Lecture 4

DFS (Revisited) & Topological Sort

 DFS(V, E)
1. for each u V u v w

2. do color[u] WHITE
3. prev[u] NIL x y z
4. time 0
5. for each u V
6. do if color[u] = WHITE
7. then DFS-VISIT(u)

Every time DFS-VISIT(u) is called, u becomes the

root of a new tree in the depth-first forest

2
 DFS-VISIT(u)
1. color[u] GRAY u v w

2. time time+1
3. d[u] time
x y z
4. for each v Adj[u] time = 1
5. do if color[v] = WHITE u v w
1/
6. then prev[v] u
7. DFS-VISIT(v) x y z
8. color[u] BLACK u v w

9. time time + 1
1/ 2/

10. f[u] time

x y z
3
 Example
u v w u v w u v w
1/ 1/ 2/ 1/ 2/

3/
x y z x y z x y z

u v w u v w u v w
1/ 2/ 1/ 2/ 1/ 2/
B B
4/ 3/ 4/ 3/ 4/5 3/
x y z x y z x y z

u v w u v w u v w
1/ 2/ 1/ 2/7 1/ 2/7
B B F B
4/5 3/6 4/5 3/6 4/5 3/6
x y z x y z x y z
4
 Example (cont.)
u v w u v w u v w
1/8 2/7 1/8 2/7 9/ 1/8 2/7 9/
C
F B F B F B

4/5 3/6 4/5 3/6 4/5 3/6

x y z x y z x y z

u v w u v w u v w
1/8 2/7 9/ 1/8 2/7 9/ 1/8 2/7 9/
C C C
F B F B F B
B B
4/5 3/6 10/ 4/5 3/6 10/ 4/5 3/6 10/11
x y z x y z x y z

u v w The results of DFS may depend on:

1/8 2/7
C
9/12 The order in which nodes are explored
F B in procedure DFS
B
4/5 3/6 10/11 The order in which the neighbors of a
x y z vertex are visited in DFS-VISIT
5
 Edge Classification
Tree edge (reaches a WHITE u v w
1/
vertex):
(u, v) is a tree edge if v was first
discovered by exploring edge (u, v) x y z

Back edge (reaches a GRAY

u v w
vertex):
1/ 2/
(u, v), connecting a vertex u to an B
ancestor v in a depth first tree 4/ 3/
x y z
Self loops (in directed graphs) are
also back edges

6
 Edge Classification
Forward edge (reaches a BLACK u v w
vertex & d[u] < d[v]): 1/ 2/7
F B
Non-tree edges (u, v) that connect a vertex 4/5 3/6
u to a descendant v in a depth first tree x y z

Cross edge (reaches a BLACK vertex

& d[u] > d[v]): u v w
1/8 2/7 9/
Can go between vertices in same depth-first F B
C

tree (as long as there is no ancestor / 4/5 3/6

descendant relation) or between different x y z
depth-first trees

7
 Analysis of DFS(V, E)
1. for each u V
2. do color[u] WHITE (V)
3. [u] NIL
4. time 0
5. for each u V (V) exclusive
6. do if color[u] = WHITE of time for
7. then DFS-VISIT(u) DFS-VISIT

8
 Analysis of DFS-VISIT(u)
1. color[u] GRAY DFS-VISIT is called exactly
2. time time+1 once for each vertex

3. d[u] time
4. for each v Adj[u]
5. do if color[v] = WHITE Each loop takes
6. then [v] u |Adj[v]|
7. DFS-VISIT(v)
8. color[u] BLACK
9. time time + 1 Total: vV |Adj[v]| + (V) = (V + E)
10. f[u] time (E)
9
 Properties of DFS
u = prev[v] DFS-VISIT(v) was
called during a search of us
u v w
adjacency list 1/ 2/

3/
x y z
Vertex v is a descendant of vertex u
in the depth first forest v is
discovered during the time in which
u is gray
10
 Parenthesis Theorem
y z s t
In any DFS of a graph G, for 3/6 2/9 1/10 11/16
all u, v, exactly one of the
4/5 7/8 12/13 14/15
following holds: x w v u
1. [d[u], f[u]] and [d[v], f[v]] are
s t
disjoint, and neither of u and v
is a descendant of the other z v u

2. [d[v], f[v]] is entirely within y w

[d[u], f[u]] and v is a
x
descendant of u
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
3. [d[u], f[u]] is entirely within (s (z (y (x x) y) (w w) z) s) (t (v u) (u u) t)

[d[v], f[v]] and u is a Well-formed expression: parenthesis are

descendant of v properly nested
11
 Other Properties of DFS
Corollary
Vertex v is a proper descendant of u u
1/8 2/7 9/12
d[u] < d[v] < f[v] < f[u]
C
F B
B
4/5 3/6 10/11
v

Theorem (White-path Theorem)

In a depth-first forest of a graph G, vertex u
v is a descendant of u if and only if at time 1/ 2/

d[u], there is a path u v consisting of

only white vertices. v

12
 Directed Acyclic Graph
DAG Directed graph with no cycles.
Good for modeling processes and structures that
have a partial order:
a > b and b > c a > c.
But may have a and b such that neither a>b nor b>a.
Can always make a total order (either a>b or
b>a for all a b) from a partial order.
 Characterizing a DAG
Lemma 22.11
A directed graph G is acyclic iff a DFS of G yields no back edges.

T T T
v u

Already Proved in Previous Class !

 Topological Sort
Topological sort of a directed acyclic graph G =
(V, E): a linear order of vertices such that if there
exists an edge (u, v), then u appears before v in
the ordering.

Directed acyclic graphs (DAGs)

Used to represent precedence of events or processes
that have a partial order
a before b b before c What about
a before c
b before c a before c a and b?

Topological sort helps us establish a total order

15
 Topological Sort
Want to sort a directed acyclic graph (DAG).

A B D

C E

A B C D E

Think of original DAG as a partial order.

Want a total order that extends this partial order.

 Topological Sort - Application
Application 1
in scheduling a sequence of jobs.
The jobs are represented by vertices,
there is an edge from x to y if job x must be
completed before job y can be done
(for example, washing machine must finish before we put the
clothes to dry). Then, a topological sort gives an order in
which to perform the jobs
Application 2
In open credit system, how to take courses (in order)
such that, pre-requisite of courses will not create any
problem

17
 Topological Sort (Fig Cormen)
undershorts 11/ 16 17/18 socks TOPOLOGICAL-SORT(V, E)
1. Call DFS(V, E) to compute
12/15
pants shoes 13/14 finishing times f[v] for each
shirt 1/8 vertex v
6/7 belt 2. When each vertex is finished,
watch 9/10
tie
insert it onto the front of a
2/5
linked list
3. Return the linked list of
jacket 3/4
vertices

socks undershorts pants shoes watch shirt belt tie jacket

Running time: (V + E)
18
Solutions

19
Solutions

20
Solutions

21
 Next Class

Strongly Connected Component

22