Programming Languages

This page examines the differences among programming languages.

Take the Brainbench certification test on OO Concepts

There are two ways of constructing a software design: one way is to make it so simple that there are obviously no deficiencies; the other is to make it so complicated that there are no obvious deficiencies. —C.A.R. Hoare

Topics this page:

Site Map  
About this site  

Language Features Compared

Adapted from John December's Presenting Java
Feature	C++	Java	C#
Inheritance	Multiple	Single	Multiple
Preprocessor	Yes	No	Yes
Header files	Yes	No [1]	Yes
Operator Overloading	Yes	No	Yes
Pointers	Yes	No	Yes
Auto. Garbage Collection	No	Yes	Yes*
Native multi-threading	No	Yes	Yes
Exception handling	Yes	Yes	Yes
Encapsulation	?	?
Polymorphism	?	?

Inheritance "is a" relationship

An object that is inherited from another object automatically receives all of the variables and methods defined by the parent. The extends keyword is used to indicate class inheritance.

Unlike C, a new class written in Java can extend only one class. Java supports single inheritance. However, a Java program can use many interfaces. Interfaces can be used as data types. Interfaces specify the methods the class must have.

The outcome from an Object Oriented (OO) approach to Analysis and Design is an Object Model. During the design of an application, functions are identified and grouped based on common objectives and their relationship to each other:

• is a subfunction (or subclass), and
• Guitarist has a guitar. contained within the other function (or class)

Preprocessor Compiler Directive $Include

Preprocessors handle compiler directives that are written as comments in code. An example:
'$Include header files.

In C, implementation-specific values are defined in the C header file limits.h. Constants are also defined in a header file.

Define an include (.inc) file that defines constants for return values. Include this file in the source code for both the module which calls a function and the module which services the function. This way, if the value changes, the changes will be automatically reflected in all using and consuming functions.

The RobotSQA uses

In pure Java, implementation-specific values are defined in .ini files. Constants are defined with the final modifier.

$156 Object Oriented Systems Development: Using the Unified Modeling Language by Bahrami, Ali. (Irwin McGraw-Hill, 1999)

Encapsulation

Unlike C, COBOL, and other procedureal languages that focus on data members being manipulated by functions, Java and other Obect Oriented languages focus on the relationships among classes (also called user defined types). Data is hidden (encapsulated) and methods are instantiated (created) with data.

Java code is modular: each class implements one function and functions are not intertwined. In other words, the members of one method must not be supporting the accomplishment of the function of another class.

Each OO function is implemented as a class. Access is isolated One class can only see what it needs to access. Access is only available if there is a need to know or a need to use the class or the member of the other class.

States should only be represented by variables that are directly accessible by the object itself, and behaviors should only be represented through methods. Variables of an object should only be accessible outside of the object through methods. That way, if changes are needed within the structure or operation of the object, they can happen without affecting any code outside of the object.

The benefits of this:

• Hides details of implementation. You can modify one class without having to worry about the changes affecting the behavior of another class. The only factor that can affect the other class is the interface that enables the classes to communicate.
• Extensibility. You can inherit a class and its behavior in a subclass, and add to its behavior in the subclass. It keeps the code small and efficient.
• Modulariy. You can manage and maintain modular code much better than you can if it were all one big class.
• Safety. One class does not have access to members of classes that it does not need have access to.

There are two ways Programmers reuse objects within their applications without having to always rewrite everything from scratch.

• Composition "has a" relationship
• Composition is using an object within another object.

Polymorphism - Interfaces

Each set of attribute definitions defines a signature to an interface. Each interface defines a cohesive set of behaviors performed by classes and components. Interfaces are realized when implementing the attributes and operations of the interface.

A class or component may not implement an interface. And several classes or components can implement the same interface.

Java interface names (such as Serializable) end in -able or -ible
Microsoft coding practice typcially prefix interface names with a capital I, resulting in names such as IComponent.

Naming Conventions

Abbreviation	Item
mnu	Menu
btn	Button

Pascal casing means that the first letter of each word is capitalized.

Camel casing — where the first letter of the first word is NOT capitalized — is used to distinguish between two items that would otherwise have the same name, such as when defining property set and get.

Reserved Words

Note: VB keywords have initial caps unless otherwise indicated.

