Package de.aristaflow.adept2.model.processmodel.tools

Class NodeRelations

java.lang.Object

de.aristaflow.adept2.model.processmodel.tools.NodeRelations

public class NodeRelations
extends Object

Class which allows for work on execution orders between nodes. FIXME must handle or join of and-or-blocks like and-joins(!) Requires: correct branch IDs, topological order.

Nested Class Summary

Nested Classes

Modifier and Type	Class	Description
static class	NodeRelations.NodeRelation	This represents the relation between two nodes.

Field Summary

Fields

Modifier and Type	Field	Description
protected Logger	logger	my logger

Constructor Summary

Constructors

Constructor	Description
NodeRelations(Template template)	Create a new NodeRelations object for the given template.

Method Summary

Modifier and Type	Method	Description
boolean	atLeastOneIsExecutedBefore(Collection<Integer> optionalPreds, int nodeID)	Determines that at least one of the designated preceding nodes is really executed before the designated reading node.
NodeRelations.NodeRelation	getNodeRelation(int nodeID1, int nodeID2)	Get the BEFORE relation between the given nodes, or what the relation is like, if node1 is not before node2.
String	getStatsString()	Print stats.
static boolean	isBefore(NodeRelations.NodeRelation nodeRelation)
protected static boolean	nodesAreOnDifferentBranchesOfSplit(Template template, int nodeID1, int nodeID2, int commonSplitNodeID)	Returns true, if the given nodes are on different branches relative to the given split node.

Methods inherited from class java.lang.Object

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

Field Detail

logger

protected final Logger logger

my logger

Constructor Detail

NodeRelations

public NodeRelations(Template template)

Create a new NodeRelations object for the given template.

Parameters:

template -

Method Detail

getNodeRelation

public NodeRelations.NodeRelation getNodeRelation(int nodeID1,
                                                  int nodeID2)

Get the BEFORE relation between the given nodes, or what the relation is like, if node1 is not before node2.

Parameters:

nodeID1 - The first node, preferably it is before node2.

nodeID2 - The second node, preferable after node1.

Returns:

The relation between node1 and node2.

See Also:

NodeRelations(Template)

nodesAreOnDifferentBranchesOfSplit

protected static boolean nodesAreOnDifferentBranchesOfSplit(Template template,
                                                            int nodeID1,
                                                            int nodeID2,
                                                            int commonSplitNodeID)

Returns true, if the given nodes are on different branches relative to the given split node.

 false:
        n1          n2
      /    \      /   \
    x ----- x ---x-----x--...
   /
 sn --- ...


 true:
     ... n1
    /
 sn
    \
     ... n2
 

FIXME this is just quick'n'dirty low-performance garbage code, rewrite

Parameters:

template - TODO

nodeID1 -

nodeID2 -

commonSplitNodeID -

Returns:

isBefore

public static boolean isBefore(NodeRelations.NodeRelation nodeRelation)

getStatsString

public String getStatsString()

Print stats.

atLeastOneIsExecutedBefore

public boolean atLeastOneIsExecutedBefore(Collection<Integer> optionalPreds,
                                          int nodeID)

Determines that at least one of the designated preceding nodes is really executed before the designated reading node. The designated preceding nodes may be in an alternative branch and thus only executed "optionally". The reason behind this is that the designated node reads a data element mandatorily which is written by the designated preceding nodes. So each execution path has to contain at least one of these preceding nodes to ensure the data element has been written.

Parameters:

optionalPreds - Predecessors of node, which have an OPTIONAL before relation to node and which write the data element.

nodeID - ID of the node which reads the data element.

Returns:

Whether all execution paths to the designated reading nodes contains at least one of the designated preceding nodes.