The BBC and Master Computer Public Domain Library

Check Number and Convert to Integer if Float

Submitted by Steve Fewell

Routine: CheckNum&ConvFloat
Name: Check Number and Convert to Integer if Number is Floating-Point
Starting Address: &96BE
Entry criteria: The Accumulator specifies which type of value has just been processed.
Exit: If the value was Floating-Point, then the FWA's value will be converted to Integer and stored in the IWA.

Description:
Transfers the current Variable Type [stored in A] to Y in order to update the status flags.

If the current Variable Type = 0 (String) then a Type Mismatch error is given.

If the current Variable Type = positive (i.e. &40, Integer) then nothing to do as the number is already integer, so exit with the IWA's value intact.

Otherwise, the current Variable Type must be negative (i.e. &FF, Floating-Point), so the number is floating-point. So we need to convert the number into an Integer and return it in the IWA.

The Convert Float to Integer routine is used to return a Two's-complement Integer value in the FWA Mantissa (stored Most significant byte first, least significant byte last). Now, store the Integer Number in the IWA - stored the right way around (least significant byte first, and Most significant byte last), and exit.

Disassembly for the Check Number and Convert to Integer if Float routine