Switch Fallthrough and Scoping

Switch Fallthrough and Scoping

Two subtle switch behaviors that trip up beginners: fallthrough and case scope.

Fallthrough: running the next case by accident

A case normally ends with break. If you forget it, execution keeps going straight into the next case, regardless of its value. That is fallthrough.

switch (x) {
  case 1:
    cout << "one";
    // no break here
  case 2:
    cout << "two";
    break;
}

If x is 1, this prints "onetwo", because with no break at the end of case 1, execution falls through into case 2.

Accidental vs actual intentional

Most of the time fallthrough is a mistake: you simply forgot the break. If fallthrough is truly what you want, mark it explicitly with the C++17 attribute [[fallthrough]]:

case 1:
  cout << "one";
  [[fallthrough]];
case 2:
  cout << "two";
  break;

The attribute tells readers and compilers: this skip is intentional.

Variables and case scope

All the cases share one block as far as variables go. A variable declared in a case can jump into the next case too, which can cause confusion. LearnCpp 8.6 suggests wrapping a case body in curly braces { } to give it its own scope:

case 1: {
  string msg = "One";
  cout << msg << endl;
  break;
}
case 2:
  // msg is not declared here
  break;

TL;DR

• Missing break makes the case fall through into the next one.
• Mark intentional fallthrough with [[fallthrough]] (C++17).
• A variable declared in a case belongs to the whole switch scope.
• Use { } inside a case to limit where the variable lives.