pub fn parse_float(s: &str) -> Option<f64>Expand description
Go strconv.ParseFloat(s, 64): out of range is an error rather than ±Inf, NaN takes
no sign, and underscores may separate digits.
ponytail: Go’s hexadecimal form (0x1p-2) is rejected; plugin flags are decimal in
practice. Upgrade by porting atofHex if a plugin needs it.