michael-hartmann/parsefloat

Convert a string to a double similar to atof

Parse a string to a float

The function strtodouble converts a string to a double similar to atof. It is written in C99 and only depends on libc.

Why?

strtodouble is a locale-independent version of atof and strtodouble. atof and strtod respect the locale which means that the digit point is locale-dependent. While in some applications this is exactly what you want, it is not suited for reading floats from data files or configurations files. Also, changing the environment and thus changing the locale is not always possible.

Usage

The prototype of the function is:

double strtodouble(const char *str, int *success)

The string you want to parse is given as the pointer str. The parsing stops at the first invalid character. The function returns the parsed value or NAN (not a number) if an error occured. If success is not NULL, at exit success will be 1 if the conversion was successfull or 0 if an error occured. Note that checking success is the only way to tell the difference between an error and parsing a NAN.

Limitations

There is a number of limitations:

• Hexadecimal floats are not supported.
• Rounding-off errors may occur; this is most likely not an issue in most cases.
• The function calls pow which might be slow.

Testing

For tests see tests.c.