Iterating a Map

Direct Answer

Use entrySet() or Map.forEach((key, value) -> ...) when you need both key and value, because the entry already carries both. Use keySet() for keys only and values() for values only. Iterating keySet() and then calling get(key) performs an avoidable lookup for every entry; iterating the keys itself is not another hash lookup.

Code Example

Choose the view that already contains the data the loop needs.

import java.util.*;

public class MapIteration {
    public static void main(String[] args) {
        Map<Long, String> orders = new HashMap<>();
        orders.put(101L, "Laptop");
        orders.put(102L, "Mouse");

        // Both parts: each entry already contains its key and value.
        for (var entry : orders.entrySet()) {
            System.out.println(entry.getKey() + " -> " + entry.getValue());
        }

        // A concise alternative for a simple action.
        orders.forEach((id, item) -> System.out.println(id + ": " + item));

        // This works, but get(id) is an avoidable lookup when both parts are needed.
        for (Long id : orders.keySet()) {
            String item = orders.get(id);
            System.out.println(id + " => " + item);
        }

        for (Long id : orders.keySet()) System.out.println(id);
        for (String item : orders.values()) System.out.println(item);
    }
}

The third loop does one extra lookup per key. It does not perform two hash lookups merely by walking keySet().

Common Mistakes & Pitfalls

• Using keySet() plus get() for pairs: it works, but each get is avoidable when entrySet() already provides the value.
• Calling entrySet() a hash lookup: it is a live view of the mappings; traversal and key lookup are different operations.
• Assuming every map has the same order: HashMap, LinkedHashMap, and TreeMap have different iteration rules.
• Changing a map without checking its iterator contract: mutation behavior differs across ordinary, unmodifiable, and concurrent maps.