Integer.toString(int i) vs String.valueOf(int i)

Question

In String type we have several method valueOf

static String   valueOf(boolean b) 
static String   valueOf(char c) 
static String   valueOf(char[] data) 
static String   valueOf(char[] data, int offset, int count) 
static String   valueOf(double d) 
static String   valueOf(float f) 
static String   valueOf(int i) 
static String   valueOf(long l) 
static String   valueOf(Object obj) 

As we can see those method are capable to resolve all kind of numbers

every implementation of specific method like you have presented: So for integers we have

Integer.toString(int i)

for double

Double.toString(double d)

and so on

In my opinion this is not some historical thing, but it is more useful for a developer to use the method valueOf from the String class than from the proper type, as it leads to fewer changes for us to make when we want to change the type that we are operating on.

Sample 1:

public String doStuff(int num) {

  // Do something with num...

  return String.valueOf(num);

 }

Sample2:

public String doStuff(int num) {
  
 // Do something with num...
 
 return Integer.toString(num);

 }

As we see in sample 2 we have to do two changes, in contrary to sample one.

In my conclusion, using the valueOf method from String class is more flexible and that's why it is available there.

One huge difference is that if you invoke toString() in a null object you'll get a NullPointerException whereas, using String.valueOf() you may not check for null.

Just two different ways of doing the same thing. It may be a historical reason (can't remember if one came before the other).

The String class provides valueOf methods for all primitive types and Object type so I assume they are convenience methods that can all be accessed through the one class.

NB Profiling results

Average intToString = 5368ms, Average stringValueOf = 5689ms (for 100,000,000 operations)

public class StringIntTest {


    public static long intToString () {
        long startTime = System.currentTimeMillis();
        for (int i = 0; i < 100000000; i++) {
            String j = Integer.toString(i);
        }
        long finishTime = System.currentTimeMillis();

        return finishTime - startTime;
    }

    public static long stringValueOf () {

        long startTime = System.currentTimeMillis();
        for (int i = 0; i < 100000000; i++) {
            String j = String.valueOf(i);
        }
        long finishTime = System.currentTimeMillis();

        return finishTime - startTime;
    }

    public static void main(String[] args) {
        long intToStringElapsed = 0;
        long stringValueOfElapsed = 0;
        for (int i = 0; i < 10; i++) {
            intToStringElapsed += intToString();
            stringValueOfElapsed+= stringValueOf();
        }
        System.out.println("Average intToString = "+ (intToStringElapsed /10));
        System.out.println("Average stringValueOf = " +(stringValueOfElapsed / 10));
    }
}

From the Java sources:

/**
 * Returns the string representation of the {@code int} argument.
 * <p>
 * The representation is exactly the one returned by the
 * {@code Integer.toString} method of one argument.
 *
 * @param   i   an {@code int}.
 * @return  a string representation of the {@code int} argument.
 * @see     java.lang.Integer#toString(int, int)
 */
public static String valueOf(int i) {
    return Integer.toString(i);
}

So they give exactly the same result and one in fact calls the other. String.valueOf is more flexible if you might change the type later.

If you look at the source code for the String class, it actually calls Integer.toString() when you call valueOf().

That being said, Integer.toString() might be a tad faster if the method calls aren't optimized at compile time (which they probably are).