Summary

Conclusions

An array has a fixed size, and specialized IntArray and similar types avoid boxing values on the JVM. The contents of arrays are compared with contentEquals, and the rectangular shape of a two-dimensional array must be checked separately. The collection hierarchy separates reading from modification, so functions accept the minimal sufficient interface. A read-only interface does not make an object immutable, and toList() creates a snapshot of the structure but does not copy mutable elements. A list preserves order and duplicates, a set ensures uniqueness by equals and hashCode, and a map holds one value per key. Keys and elements of hash structures must remain stable with respect to equality, and a collection must not be modified during an ordinary traversal. ArrayDeque expresses a stack and a queue, and PriorityQueue selection by a comparator, which must define ties. A structure is chosen by its operations, the data volume, and the ownership contract, and it is tested on an empty set, boundaries, duplicates, and snapshot independence.

Self-check questions

1. How does Array<Int> differ from IntArray?
2. How do you compare the contents of two arrays?
3. Does Array<IntArray> guarantee a rectangular shape?
4. Why does val not make a list immutable?
5. What is the difference between a live view and a snapshot?
6. How does remove differ from removeAt?
7. Why can subList be dangerous after the parent changes?
8. What is the relationship between equals and hashCode?
9. How do you distinguish a missing key from a nullable value?
10. When does getOrPut call the initial function again?
11. How do you remove elements during traversal?
12. Why is traversing a PriorityQueue not sorting?

Useful links

• https://kotlinlang.org/docs/arrays.html.
• https://kotlinlang.org/docs/collections-overview.html.
• https://kotlinlang.org/docs/list-operations.html.
• https://kotlinlang.org/docs/set-operations.html.
• https://kotlinlang.org/docs/map-operations.html.