Structure of programs
@2012-2013
Introduction
Software engineering = problem solving activity:
Understanding the problem
Designing an algorithm as a solution
Implementing the algorithm in a computer program
Algorithm: sequence of steps that take from the
input to the output for solving a problem.
Correct: provide a correct solution according to the
specifications
Finite : terminate
General: work for every instance of a problem
Efficient : use few resources (time, memory, etc.)
Structure of computer programs
Program
Library, package
File, class
Function, procedure,
method
Block
Statement
Expression
Word,
token
Book
Part
Chapter
Section
Paragraph
Sentence
Phrase
Word
Covered topics
Control structures
Data structures
Functions and procedures
Separation of concerns
I. CONTROL STRUCTURES
1. Sequence
2. Selection
3. Repetition
5
Introduction
Computer program represents an algorithm
resolving a given problem.
All computer programs, no matter how simple or
how complex, are written using one or more of
three basic structures:
Sequence
Selection
Repetition
These structures are called control structures or
logic structures, because they control the
program logic
6
1. Sequence structure
The sequence structure in a computer program
directs the computer to process the statements
one after another, in the order listed in the
program
1. Sequence structure
A statement may be:
Assignment statement
Input /output statement
Composite statement
2. Selection structure
Make a decision, and then take an appropriate
action based on that decision
Provide the appropriate action to take based on
the result of that decision
The decision depends on various condition values
2. Selection structure
Single input
single output
Single input
double output
condition
Single input
multiple output
condition
TRUE
S1
TRUE
S1
value
FALSE
S2
v1
S1
vn
Sn
10
2. Selection structure
11
3. Repetition structure
Allow the programmer to specify that an action
should be repeated, depending on the condition
value
When used in a program, the repetition structure,
also referred to as a loop, directs the computer to
repeat one or more statements until some
condition is met, at which time the computer
should stop repeating the statements
12
3. Repetition structure
FALSE
S1
..
Sn
condition
TRUE
S1
..
Sn
condition
FALSE
TRUE
13
3. Repetition structure
Case 1: The repetition number is known in
advance
14
3. Repetition structure
Case 2: The repetition number is not known in advance
Statements in the body of this repetition structure are
executed repeatedly as long as the loop-continuation test is
evaluated to false
The condition is first evaluated: may be none of these statements
is executed
Otherwise, these statements are executed at least one time
15
II. Data structures
How to choose or devise the appropriate data
structures for a problem ?
Algorithms will have to manipulate data in some way
The way we choose to store and organize our data (i.e.
data structure) directly affects the efficiency of our
algorithm
Classic data structures : design, implementation and use
16
Type
Primitive type:
Integer, Boolean, String,
Composite type
Tuple
Abstract data type: data structure that is defined
indirectly by the operations that may be performed on
it, and the mathematical properties of those
operations
Array
List
Tree
Hash
Graph
17
III. Functions and procedures
Functions and procedures are part of computer
program. They can be custom defined.
A function or a procedure is built out of control
structures in order to manipulate on the
determined data structures.
Functions are really mathematical relations that
map every input to exactly one output. Functions
are designed to return their output value.
Procedures are recipes for computation that
perform side effects.
Either function or procedure is used to represent
a concern.
18
Example: Functions and procedures
C, C++, Java : no distinct
Pascal, .NET:
function returns value
procedure doesnt return value.
DBMS:
procedures (SPROCs) : stored compiled queries
functions (UDFs): built-in piece of expressions used to
build queries
19
IV. SEPARATION OF
CONCERNS
1.
2.
3.
4.
Principles
Concerns
Types of separation
Stakeholders of concerns
20
1. Principles
The principle of separation of concerns states that
software should be organized so that each program
element does one thing and one thing only.
Each program element should therefore be
understandable without reference to other elements.
Program abstractions (procedures, objects, etc.)
support the separation of concerns.
Procedural programming languages such as C and Pascal
can separate concerns into procedures.
