Package org.processmining.framework.util.search

Interface NodeExpander<N>

Type Parameters:
N -
public interface NodeExpander<N>
This class is used by the MultiThreadedSearcher to handle callbacks. Each time the searcher considers a node, it asks the registered expander to expand the node. If a node cannot be expanded, then a call to processLeaf is made. The expander should not process any leafs from within the expandNode method.
Author:
bfvdonge

  • Method Summary

    Modifier and Type
    Method
    Description
    Collection<N>
    expandNode(N toExpand, Progress progress, Collection<N> unmodifiableResultCollection)
    This method is called by the searcher when a node in the search tree has to be considered for expansion.
    void
    processLeaf(N leaf, Progress progress, Collection<N> resultCollection)
    This method is called by the searcher to which this expander is registered each time a leaf was encountered.

  • Method Details

    • expandNode

      Collection<N> expandNode(N toExpand, Progress progress, Collection<N> unmodifiableResultCollection)
      This method is called by the searcher when a node in the search tree has to be considered for expansion. The node to expand is given as a parameter and the method should return a collection of nodes representing the children of this node. If the node turns out to be a leaf, an empty collection has to be returned.
      Parameters:
      toExpand - The node to expand.
      progress - The progress indicator provided to the searcher in which this expander is registered. The expander may increment the progress, but it should check for cancellation, especially when doing long computations.
      unmodifiableResultCollection - the leafs of the search space that have been added to the resultCollection by the processLeaf method. Note that any access to this collection should be synchronized and this collection should not be modified in this method. However the provided collection is modifiable, as for synchronization purposes, the actual resultCollection is provided.
      Returns:
      A collection of child nodes, or an empty collection if this node turns out to be a leaf.

    • processLeaf

      void processLeaf(N leaf, Progress progress, Collection<N> resultCollection)
      This method is called by the searcher to which this expander is registered each time a leaf was encountered. The provided resultCollection was specified in the startSearch method of the searcher and can be used to store this result. A typical implementation is: synchronized(resultCollection) { resultCollection.add(leaf); } However, more advanced implementations are possible. Finally, note that calls to this method are not thread-safe, i.e. the implementing class should take care of synchronization on the resultCollection if necessary (as in the example above).
      Parameters:
      leaf - The node that was found to be a leaf by the expandNode method (i.e. the expandNode method returned and empty collection)
      progress - The progress indicator provided to the searcher in which this expander is registered. The expander may increment the progress, but it should check for cancellation, especially when doing long computations.
      resultCollection - The collection to which to add the leaf node. More advanced computations are allowed here, i.e. the resultCollection can be changed. Note however that this requires syncrhonization on the collection, as calls to this method are not thread-safe.