McCabe's Cyclomatic Complexity Number - Software Measurement

COMPSCI 702: Software Measurement
McCabes Cyclomatic Complexity Number

Ewan Tempero e.tempero@cs.auckland.ac.nz
www.cs.auckland.ac.nz:ewan
Emilia Mendes (supervisor) emilia@cs.auckland.ac.nz
COMPSCI 702 c Ewan Tempero p.1/14
Lecture Overview
Structure complexity McCabes Cyclomatic Complexity Number References Fenton and Peeger, Chapters 7 & 8 Potential Exam Question
Explain the intuition that supports using McCabes Cyclomatic Complexity Number as a metric for complexity.
Structure complexity: Intuition

The more complex the structure of code . . . the harder it is to understand . . . the more likely it will have a defect . . . the harder it will be to change . . . the longer it will take to produce . . . the more difcult it will be to reuse
What is structure?
control-ow the sequence of instructions that are
executed
data-ow the creation and movement of data between
components of the code

data organisation the relationship of data items to each
other
Directed Graphs
a mathematical structure (with an appealing visual
representation) for representing things that are related consists of vertices (or nodes, or points), connected by edges (or line segments, or arcs). Vertices: A, B, C, D, E, F Edges: (A,C), (A,C), (A,E), (B,C), (B,F), (C,D), (D,A), (E,F),
A
D E F B
Also: undirected graphs, rules restricting edges between
vertices, classication of vertices Graph Theory the study of properties of graphs
Flowgraphs
Directed graphs (owgraphs) can be used to model control
ow of a program
vertex = statement, edge = (A, B) if control ows from
statement A to B
properties of owgraphs may provide information about
properties of the code
Example
A B C D E F G public static boolean isPrime(int n) { boolean prime = true; A int i = 2; while (i < n) { B if (n % i == 0) { prime = false; C } i++; } D return prime; } E
F G
McCabes Cyclomatic Complexity Number (CCN)

measures the number of linearly independent paths through
the owgraph
v(F ) = e n + 2, F the owgraph of the code, n the number
of vertices, e the number of edges

Intuition the larger the CCN the more complex the code Various sources recommend a CCN of no more than 10-15 Example: CCN = 8 7 + 2 = 3
Example 2
A B C D public static boolean isPrimeA(int n) { for (int i = 2; i < n; i++) { A if (n % i == 0) { return false; B } } return true; C }
(counter) Example
if (a) if (b) goto d: else goto e: else if (c) goto d: else goto e: d: print "Hi. " e: print "How are you?"
if (u) if (v) if (w) print "Hi" print "How are you?"
CNN= 11 9 + 2 = 4
CNN= 7 5 + 2 = 4
(counter) Example (2)

From graph theory, v(f ) = d + 1, where d is the number of
predicate nodes (tests/conditions)

public static boolean isPrimeA(int n) { for (int i = 2; i < n; i++) { if (n % i == 0) { return false; } } return true; } public static boolean dividableBy4(int n) { if (n % 2 == 0) { if ((n/2) % 2 == 0) { return true; } } return false; }
Switch statements
switch (a) { case one: break; case two: // fall through case three: break; default: } .... if (a == one) { // do case one } else if (a == two) { // do case two // do case three } else if (a == three) { // do case three } else { // default }
Exceptions
public void aMethod(Object obj) { // some code without conditionals obj.toString(); // more code without conditionals }
... try { // some code FileInputStream f = new FileInputStream("foo") // more code } catch (FileNotFoundException e) { e.printStackTrace(); } catch (SecurityException e2) { e2.printStackTrace(); } finally { System.out.println("finally!"); } ...
Validation
Attribute relation model: direct metric Unit Denition Model: number of linearly independent paths Instrumentation Model: construct the owgraph of the code,
determine the number of edges and vertices compute or (much easier) identify condition vertices compute
Measurement protocol model: for loops, exceptions, switch
statements
Entity population Model: does it make sense to talk about
average CCN?
Representation Condition: for complexity dubious for number of linearly independent paths valid

McCabe's Cyclomatic Complexity Number - Software Measurement

Uploaded by

Copyright:

Available Formats

McCabe's Cyclomatic Complexity Number - Software Measurement

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

McCabe's Cyclomatic Complexity Number - Software Measurement

Uploaded by

Copyright:

Available Formats

COMPSCI 702: Software Measurement

McCabes Cyclomatic Complexity Number

COMPSCI 702 c Ewan Tempero p.1/14

COMPSCI 702 c Ewan Tempero p.2/14

Structure complexity: Intuition

COMPSCI 702 c Ewan Tempero p.3/14

components of the code

COMPSCI 702 c Ewan Tempero p.4/14

Also: undirected graphs, rules restricting edges between

vertices, classication of vertices Graph Theory the study of properties of graphs

COMPSCI 702 c Ewan Tempero p.5/14

properties of the code

COMPSCI 702 c Ewan Tempero p.6/14

COMPSCI 702 c Ewan Tempero p.7/14

McCabes Cyclomatic Complexity Number (CCN)

of vertices, e the number of edges

COMPSCI 702 c Ewan Tempero p.8/14

COMPSCI 702 c Ewan Tempero p.9/14

if (u) if (v) if (w) print "Hi" print "How are you?"

COMPSCI 702 c Ewan Tempero p.10/14

(counter) Example (2)

predicate nodes (tests/conditions)

COMPSCI 702 c Ewan Tempero p.11/14

COMPSCI 702 c Ewan Tempero p.12/14

COMPSCI 702 c Ewan Tempero p.13/14

COMPSCI 702 c Ewan Tempero p.14/14

You might also like