Working with Streams and Lambda Expression

---

title: "Working with Streams and Lambda Expression" date: "2024-09-14" excerpt: "High-density notes on Java functional interfaces, stream pipelines, Optionals, reductions, collectors, and parallel processing rules."

tags: "java,streams,lambda" source: "https://shrikant-havale.in/2024/09/14/working-with-streams-and-lambda-expression/"

A practical checklist for Java functional programming and stream APIs.

Standard Functional Interfaces

• Consumer<T>: accepts input, returns nothing.
• Supplier<T>: accepts nothing, returns a value.
• Function<T, R>: transforms input to output.
• Predicate<T>: tests and returns boolean.
• Primitive specializations reduce boxing (IntFunction, ToIntFunction, DoubleSupplier, etc.).

Streams Basics

• Streams are single-use; reusing a consumed stream throws IllegalStateException.
• Infinite sources (generate, iterate) typically require limit().
• peek() does nothing unless a terminal operation exists.
• sorted(...) is valid; sort(...) is not a Stream method.

Finding Elements

• findAny() may return any element.
• findFirst() respects encounter order on ordered streams.

Primitive Streams and Optionals

• DoubleStream.average() returns OptionalDouble.
• Empty optional values throw on direct get...() calls.

IntStream.empty().average().getAsDouble(); // throws NoSuchElementException

map() vs flatMap()

• map() => one output per input.
• flatMap() => zero or more outputs per input (must return streams).

Reduction

• Reductions combine values into one result.
• sum, min, max, count are common reductions.
• Accumulators should be associative and stateless, especially in parallel streams.

Grouping and Partitioning

• groupingBy(...) groups by a classifier and returns a Map.
• partitioningBy(...) groups by boolean predicate.
• For parallel collection scenarios, use concurrent collectors when appropriate.

Lambdas and Method References

• Lambdas are compact implementations of functional interfaces.
• Avoid stateful lambdas in stream pipelines.
• Method references use type inference and can target constructors and methods.

Parallel Stream Notes

• parallel() is for streams; parallelStream() is for collections.
• Some operations (limit, skip, forEachOrdered) can reduce parallel gains.
• Parallel behavior requires careful collector and ordering choices.

Source

Original post: https://shrikant-havale.in/2024/09/14/working-with-streams-and-lambda-expression/