base-utils

Cross platform library for conversion between bases in C99

Overview:

• The initial project intention was to offer specialized functions for conversion into decimal numbers from much bigger bases. (Mainly base 36 and base 62)
• The idea to add more generic functions came up since base conversion is not straightforward in C language at all and there is no high level built-in functions for that purpose.

base-utils is currently used on a much bigger project for hash processing.

Basic usage:

When using the specialized functions, knowing which types will be handled by the application can give the code a good speed boost.

Over a million iterations, benchmarks shows that the unsigned flavor is 7% faster than the signed one.

    // Unsigned conversion from base 36 to base 10
    uint64_t dec = uconvb36tob10("ab16j901x10");
    // Unsigned conversion from base 62 to base 10
    uint64_t dec = uconvb62tob10("8AmbA1i1KKaTam68");

Only use the signed function if sure that negative representation should be handled as well.

    // Signed conversion from base 36 to base 10
    int64_t dec = convb36tob10("-ab16j901x10");
    // Signed conversion from base 62 to base 10
    int64_t dec = convb62tob10("-8AmbA1i1KKaTam68");

Converting from a base to another can be done using the following function, if both are below base 62

    // Conversion from base 10 to base 2 (It even supports negative representations too!) 
    char* bin = basetobase("10", 10, 2);

Using a custom charset, you can hop between custom base representations easily.

    // Instead of using the default 0 and 1 within base 2,
    // the function will use the given charset to represent the value
    char* num = basetobasecustom("10", 10, 2, "XyZ");

The library comes with a specialized "atoi" function which can run 2.2x faster than the standard implementation (GCC 7.3.0 Windows 8.1)

    // Casts an unsigned char representation of a number into an actual integer
    int i = uatoi('1');

Warnings:

1. Take care of your representation sizes, if the representation to be converted is bigger than the biggest C data types the application will probably crash.
2. When using the function "basetobasecustom", performance may be poor depending on the length of the custom charset.

Known issues (Feel free to help!):

Currently there are no known issues.