Javascript	Java	MS C++	MS VB	MS C#
abstract
				as
				base
boolean			bool
break
byte
case
cast
catch
char
				checked
class
const
continue
default
				delegate
delete
do
double
else
				enum
extends
				event
				explicit
				extern
false
final
finally
				fixed
float
for
				foreach
function
	future
	generic
goto
if
implements
				implicit
import			Imports
in
	inner
instanceof
int			Integer	int
interface
				internal
				is
				lock
long
				namespace
native
new
null
				object
	operator
	outer			out
				override
package
				params
private
protected
public
	rest
				readonly
				ref
return
				sbyte
				sealed
short
				sizeof
				stackalloc
static
	strictfp
				string
				struct
super
switch
synchronized
this
throw
throws
transient
	true
try
	typeof		TypeOf
				uint
				ulong
				uchecked
				unsafe
				ushort
				using
var
				virtual
void
	volatile
while
	with

VB.NET keywords:


AddHandler
AddressOf
Alias
And
AndAlso
Ansi
As
Assembly
Auto
ByRef
ByVal
Call
Case
CBool
CByte
CChar
CDate
CDec
CInt
CLng
CObj
CShort
CSng
CStr
CType
Date
* Decimal
* Declare
Delegate
Dim
DirectCast
Each
Elseif
End
Enum
Erase
Error
Event
Exit
Friend
Get
GetType
Handles
In
Inherits
Integer
Is
Lib
Like
Loop
Me
Mod
Module
MustInherit
MustOverride
MyBase
MyClass
Namespace
Next
Not
Nothing
NotInheritable
NotOverridable
Object
On
Option
Optional
Or
OrElse
Overloads
Overrides
ParamArray
Preserve
Property
RaiseEvent
ReadOnly
ReDim
Rem
RemoveHandler
Resume
Select
Set
Shadows
Shared
Single
Step
Stop
String
Structure
Sub
SyncLock
Then
To
Unicode
Until
When
With
WithEvents
WriteOnly
Xor

With C#, a keyword can be used as a variable name by appending the "at" symbol @ in front of the name, such as @Xor.

Programming Control Logic

Bohm and Jacopini demonstrated that all programs could be written using only three control structures:
Sequence structures,

Selection structures:

• If-Else statements
• Case sequence structure

Repetition structures:
• For Loops
• While Loops
• Recursive Loops

Complexity

August 1996 NIST Special Publication 500-235 by Arthur H. Watson and Thomas J. McCabe

Javaprepare.com by Naveen Gabrani

Data Structures & Algorithms in Java by Robert Lafore (Sams, 1998), 617 pages.

Typically, 50% of the run time within a program is concentrated within only 4% of the code. —Knuth, D., "An Empirical Study of FORTRAN Programs," Software Practice and Experience, April-June 1971.

Nested If-Else Statements

C

In C and Java there are several formats:
if studentGrade >= 60 {
    "Passed"
} else {
    "Failed"
}

Tertiary operator:

studentGrade >= 60 ? "Passed" : "Failed"

A if-else-if ladder can be more efficient and easier to read than a case statement:

if(month == 12 || month == 1 || month == 2)
season = "Winter";
else if(month == 3 || month == 4 || month == 5)
season = "Spring";
else if(month == 6 || month == 7 || month == 8)
season = "Summer";
else if(month == 9 || month == 10 || month == 11)
season = "Autumn";
else
season = "Bogus Month";

This is useful for logic based on strings and other objects, because Java case statements only handle numbers.

Where ELSE IF is employed in C, PL/SQL uses ELSIF but standard PSM (Persistent Stored Modules) expects ELSEIF.

If you need to add an item upon not finding it in a Hashtable, avoid two operations and use the get method to get the value within the if statement instead of a containskey method followed by a put.

if ( (md = (MyData)htable.get (keyCode) != null ){
return md;
} else {
// insert New Element into Hashtable here.
}

Unix Shell Scripts

In UNIX shells, if commands are ended with if spelled backward. Multiple branching is specified using "elif".

echo "read word 1 \c"
read word1
echo "read word 2 \c"
read word2
if [[ test !-r file1 ]]  
	echo "cannot open file1"
else
	elif test "$word1" = "$word2" then
		echo "The words are the same"
	else
		echo "The words are not the same"
fi 

A space is mandatory before and after each pair of double brackets used to access extended tests.

SAP ABAP IF/ELSE Conditional Programing

Remember the periods (not commas):

if expression1.
ABAP statements.
[elseif expression2.
ABAP statements.]
[else. 
ABAP statements.]
endif.

In UML (and traditionally), the diamond is used to illustrate if/else statements.

