Skip to content
Tutorials Center
My learning

Topics

Java Reference ยท Streams and Functional

Java Stream limit() and skip()

Reference for Java Stream.limit() and skip(): take the first n elements, drop the first n, build pages, and understand the cost on ordered and parallel streams.

limit(n) truncates a stream to at most n elements, and skip(n) discards the first n. Combined they cut out a window, which is how you implement paging over a stream. Both depend on the stream's encounter order: "first" means first in that order.

limit() is a short-circuiting operation, so it is what makes infinite streams usable. skip() is not: it must still produce the elements it throws away.

Java
List<Integer> page = numbers.stream().skip(20).limit(10).toList();

Try it

Java Windows over a stream

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with IO.readln() or a Scanner), it gets the next line from this box.

Java Infinite streams and invalid arguments

Press Run (or Ctrl + Enter) to see the result here.

Program input ยท what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with IO.readln() or a Scanner), it gets the next line from this box.

Syntax

Java
Stream<T> limit(long maxSize)
Stream<T> skip(long n)
Name Description
maxSize Maximum number of elements to keep. Must not be negative.
n Number of leading elements to discard. Must not be negative.

Returns: the sliced stream. Throws: IllegalArgumentException for a negative argument, thrown when the method is called.

Usage notes

  • Counts are long. Convert page arithmetic to long before multiplying to avoid int overflow.
  • Over- and under-sized: limit larger than the stream, or skip beyond its end, is fine; you get all elements or none.
  • Short-circuit: limit stops pulling from upstream once it has enough, but stages before it run only for the elements actually used.
  • Parallel and ordered streams: both operations must respect encounter order, which can be expensive in parallel. Call unordered() first if you only need any n elements.
  • Not a database page. Skipping 100,000 elements still visits them; for large offsets query by key instead.
  • Sorting first: sorted().limit(n) gives the n smallest, but buffers all input.
  • Dynamic prefix: to stop at a condition rather than a count use takeWhile().
  • List.subList(from, to) is a cheaper O(1) window when the source is already a list.