/30/73 |
| 2 | .sh NAME |
| 3 | ecvt \*- output conversion |
| 4 | .sh SYNOPSIS |
| 5 | .ft B |
| 6 | jsr pc,ecvt |
| 7 | .s3 |
| 8 | jsr pc,fcvt |
| 9 | .s3 |
| 10 | char *ecvt(value, ndigit, decpt, sign) |
| 11 | .br |
| 12 | double value; |
| 13 | .br |
| 14 | int ndigit, *decpt, *sign; |
| 15 | .s3 |
| 16 | char *fcvt(value, ndigit, decpt, sign) |
| 17 | .br |
| 18 | .li |
| 19 | ... |
| 20 | .ft R |
| 21 | .sh DESCRIPTION |
| 22 | .it Ecvt |
| 23 | is called with a floating point number in fr0. |
| 24 | .s3 |
| 25 | On exit, the number has been converted into a |
| 26 | string of ascii digits in a buffer pointed to by r0. |
| 27 | The number of digits produced is controlled |
| 28 | by a global variable \fI\*_ndigits\fR. |
| 29 | .s3 |
| 30 | Moreover, the position of the decimal point is |
| 31 | contained in r2: r2=0 means the d.p. is at the |
| 32 | left hand end of the string of digits; |
| 33 | r2>0 means the d.p. is within or to the right |
| 34 | of the string. |
| 35 | .s3 |
| 36 | The sign of the number is indicated by r1 (0 for +; 1 for \*-). |
| 37 | .s3 |
| 38 | The low order digit has suffered decimal rounding |
| 39 | (i. e. may have been carried into). |
| 40 | .s3 |
| 41 | From C, |
| 42 | the |
| 43 | .it value |
| 44 | is converted and a pointer to a null-terminated |
| 45 | string of \fIndigit\fR digits is returned. |
| 46 | The position of the decimal point is stored indirectly |
| 47 | through \fIdecpt\fR (negative means to the left of the |
| 48 | returned digits). |
| 49 | If the sign of the result is negative, |
| 50 | the word pointed to by \fIsign\fR is non-zero, otherwise |
| 51 | it is zero. |
| 52 | .s3 |
| 53 | \fIFcvt\fR is identical to \fIecvt\fR, except that the correct digit |
| 54 | has had decimal rounding for F-style output of the number |
| 55 | of digits specified by \fI\(*_ndigits\fR. |
| 56 | .sh "SEE ALSO" |
| 57 | printf(III) |
| 58 | .sh BUGS |