Case/Switch Decision Statements

A common mistake is leaving out the break; statement.

switch (inputChar) {
   case 'A':
   case 'a':
      System.out.println("\"" + args[0] + "\" for \"Apple\"");
      break;
   case 'B':
   case 'b':
      System.out.println("\"" + args[0] + "\" for \"Boy\"");
      break;
   case 'C':
   case 'c':
      System.out.println("\"" + args[0] + "\" for \"Cat\"");
      break;
   case 'D':
   case 'd':
      System.out.println("\"" + args[0] + "\" for \"Dog\"");
      break;
   default:
      System.out.println("\"" + args[0] + "\" - Whatever!");
      break;
}

C and Java compilers are silent when it sees a “drop-through”, but the C# compiler issues an error.

Use a consistent design throughout your application for handling the situation when the case value is not anticipated.

C#

Unlike in Java, C, or C++, C# switch statements can have a string as its governing type, with a value of null is permitted as a case label constant.

Case statements in UNIX shells use *) to signal the default and case spelled backwards.

echo "Enter a, b, or c: \c"
read sletter
case $sletter in
a|A)
echo "You entered an a"
;;
b|B)
echo "You entered a b"
;;
c|C)
echo "You entered a c"
;;
*)
echo "You did not enter an a, b, or c"
;;
esac

Ruby

Ruby's range construct:

answer = case age
  when 0..17 then puts "not adult"
  else puts "adult"
end

SAP ABAP CASE Conditional Programing

Remember the periods (not commas):

case counter.
when value1.
ABAP statements.
when value2.
ABAP statements.
endcase.

For Counter-Controlled Repetition

This routine (a set of Java statements) steps through a series of values and performs some action based on whether each value meets some predefined criteria.

This loop is performed 19 times because it starts from 1:

int loopLimit = Interger.parseInt (args[0]);
outer_loop: for (int i = 1 ; i <= loopLimit ; i++ ) {
    for(int j = 0; j < 20; j++) {
        // some code
       if( boolean_expression ) {
           continue outer_loop;
       }
   }
}

To leave a loop conditionally, PL/SQL uses EXIT, or EXIT WHEN(...). PSM uses LEAVE, and puts the leave-statement in an if-statement to exit conditionally.

The generic syntax of a for loop using teritiary operators:

for( initialization value ; conditional expression ; update increment expression ) {
    statements
}

The four basic elements involved in counter-controlled repetition:

• The name of the loop counter, which is also called the loop variable (“i”)
• The initial value of the loop variable (“1”)
• The update expression, which determines how the loop variable is incremented (or decremented) each time through the loop (“i++”)
• The condition that determines when to continue the loop (“i <= loopLimit”)

The for loop exits when the condition in the middle (such as x <= loopLimit) is evaluated as false.

One of the most common errors is the one-off error, typically caused by replacing <= with < in the condition portion of the for loop.

C# provides the foreach statement to iterate through a collection without using multiple statements. This avoids the need to explicitly extracting each element from a collection. Best of all, foreach syntax is specific to the particular collection.

foreach (int number in numbers) {     
	Console.WriteLine(number); 
}

An example using the ASP Fields Collection:

For Each ofield in oRS.Fields
    Response.Write ofield.name
    Response.Write ofield.type
Next

In UNIX shells, the index variable following the "for" is used to iterate from 0 through the list after keyword "in".

for animal in dog cat rabbit iguana cow
do
	echo "I see a "$animal 
done

A space is mandatory before and after each pair of double brackets used to access extended tests.

SAP ABAP Do Loop Programing

Remember the periods (not commas):

do n times.
counter = counter + 1.
if counter = 2.
continue. "another loop pass.
if counter = 3.
exit. "terminate loop.
if counter = 4.
stop. "jump out to end-of-selection section.
else.
	ABAP statements.
enddo.

While Loops

A while loop in C may never be processed since it checks for exit condition before the code is executed. This is also why you should initialize when counters to 1 rather than zero.

i = 1, j = 1;

while (i <=loopLimit) {
   j = 1;
   while (j <= i) {
      System.out.print ("*");
      j++;
   }

   System.out.println (" ");
   i++;
}

Use do loops for code that should be processed at least once, because checking is done after the embedded code is executed.

class WhileTest {
  public static void main (String args[]) {

   int loopLimit = Integer.parseInt (args[0]);
   int i = 1;

   do {
     System.out.println ("The loop count is: " + i);
     i++;
   } while (i <=loopLimit);
  }
} 

Did I mention that while loops check for its exit condition before the embedded code is executed? ;)

A VBScript/ASP example of looping through a file until End Of File:

Response.Write "<table>"
Do while NOT oRS.EOF
   Response.Write "<tr><td>&" & oRS("LastName").Value & "nbsp;</td></tr>"
   oRS.MoveNext
Loop
Response.Write "</table>"
oRS.Close
Set oRS=Nothing

In UNIX shells:

num=0
while test $num -le 10 do
	echo $num
	let num=num+2
done
echo "The value is " $num


until false do
	commands
done

SAP ABAP While Loop Programing

Remember the periods (not commas):
Processing is terminated when the logical expression is false.

while counter < 10
ABAP statements.
counter = counter + 1.
endwhile.

