Python provides special syntax to apply higher-order functions as part of executing a def statement, called a decorator. Perhaps the most common example is a trace.

>>> def trace(fn):
        def wrapped(x):
            print('-> ', fn, '(', x, ')')
            return fn(x)
        return wrapped

>>> @trace
    def triple(x):
        return 3 * x

>>> triple(12)
->  <function triple at 0x102a39848> ( 12 )
36

In this example, A higher-order function trace is defined, which returns a function that precedes a call to its argument with a print statement that outputs the argument. The def statement for triple has an annotation, @trace, which affects the execution rule for def. As usual, the function triple is created. However, the name triple is not bound to this function. Instead, the name triple is bound to the returned function value of calling trace on the newly defined triple function. In code, this decorator is equivalent to:

>>> def triple(x):
        return 3 * x

>>> triple = trace(triple)

In the projects associated with this text, decorators are used for tracing, as well as selecting which functions to call when a program is run from the command line.

Extra for experts. The decorator symbol @ may also be followed by a call expression. The expression following @ is evaluated first (just as the name trace was evaluated above), the def statement second, and finally the result of evaluating the decorator expression is applied to the newly defined function, and the result is bound to the name in the def statement. A short tutorial on decorators by Ariel Ortiz gives further examples for interested students.