Function nom::number::streaming::double

pub fn double<T, E: ParseError<T>>(input: T) -> IResult<T, f64, E> where
    T: AsBytes + InputLength + Slice<RangeFrom<usize>>, 

Recognizes floating point number in a byte string and returns a f64

streaming version: will return Err(nom::Err::Incomplete(_)) if it reaches the end of input

use nom::number::streaming::double;

let parser = |s| {
  double(s)
};

assert_eq!(parser("11e-1;"), Ok((";", 1.1)));
assert_eq!(parser("123E-02;"), Ok((";", 1.23)));
assert_eq!(parser("123K-01"), Ok(("K-01", 123.0)));
assert_eq!(parser("abc"), Err(Err::Error(("abc", ErrorKind::Float))));

this function uses the lexical-core crate for float parsing by default, you can deactivate it by removing the "lexical" feature