From SwiftAPI

C

#include <stdlib.h>
 
char* s = "3.14";
double f = atof(s);
// f == 3.14

C++

#include <string>
#include <sstream>
using namespace std;
 
string s = "3.14";
istringstream iss(s);
double f;
iss >> f;
// i == 3.14

Java

J2SE 1.4.2

String s = "3.14";
float f;
try {
    f = Float.parseFloat(s);
}
catch(NumberFormatException e) {
 
}
// f == 3.14

JavaScript

var s = "3.14";
var f = parseFloat(s);
// f == 3.14

alternatively:

var s = "3.14";
var f = +s;
// f == 3.14

Objective-C

developer.apple.com

@property(readonly) float floatValue

NSString *s = @"3.14";
float f = s.floatValue;
// f == 3.14

Perl

There is no need. Strings and numbers are interchangeable. Strings are automatically converted to numbers when used in arithmetic.

$s = "3.14";
$f = 0 + $s;
# $f == 3.14

Note that however it stops at the first invalid character, and does not cause an error if the string contains invalid characters (it does produce a warning if "warnings" is on though):

$s = "3.14abc159";
$f = 0 + $s;
# $f == 3.14
 
$s = "xyz3.14";
$f = 0 + $s;
# $f == 0

PHP

There is usually no need. Strings and numbers are interchangeable. Strings are automatically converted to numbers when used in arithmetic.

$s = "3.14";
$f = (float)$s;
// $f == 3.14

Python

docs.python.org

s = "3.14"
f = float(s)
# f == 3.14

Ruby

Parses as much of string as valid, or returns 0 if entire string is invalid. Never raises exception. ruby-doc.org

s = "3.14"
f = s.to_f
# f == 3.14

Parses all of string; raises exception if any of it is invalid. ruby-doc.org

s = "3.14"
f = Float(s)
# f == 3.14