Package it.unimi.dsi.big.webgraph

Class NodeIterator

java.lang.Object
it.unimi.dsi.big.webgraph.NodeIterator
All Implemented Interfaces:
LongIterator, Iterator<Long>, PrimitiveIterator<Long,LongConsumer>, PrimitiveIterator.OfLong
Direct Known Subclasses:
ArcLabelledNodeIterator
public abstract class NodeIterator extends Object implements LongIterator
This interface extends LongIterator and is used to scan a graph, that is, to read its nodes and their successor lists sequentially. The LongIterator.nextLong() method returns the node that will be scanned. After a call to this method, calling successors() or successorBigArray() will return the list of successors.

Implementing subclasses can override either successors() or successorBigArray(), but at least one of them must be implemented.

  • Field Details

    • EMPTY

      public static final NodeIterator EMPTY
      An empty node iterator.

  • Constructor Details

    • NodeIterator

      public NodeIterator()

  • Method Details

    • outdegree

      public abstract long outdegree()
      Returns the outdegree of the current node.
      Returns:
      the outdegree of the current node.

    • successors

      public LazyLongIterator successors()
      Returns a lazy iterator over the successors of the current node. The iteration terminates when -1 is returned.
      Returns:
      a lazy iterator over the successors of the current node.
      Implementation Specification:
      This implementation just wraps the array returned by successorBigArray().

    • successorBigArray

      public long[][] successorBigArray()
      Returns a reference to an array containing the successors of the current node.

      The returned array may contain more entries than the outdegree of the current node. However, only those with indices from 0 (inclusive) to the outdegree of the current node (exclusive) contain valid data.

      Returns:
      an array whose first elements are the successors of the current node; the array must not be modified by the caller.
      Implementation Specification:
      This implementation just unwrap the iterator returned by successors().

    • copy

      public NodeIterator copy(long upperBound)
      Creates a copy of this iterator that will never return nodes ≥ the specified bound; the copy must be accessible by a different thread. Optional operation (it should be implemented by all classes that allow to scan the graph more than once).
      Parameters:
      upperBound - the upper bound.
      Returns:
      a copy of this iterator, with the given upper bound.
      Implementation Specification:
      This implementation just throws an UnsupportedOperationException. It should be kept in sync with the result of ImmutableGraph.hasCopiableIterators().

    • skip

      public long skip(long n)
      Skips the given number of elements.

      The effect of this call is exactly the same as that of calling LongIterator.nextLong() for n times (possibly stopping if Iterator.hasNext() becomes false).

      This method is a big version of LongIterator.skip(int).

      Parameters:
      n - the number of elements to skip.
      Returns:
      the number of elements actually skipped.
      See Also: