com.complexible.stardog.repair

Interface RepairPlanMetric

Deprecated.

This functionality is deprecated as of 7.5.0

public interface RepairPlanMetric

Cost metric for repair plans. The cost metric has two purposes: 1) Compute the cost for a given plan so two plans can be compared. 2) Compare the operations such that an action that will yield a better plan for this metric can be selected. The plans evaluated by a metric may not be complete

Cost metrics should be monotonic; that is adding a new operation to an existing plan should never decrease its cost. More formally, for two plans plan1 and plan2, plan1.containsAll(plan2) implies metric.computeCost(plan1) >= metric.computeCost(plan2).

Since:

2.3

Version:

4.0

Author:

Evren Sirin

Method Summary

Modifier and Type	Method and Description
Optional<Comparator<com.complexible.stardog.api.update.UpdateOperation>>	comparator() Deprecated. Returns a comparator that will compare two operations and order them such that operations that will yield better, i.e.
double	computeCost(RepairPlan thePlan) Deprecated. Returns the cost of the given plan.

Method Detail

computeCost

double computeCost(RepairPlan thePlan)

Deprecated.

Returns the cost of the given plan. Plans with a smaller cost value are preferred over plans with higher costs. There is no requirement on the range of values this metric returns and even negative values can be returned to indicate highly preferred plans. The monotonicity of values need to be ensured as explained above. Otherwise, the minimality and completeness of plans generated cannot be guaranteed.

comparator

Optional<Comparator<com.complexible.stardog.api.update.UpdateOperation>> comparator()

Deprecated.

Returns a comparator that will compare two operations and order them such that operations that will yield better, i.e. smaller cost, plans with respect to this metric will be less than operations that will yield worse, i.e. higher cost, plans.

This comparator is optional. This function may return an empty value or the comparisons it performs for operations may not be consistent with the costs for containing these operations. In either case, the plans will be computed correctly and the optimal plans will be found. A more accurate operation comparator will speed up the planning process as better plans can be found sooner but the exact same plans will be found with a missing or inaccurate comparator too.