Choosing Between Loop Types

Once you learn that control structures can be placed inside one another, it is tempting to conclude that any arrangement is acceptable. If the language lets you write something, the reasoning goes, then it must also be good style. Keep that belief in mind as you read, because this article examines whether it holds.

You have already met two familiar combinations. An if statement can be placed inside a loop, and a while loop can be placed inside another while loop. The source makes a broader point, though: any combination of one control structure inside another is possible. Nesting is not restricted to a pair of structures. It can continue through several levels. A while loop inside an if statement, with that if statement inside another while loop, is one such case.

The easiest way to read a deeply nested structure is from the outside in. Start with the outermost container, then move to what it holds, and so on down. Each layer wraps the one beneath it, so the innermost statement is governed by every container around it.

Among all the ways to nest control structures, the nested for loop deserves particular attention. The source states that nested for loops arise naturally in many algorithms and that understanding how they work is important. This is not an unusual trick reserved for rare programs. When a problem has two dimensions or two independent counts, a loop inside a loop is often the natural way to express it. Printing a multiplication table is a standard example, because it needs one count for rows and another count for columns.

The key to reading a nested for loop is remembering that the inner loop is not a one-time event. Each full pass of the outer loop causes the inner loop to start again from its beginning and run through all of its own iterations before control returns to the outer loop. The outer loop therefore controls how many times the whole inner loop runs, and the inner loop controls how many times the body runs within each of those passes.

Trace the execution step by step. With row equal to 1, the inner loop runs with col equal to 1, 2, and 3, printing 1, 2, and 3. When the inner loop's condition becomes false, control leaves the inner loop and reaches the println, which ends that line. Then the outer loop's update step runs, row becomes 2, and the inner loop starts again from col equal to 1. It prints 2, 4, and 6, and the inner loop finishes once more before the outer loop's condition is checked and the program ends.

Predicting how many times the inner body runs is the central skill here. Because the inner loop runs fully on every outer pass, you can count in two stages. First count how many times the inner body runs in one complete inner loop. Then multiply by the number of outer passes. Each outer pass contributes the same full set of inner iterations.

Java places no limit on how deeply control structures may be nested. The compiler accepts any depth that is syntactically valid, so the real constraint is human. The source explains that, as a practical matter, a program with more than a few levels of nesting is difficult to understand. A reader must keep each enclosing condition in mind while following the code inside it, and each added level increases that burden.

A useful test is to ask whether a reader can follow the code without repeatedly scrolling back to see which enclosing structure governs the current line. When the answer is no, the nesting has likely grown past the point of comfortable reading. The source does not prescribe a fixed numeric limit, so judgment about the specific program is part of the skill.