51 lines
1.2 KiB
Go
51 lines
1.2 KiB
Go
package topological
|
||
|
||
import (
|
||
"algorithms-go/DataStructure/graph"
|
||
"algorithms-go/DataStructure/stack"
|
||
)
|
||
|
||
//拓扑排序指的是将有向无环图（又称“DAG”图）中的顶点按照图中指定的先后顺序进行排序。
|
||
|
||
//拓扑排序实质上就是一种偏序到全序的排序算法
|
||
|
||
//应用场景:排课问题，课程有依赖关系
|
||
|
||
//对有向无环图进行拓扑排序，只需要遵循两个原则：
|
||
//在图中选择一个没有前驱的顶点 V；
|
||
//从图中删除顶点 V 和所有以该顶点为尾的弧。
|
||
|
||
func Sort(g *graph.DirGraph) *stack.Stack {
|
||
var visit func(v graph.VertexId)
|
||
sorted := stack.New()
|
||
visited := make(map[graph.VertexId]bool)
|
||
marked := make(map[graph.VertexId]bool)
|
||
|
||
visit = func(v graph.VertexId) {
|
||
//判断是不是一个无环图
|
||
if marked[v] {
|
||
panic("not a DAG")
|
||
}
|
||
marked[v] = true
|
||
//获取出度的顶点
|
||
successors := g.GetSuccessors(v).VerticesIter()
|
||
for successor := range successors {
|
||
if !marked[successor] && !visited[successor] {
|
||
visit(successor)
|
||
}
|
||
}
|
||
visited[v] = true
|
||
marked[v] = false
|
||
sorted.Push(int(v))
|
||
}
|
||
|
||
vertices := g.VerticesIter()
|
||
for vertex := range vertices {
|
||
if !visited[vertex] {
|
||
visit(vertex)
|
||
}
|
||
}
|
||
|
||
return sorted
|
||
}
|