Texas Instruments TI89 Developer Guide - Page 434

392 Appendix A: System Routines - Direct Floating Point Operations ck_valid_float Declaration: BOOL ck_valid_float (BCD16 * num) Category: Direct Floating Point Operations Description: Verifies that a floating-point value is within the valid user range. The internal TI BCD floating-point format has 16 mantissa digits with an exponent range of -16384 to 16383 while a valid user float is 14 mantissa digits with an exponent range of -999 to 999. ck_valid_float rounds the input BCD16 value num to 14 digits and underflows the value to 0 if the exponent is less than -999. If num is NAN or the exponent is greater than 999, FALSE is returned and num will contain a NAN. (See section 2.9.4. Floating-Point Representations in the TI-89 / TI-92 Plus Sierra C Assembler Reference Manual.) Inputs: num - Pointer to a TI BCD floating-point value (BCD16). Outputs: num is rounded to 14 mantissa digits or underflowed to 0 if necessary to insure that its contents are within the range of a user floating-point value. In this case, TRUE will be returned. If the input num is NAN or the exponent is greater than 999, the output value of num will be NAN and FALSE will be returned. Assumptions: None Side Effects: None Availability: All versions of the TI-89 / TI-92 Plus. TI-89 / TI-92 Plus Differences: None See Also: is_float_transfinite, push_Float Example: /* convert an internal floating point value to the range of a user float */ if( !ck_valid_float( &temp )) /* does round14 and underflows if necessary */ ER_throw( ER_OVERFLOW ); /* could not convert to valid user float range */ TI-89 / TI-92 Plus Developer Guide Not for Distribution Beta Version January 26, 2001