Package tools.refinery.interpreter.localsearch.planner

Class LocalSearchRuntimeBasedStrategy

java.lang.Object

tools.refinery.interpreter.localsearch.planner.LocalSearchRuntimeBasedStrategy

public class LocalSearchRuntimeBasedStrategy
extends Object

This class contains the logic for local search plan calculation based on costs of the operations. Its name refers to the fact that the strategy tries to use as much information as available about the model on which the matching is initiated. When no runtime info is available, it falls back to the information available from the metamodel durint operation cost calculation. The implementation is based on the paper "Gergely Varró, Frederik Deckwerth, Martin Wieber, and Andy Schürr: An algorithm for generating model-sensitive search plans for pattern matching on EMF models" (DOI: 10.1007/s10270-013-0372-2)

Constructor Summary

Constructors

Constructor	Description
LocalSearchRuntimeBasedStrategy()

Method Summary

Modifier and Type	Method	Description
protected SubPlan	convertPlan(Set<PVariable> initialBoundVariables, PlanState searchPlan)	Converts a plan to the standard format
protected PlanState	plan(PBody pBody, Set<PVariable> initialBoundVariables, IQueryBackendContext context, ResultProviderRequestor resultProviderRequestor, LocalSearchHints configuration)	The implementation of a local search-based algorithm to create a search plan for a flattened (and normalized) PBody

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

LocalSearchRuntimeBasedStrategy

public LocalSearchRuntimeBasedStrategy()

Method Details

convertPlan

protected SubPlan convertPlan(Set<PVariable> initialBoundVariables,
 PlanState searchPlan)

Converts a plan to the standard format

plan

protected PlanState plan(PBody pBody,
 Set<PVariable> initialBoundVariables,
 IQueryBackendContext context,
 ResultProviderRequestor resultProviderRequestor,
 LocalSearchHints configuration)

The implementation of a local search-based algorithm to create a search plan for a flattened (and normalized) PBody

Parameters:

pBody - for which the plan is to be created

initialBoundVariables - variables that are known to have already assigned values

context - the backend context

resultProviderRequestor - requestor for accessing result providers of called patterns

configuration - the planner configuration

Returns:

the complete search plan for the given PBody

Since:

2.1