uglycalc's Introduction

UglyCalc

Simple calculator with add, subtract and multiply that supports lazy evaluation.

Setting up

Install

Compile: (Windows)

g++ -std=gnu++11 uglycalc.cpp -o uglycalc.exe

On linux you can use the makefile: make.

Run

Run: (Windows)

uglycalc.exe

to read from the standard input, or

uglycalc.exe < [filepath]

if the input is in a file, or

uglycalc.exe [filepath]

to read from a file.

On linux you can use make test and put the test files in the input folder.

Assumptions & Limitations

If there are cycles in the dependecy graph of the input variables, then there can be contradictions. For example:

a add b
b add 10
b add a
print a

will result in a = 10 + a, a contradiction. So, my calculator throws an exception whenever there is a cycle in the dependency graph of the input variables. Even the case of a self-cycle like a add a.

The problem specification does not ban full numeric register names, like 123. I assume this is not allowed since it causes ambiguity. For example the following input:

x add 10
10 add 20
x add 10
print x

can return 20 or 30 or 40. So, my calculator does not allow full numeric register names.

My calculator supports integer numbers that fit in c++ int data type (including negative numbers), which usually corresponds to a 32-bit integer and returns a warning if an input number is out of the range of int. It also assumes that all the evaluations whenever print is called stay within the long long range, otherwise the printed value may be wrong.
I use map<string, int> map_values for memoization, i.e. to avoid computing the value of a register more than once during each evaluation.

Recommend Projects