FixedInt - Unsigned fixed sized Integers
use FixedInt; my \fixedint = FixedInt.new(:8bit); say fixedint; # 0 say fixedint -= 12; # 244 say fixedint.signed; # -12 say fixedint.bin; # 0b11110100 say fixedint.hex; # 0xF4 say fixedint += 2500; # 184 say fixedint.signed; # -72 say fixedint.bin; # 0b10111000 say fixedint.hex; # 0xB8
FixedInt provides an easy way to work with fixed-sized unsigned integer in Raku. Rakus Integers by default are unfixed, arbitrary size. Doing unsigned fixed-size bitwise operations requires a bunch of extra bookkeeping. This module hides that bookkeeping behind an interface.
One major caveat to be aware of when using this module. The class instance may not be instantiated in a $ sigiled variable.
Raku $ sigiled scalar variables do not implement a STORE method, but instead do direct assignment; and there doesn't seem to be any easy way to override that behaviour.
An implication of that is that classes that do implement a STORE method can not be held in a $ sigiled variable. (Well, they can, they just won't work correctly. The first time you try to store a new value, the entire class instance will be overwritten and disappear.)
There are two work-arounds. One is to use an unsigiled variable. Unsigiled variables have no preconception of what may or may not be allowed, so don't override a STORE method. They work well, but if you really want a $ sigiled container, your only real option is to fake it using a constant term (rather than a variable.)
my \int32 = FixedInt.new; # unsigiled constant # or constant term:<$int32> = FixedInt.new; # "sigiled" constant term # looks like a variable but is really a constant term
The FixedInt module allows creating and working with any sized fixed size Integer. Not only the "standard" sizes: 8, 16, 32, 64, etc., but also: 11, 25, 103, whatever. Once instantiated, it can be treated like any other variable. You can add to it, subtract from it, multiply, divide, whatever; the value stored in the variable will always stay the specified bit size. Any excesses will "roll over" the value. Works correctly for any standard bitwise or arithmatic operator, though you must remember to assign to the variable to get the fixed sized properties.
Provides some bit-wise operators that don't make sense for non-fixed size integers. Provides a few "format" operators that return a formatted IntStr; useful for display but also directly usable as an integer value.
Note that most of the method examples below show binary representations of the value, that is just for demonstration purposes, returns decimal numbers by default. All of the method examples below assume an 8 bit fixedint.
Specify the number of bits. May be any positive integer. Defaults to 32. Accepts :bit or :bits. Defaults to 0; will not accept a Nil value.
my \fixedint = FixedInt.new(:8bit); # or my \fixedint = FixedInt.new(:bits(8)); # or my \fixedint = FixedInt.new(bits => 8);
.ror (Int $bits)
Rotate right by the given number of bits
# before 00000011 (3) fixedint.=ror(1); # after 10000001 (129)
.rol (Int $bits)
Rotate left by the given number of bits
# before 10000001 (129) fixedint.=rol(3); # after 00001100 (12)
Ones complement. Every bit is negated.
# before 00001100 (12) fixedint.=C1; # after 11110011 (243)
Twos complement. A "negated" integer.
# before 00001100 (12) fixedint.=C2; # after 11110100 (244)
Treats the fixed size unsigned integer as a signed integer, returns negative numbers for FixedInts with a set most significant bit.
say fixedint # 244 say fixedint.signed # -12
Returns a binary formatted IntStr;
say fixedint # 244 say fixedint.bin # 0b11110100
Returns a octal formatted IntStr;
say fixedint # 244 say fixedint.oct # 0o0364
Returns a hex formatted IntStr;
say fixedint # 244 say fixedint.hex # 0xF4
Steve Schulze (thundergnat)
COPYRIGHT AND LICENSE
Copyright 2020 Steve Schulze
This library is free software; you can redistribute it and/or modify it under the Artistic License 2.0.