← Java Minute

How can you get the number of characters in a String?

Published on 2026-07-21·v1.0

It's harder than it looks. The answer depends on what you call a "character" — the length() method returns the number of Unicode code units, which isn't always the same as the number of characters you actually see on screen.

Question

How can you get the number of characters in a String?

What It Is

The length() method on String returns the number of Unicode code units in the string. A code unit is a 16-bit value — the type used internally to store char values in Java.

For most everyday text, one code unit maps to exactly one visible character, so length() behaves the way you'd expect.

The Surrogate Pair Problem

Some Unicode characters — like many emoji and characters from certain historical or symbolic scripts — don't fit in a single 16-bit code unit. They need two or more code units to be represented, a pair known as a surrogate pair.

When that happens, length() no longer matches the number of characters visually displayed on the screen: it counts code units, not "characters" in the everyday sense.

Code Unit vs Code Point

Two concepts are essential here:

  • Code unit: corresponds to the char type in Java. It's a 16-bit piece of a character's encoding, and it's what length() actually counts.
  • Code point: corresponds to the int type in Java. It represents the real, complete character — though not every int value is a valid code point.

A code point may be made up of one or two code units, depending on the character.

Practical Example

java
String text = "a😀b"; // "a" + 😀 (grinning face emoji) + "b"

System.out.println(text.length());        // 4 — counts code units
System.out.println(text.codePointCount(0, text.length())); // 3 — counts code points

The emoji 😀 is encoded as a surrogate pair, so it consumes two code units — but it's still just one visible character.

Solution and Conclusion

If you need the true number of characters as perceived by a user, don't rely on length() alone. Use codePointCount() to count Unicode code points instead, and prefer code-point-aware iteration (such as codePoints()) when processing strings that may contain characters outside the basic 16-bit range.

References