Object-oriented programming languages such as Java can
separate concerns into objects.
Service-oriented architecture can separate concerns into
services.
21
2. Concerns
A concern is an area of interest or focus in a
system.
Concerns are the primary criteria for
decomposing software into smaller, more
manageable and comprehensible parts that have
meaning to a software engineer.
Procedural programming, describing concerns as
procedures
Object-oriented programming, describing concerns as
objects
22
3. Types of separation
Quality: deal separately different quality aspects
of the system
E.g.: security
Time: plan the activity of a system
E.g.: software life cycle
View: consider & analyze separately the system
E.g.: control flow, data flow
Size: dominate the system complexity
E.g.: component
23
4. Stakeholder concerns
Functional concerns: related to specific functionalities to be
included in a system
Quality of service concerns: related to the non-functional
behaviors of a system
Policy concerns: related to the overall policies that govern
the use of the system
System concerns: related to attributes of the system as a
whole, such as its maintainability or its configurability
Organizational concerns: related to organizational goals and
priorities such as:
producing a system within budget
making use of existing software assets
maintaining the reputation of an organization
24
5. Example: Banking system
Concerns are not program issues but reflect the system
requirements and the priorities of the system stakeholders.
By reflecting the separation of concerns in a program, there
is clear mapping from requirements to implementation.
Transfer
Deposit
Withdraw
25
Example:
Banking systems core concerns
Core concerns: functional concerns relating to the
primary purpose of a system
Transfer
Transfer
concern
Deposit
Deposit
concern
Withdraw
Withdraw
concern
26
Example: Transfer concern
Core concern allows bank customers to transfer
an amount of money from a given account to
another account
void transfer(Account fromAccount, Account toAccount, int amount) {
if (fromAccount.getBalance() < amount) {
throw new InsufficientFundsException();
}
fromAccount.withdraw(amount);
toAccount.deposit(amount);
}
27
Example:
Banking systems secondary concerns
Secondary concerns: functional concerns that
reflect non-functional and QoS requirements
Authentication Concern
Authentication Concern
Logging Concern
Logging Concern
Transfer
Transfer
Deposit
Deposit
concern
concern
Transaction Concern
Transaction Concern
Authentication Concern
Logging Concern
Withdraw
Withdraw
al
concern
Transaction Concern
28
Example: authentication, transaction
and logging concerns
void transfer(Account fromAccount, Account toAccount, int amount) throws Exception {
if (!getCurrentUser().canPerform(OP_TRANSFER)) {
Authentication Concern
throw new SecurityException();
}
Transaction Concern
Transaction tx = database.newTransaction();
try {
if (fromAccount.getBalance() < amount) {
throw new InsufficientFundsException();
Transfer Concern
}
fromAccount.withdraw(amount);
toAccount.deposit(amount);
Transaction Concern
tx.commit();
Logging Concern
systemLog.logOperation(OP_TRANSFER, fromAccount, toAccount, amount);
}
catch(Exception e){
Transaction Concern
tx.rollback();
throw e;
}
CONCLUSION
30
Key points: control structures
Control structures:
Sequence
Selection
Repetition
Data structures
Abstracted by means of operations that can be
performed on a domain of values
Functions and procedures
Operations representation, depending on different
contexts of use: programming languages, technologies,
etc.
31
Key points: separation of concerns
Dominate complexity
Separate the issues to concentrate on one at a
time
Divide & conquer
Support parallelization of efforts and separation
of responsibilities
32
Discussion
What is a well structured program ?
Correct control structures
Adequate data structures
Concise functions and procedures
Easy to read, understand, modify and verify
How to write a well structured program ?
Algorithm
Code
Documentation
33
Exercises
Write a well structured program that
reads a set of floating point values from the user in the
range [0, 100]
computes and prints the average,
finds and prints the maximum value,
finds and prints the minimum value,
prints the values in sorted order
Using
Sequence, selection, and repetition structures
At least 2 different programming languages
34
