Skip to content

The-Last-Cookie/stringcalc

Repository files navigation

StringCalc

Content

  1. Description
  2. Main features
  3. Technical documentation
  4. Testing
  5. Planned features
  6. Links
  7. Licence

Description

Integers and other data types alike have a specific set of values which they can have. If you want to use bigger numbers than these data types offer, you will need a workaround. Strings can theoretically be infinite (practically, they are limited by the maximum available memory space). This C++ library offers the ability to use strings and calculate equations with them as if they were numbers.

The maximum size of a string is given by std::string().max_size(). In standard C++ it is around 2^32 most of the times. However, this can vary due to compiler and memory implementation. I will try to optimize the code, so that strings can be as long as possible. An idea might be to use the rope data structure. This will need further investigation though.

Every main function can only have positive numbers as input and returns the calculated number as a string. The base parameter can be between 2 and 36.

Main features

This section lists a few examples of calculations that can be done with the library. More examples can be found in the example.cpp file.

Mathematical operations

// Addition
std::cout << "5 + 6 = " << StringCalc::add(10, "5", "6") << "\n";

// Subtraction
std::cout << "12 - 7 = " << StringCalc::sub(10, "12", "7") << "\n";

// Binary multiplication
std::cout << "10111 * 111 = " << StringCalc::mult(2, "10111", "111") << "\n";

// Binary division
std::cout << "10110 / 111 = " << StringCalc::div(2, "10110", "111") << "\n";

Converter

With the converter, you can convert numbers from one base to another. The bases may be between 2 and 36. So there are the numbers 1 to 9 as well as the letters A to Z to represent a number. If you want to use one number several times, it is recommended to parse this number once and then call convertTo each time for a new base. Unfortunately, the converter can get quite slow with big numbers (more than 30 digits).

StringCalc::Converter conv;
conv.parse("1679615", 10);
std::cout << "1679615 Base 10 in Base 16 is: " << conv.convertTo(16);

Technical documentation

Overview

Main functions

Name Description Input Output
add Addition Number base and two positive integers Sum
sub Subtraction Number base and two positive integers Difference
mult Multiplication Number base and two positive integers Product
div Integer division (without remainder) Number base and two positive integers Quotient
mod Modulo Number base and two positive integers Remainder
pow Exponentiation Number base and two positive integers Power
fact Factorial Number base and one positive integer Factorial

Converter methods

Name Description Input Output
parse Parse a number into the converter Number base and one positive integer Nothing
convertTo Convert the number which was given via parse to a specific base Number base Number in given number base

Helper functions

Name Description Input Output Commentary
min Find the string with the smaller value Two positive integers String with smaller value Can lead to unexpected behaviour if the given numbers are invalid
max Find the string with the higher value Two positive integers String with the higher value Can lead to unexpected behaviour if the given numbers are invalid
intToString Convert the given number to its string value One positive integer String value of integer (10 and higher will be converted to A, B, etc. respectively)
charToInt Convert a string value to its integer value Char value Actual integer value of char E.g. A will be converted to 10
removeLeadingZeros Remove zeros in front of a string One positive integer Positive integer without any zeros in front

Private functions

Notice: These functions here are available in the library, however they are used for internal calculations and therefore shouldn't be used when implementing this library as they have no input validation. Though, the functions are listed here, so that the documentation is complete. Additionally, using them without prior caution can lead to unexpected behavior.

Name Description Input Output Commentary
h_add Helper function for addition Number base and two positive integers Sum
h_sub Helper function for subtraction Number base and two positive integers Difference
h_mult Helper function for multiplication Number base and two positive integers Product
h_div_sub Normal integer division with naive subtraction Two positive integers Quotient without remainder Needed as a component for the actual division algorithm
h_div Helper function for integer division Number base, two positive integers and a parameter if quotient or remainder should be returned Quotient or remainder
h_mod Helper function for modulo Number base and two positive integers Remainder
h_pow Helper function for exponentiation Number base and two positive integers Power
h_fact Helper function for factorial Number base and one positive integer Factorial

Testing

This library uses doctest for testing its functionalities. All tests can be found in the tests folder.

However, there is one problem while executing tests. Since I decided to not use custom exceptions, the program will print out errors for every test (only if there are any of course). I'm not sure at the moment how to deactivate/activate these error messages depending on the context, so I will leave it like this for now (I'm aware that you can define a global variable for such things, but I'm not fond of this approach).

To add new tests, you need to create a header file (.h) in the tests subfolder. Every file in there has the following structure:

// Test.h
#pragma once

#include "../doctest.h"

TEST_SUITE("test") {
    // Your test cases
}

As a final step, you need to include the header file in the main.cpp file in the testing directory, like so:

#include "tests/Test.h"

Planned features

Major release v2

  • Merge all number categories into one namespace and use base as indicator
  • Support for negative numbers
  • Use more pointers (call-by-reference)

Links

Licence

This project is licensed under the MIT Licence. See the LICENCE file for more information.