spliterator

Spliterator is an interface in the Java programming language (in the java.util package) that represents a source of elements which can be traversed and optionally partitioned into multiple parts. It was introduced with the Streams API in Java 8 to enable efficient and scalable parallel processing of data sources such as collections, arrays, and I/O streams.

A spliterator provides both traversal and, potentially, partitioning capabilities. Core operations include:

- tryAdvance, which consumes the next element if present and applies a given action to it, returning true; otherwise returns false.

- forEachRemaining, which repeatedly applies an action to all remaining elements.

- trySplit, which attempts to partition the source into two parts and returns a new spliterator for one part, allowing parallel work; if split is not possible, it returns null.

- estimateSize, which returns an estimate of the number of remaining elements.

- characteristics, which returns a bitwise combination of properties describing the source, such as ORDERED, DISTINCT, SORTED, SIZED, NONNULL, IMMUTABLE, CONCURRENT, and SUBSIZED.

- getComparator (when the SORTED characteristic is present), which returns the element comparator or throws IllegalStateException if not applicable.

Characteristics provide information that helps the framework optimize traversal and splitting. For example, SIZED indicates that estimateSize is exact; SUBSIZED indicates that sub-spliterators will also have known sizes.

Specialized subinterfaces exist for primitive streams, namely Spliterator.OfInt, Spliterator.OfLong, and Spliterator.OfDouble, which use primitive consumers to avoid boxing.

Spliterators form a bridge between data sources and streams, enabling source-aware parallelism and efficient iteration. They are commonly implemented by collections and by utility classes, and can be constructed or adapted for custom data sources.