Defensive code can prevent unforeseen circumstances from causing fatal program behaviors. A common defensive code pattern is to guard against dereferencing the values null or undefined. However, if the situation that some defensive code guards against never can occur, then the defensive code serves no purpose and can safely be removed.

Examine the surrounding code to determine if the defensive code is worth keeping despite providing no practical use. If it is no longer needed, remove it.

The following example shows a cleanupLater function that asynchronously will perform a cleanup task after some delay. When the cleanup task completes, the function invokes the provided callback parameter cb. To prevent a crash by invoking cb when it has the value undefined, defensive code guards the invocation by checking if cb is truthy.

However, the cleanupLater function is always invoked with a callback argument, so the defensive code condition always holds, and it is therefore not required. The function can therefore be simplified to:

Guarding against the same situation multiple times is another example of defensive code that provides no practical use. The example below shows a function that assigns a value to a property of an object, where defensive code ensures that the assigned value is not undefined or null.

However, due to coercion rules, v == undefined holds for both the situation where v isundefined and the situation where v isnull, so the v == null guard serves no purpose, and can be removed:

  • Wikipedia: Defensive programming.
    •