mapss.dif.graph
Class PartialTopSortStrategy
java.lang.Object
mocgraph.analysis.strategy.CachedStrategy
mapss.dif.graph.BaseTopSortStrategy
mapss.dif.graph.PartialTopSortStrategy
- All Implemented Interfaces:
- TopSortAnalyzer, mocgraph.analysis.analyzer.Analyzer, mocgraph.analysis.analyzer.GraphAnalyzer
public class PartialTopSortStrategy
- extends BaseTopSortStrategy
A topological sorting approach by building partial orders.
One advantage of partial ordering is that optimization at later stage can make
better decision based on the partial order.
- Version:
- $Id: PartialTopSortStrategy.java 409 2007-05-13 19:47:16Z plishker $
- Author:
- Ming-Yung Ko
- See Also:
TopSortAnalysis
Constructor Summary |
PartialTopSortStrategy(mocgraph.Graph graph)
Construct a top sorting strategy for a given graph. |
Method Summary |
protected java.lang.Object |
_compute()
Compute a topological sorting order. |
protected mocgraph.Node |
_nextSource(java.util.List partialOrder,
java.util.Collection sources)
Return the next source node in determining the top sort. |
Methods inherited from class mocgraph.analysis.strategy.CachedStrategy |
_convertResult, _result, cachingStatus, disableCaching, enableCaching, getCachedResult, graph, obsolete, reset, setCachedResult |
Methods inherited from class java.lang.Object |
clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait |
Methods inherited from interface mocgraph.analysis.analyzer.GraphAnalyzer |
graph |
PartialTopSortStrategy
public PartialTopSortStrategy(mocgraph.Graph graph)
- Construct a top sorting strategy for a given graph.
- Parameters:
graph
- The given graph.
_compute
protected java.lang.Object _compute()
- Compute a topological sorting order. The result is
returned as a
List
, where the order begins at index 0.
- Overrides:
_compute
in class mocgraph.analysis.strategy.CachedStrategy
- Returns:
- A topological sorting order.
_nextSource
protected mocgraph.Node _nextSource(java.util.List partialOrder,
java.util.Collection sources)
- Return the next source node in determining the top sort.
The next source node candidate is determined by the top sort computed
so far as well the candidate sources to append to the parial order.
This method is supposed to be overriden by specific source selection
approach.
- Parameters:
partialOrder
- The partial top sort computed so far.sources
- The candidate sources to append to the partial order.
- Returns:
- The next source node.