Interview Questions: Collections

Q15 · ●●● — What is the difference between an array and an ArrayList?

Weak answer: "An array has a fixed size and an ArrayList can grow."

Strong answer: "An array is fixed at creation — you must know the size up front, and it cannot change. An ArrayList grows and shrinks as you add and remove, and it gives you methods like add, remove, contains, and size. Arrays hold primitives directly; an ArrayList holds objects. In practice I use an ArrayList for almost everything in test code, because you rarely know in advance how many results or rows you will collect. Arrays still appear in main(String[] args) and wherever a fixed set is genuinely fixed."

Follow-up: "How would you get the number of items in each?"

Where in this book: Chapter 5.

Q16 · ●●● — What is the difference between a List, a Set, and a Map?

Weak answer: "A List is ordered, a Set has no duplicates, and a Map has keys and values."

Strong answer: "A List keeps insertion order and allows duplicates, so it fits a record of everything that happened, such as every test result in order. A Set holds each item once and has no positions, so it fits questions like 'have I already seen this?' — distinct failure reasons, or pages a crawler has visited. A Map stores key-value pairs and looks a value up by its key, which fits configuration and lookups: environment name to URL, test name to result. The choice comes from the question you need to answer, not from preference."

Follow-up: "Which would you use to count how many times each status appears?"

Where in this book: Chapters 5 and 10.

Q17 · ●●● — How do you get a value out of a Map, and what happens if the key is not there?

Weak answer: "You use get(), and it returns null if the key is missing."

Strong answer: "get(key) returns the value, and a missing key returns null rather than throwing — which is exactly why it causes trouble. The null travels on and produces a NullPointerException somewhere else entirely. So I either check with containsKey first, or use getOrDefault(key, fallback), which returns my default when the key is absent. In a config loader that is the difference between a clear message and a mystery failure three classes away."

Follow-up: "What happens if you put a key that already exists?"

Where in this book: Chapter 10.

Q18 · ●●○ — Does a HashMap keep the order you inserted things in?

Weak answer: "No, it is unordered."

Strong answer: "No, and it also does not sort. The order it returns is an internal detail that you must never rely on — I have seen it come back Search, Login, Cart when inserted Login, Search, Cart. For a tester that matters because an assertion depending on HashMap order will pass for months and then fail for no visible reason. If order is part of the requirement, I say so in the type: LinkedHashMap for insertion order, TreeMap for sorted keys. Then the guarantee is in the code rather than in hope."

Follow-up: "How would you write that assertion so order does not matter?"

Where in this book: Chapter 10.

Q19 · ●●○ — How do you loop through a Map?

Weak answer: "With a for-each loop over the keys."

Strong answer: "Three views are available depending on what I need. keySet() for the keys, values() for the values, and entrySet() when I need both — which is the common case in reporting. The entry loop gives me getKey() and getValue() on each pair, so building a summary line per test is one loop. If I only need to count something, values() is enough."

Follow-up: "Show me the entry loop."

Where in this book: Chapter 10.

Q20 · ●●○ — How do you check whether a list contains a particular value?

Weak answer: "Use contains()."

Strong answer: "contains(value) returns true or false, and it uses equals under the hood, so it is case-sensitive for Strings. That is a real trap in test data: a list containing 'PASS' will report false for 'pass', and the test quietly takes the wrong branch. So I normalize case when the data is human-entered. For my own classes, contains only works correctly if I have overridden equals, because otherwise it compares object identity."

Follow-up: "What if the list holds your own TestResult objects?"

Where in this book: Chapter 5, with the equality rule in Chapters 6 and 7.

Q21 · ●●○ — What is the difference between ArrayList and LinkedList?

Weak answer: "ArrayList uses an array and LinkedList uses nodes."

Strong answer: "ArrayList stores items in a resizable array, so reading by index is fast and it is the right default. LinkedList chains nodes together, which makes inserting or removing in the middle cheaper but reading by index slower. The honest senior answer is that ArrayList wins in almost all real code, including test frameworks, because we mostly add to the end and read through. The more useful point is that both are a List, so I declare List on the left and can swap the implementation without touching any other line."

Follow-up: "So why declare List rather than ArrayList?"

Where in this book: Chapters 5 and 12.

Q22 · ●●○ — What does Set.add() return, and why is that useful?

Weak answer: "It adds an item to the set."

Strong answer: "It returns a boolean: true if the item was actually added, false if it was already present. That is genuinely useful, because it answers 'is this new?' and 'record it' in one call. A crawler tracking visited pages, or a report collecting distinct failure messages, needs exactly that. For my own classes it depends on equals and hashCode being written correctly, otherwise the set will happily store two objects I consider identical."

Follow-up: "What does the set size come to if you add the same value three times?"

Where in this book: Chapter 10.