Loops of Logic: Recursion: Factorials

Recursion is when a base class/function calls itself (in a disciplined, hierarchical manner). One reason for recursion is for the calculation of factorials (written as n!), the product of
n * ( n-1 ) * ( n-2 ) * ... * 1

For example, 5! (5 factorial) is the product of 5 * 4 * 3 * 2 * 1

“If the number is less than zero, reject it. If it isn't an integer, round it down to the next integer. If the number is zero or one, its factorial is one. If the number is larger than one, multiply it by the factorial of the next smaller number.” To calculate the factorial of any number that is larger than 1, you need to calculate the factorial of at least one other number. The function you use to do that is the function you're in the middle of already; the function must call itself for the next smaller number, before it can execute on the current number.

Here is an example written in JScript code, using the java.math package and the keyword return:

function factorial(aNumber)  {
// If the number is not an integer, round it down:
aNumber = Math.floor(aNumber);  
// If the number is less than zero, reject it:
if (aNumber < 0)  {  
    return "not a defined quantity";
    }
      // If the number is 0 or 1, its factorial is 1:
      if ((anumber == 0) || (anumber == 1))  {  
      return 1;
      }
        // Otherwise, recurse until done:
        else return (anumber * factorial(anumber - 1));  
}

Recursion was used to create this artwork

Stop It!

To unload a Java executable from memory:
System.exit(0);

The destroy() method is invoked by the browser or applet viewer to notify the applet to release any resources it has used. The applet viewer or browser decides when this is necessary.

If you are running a Java program from a Console Window and the cursor does not provide a system prompt after the program is done, terminate the program by pressing


Ctrl-C

Precedence of Operators

Below are operators listed in order of precedence, from the highest (evaluated firs), to the lowest (evaluated last).

Java operators behave slightly differently than in traditional programming languages. For example: d + (1/2) * d and d + d / 2 are not the same. When Java encounters “(1/2)” or an operation involving integer data types, it performs integer division and throws away (truncates) the remainder. To avoid this, specify “(1.0/2.0)” to force floating point division.

Unary Operators

Parentheses

Parentheses () are used to control the order of precedence. They ensure that expressions are evaluated as intended.

Boolean NOT

The ! (exclamation point character) negates whatever logic is within the parentheses to its right. For example, the result from this statement is true when the variable age contains 21. This simplifies coding an if-then-else statement into just an if-then statement, such as when testing for a lack of desired conditions.
if (!(age == 21))

Bitwise Negation

The ~ (tilde character) negates bits from false to true. Used only with integral data types (int, short, byte, char, long) .

Assignment Operators

Operator	Description	Sample Coding	Expanded Equivalent Coding	Before		After
				x	y	x	y
++	Postfix Increment	x++;	x = x + 1;	9		10
	Prefix Increment	++x;	x = x + 1;	9		10
	Assignment, then Postfix Increment	y = x++;	y = x; x = x + 1;	9	14	10	9
	Prefix Increment & Assignment	y = ++x;	x = x + 1; y = x;	9	14	10	10
--	Postfix Decrement	x--;	x = x - 1;	9		8
	Prefix Decrement	--x;	x = x - 1;	9		8
	Assignment, then Postfix Decrement	y = x--;	y = x; x = x - 1;	9	14	8	9
	Prefix Decrement & Assignment	y = --x;	x = x - 1; y = x;	9	14	8	8

With y = x++; the value assigned to y is the original value of x.
With y = ++x; the value assigned to y is 1 more than the original value of x.

Typecasting

Typecasting, also known as casting, is when programmers explicitly instruct Java to convert the value of a datatype to a new value with a new datatype. This is as opposed to conversion, or implicit type changing performed by Java without programmer intervention.

An explicit cast is needed to convert a narrowing conversion to a datatype with less digits results in a loss of precision: This statements results in an explicit convertion before the arithmetic is performed:

x = (byte) (y * 4);

Narrowing occurs with the transformation of a data type with a larger range and precision into a smaller data type, resulting in a loss of data.

Compile-time casting rules when transforming between a NewType and OldType:

Arithmetic Operators

Operator	Description	Sample Coding	Before		After
			x	y	x	y
*	Multiplication	x = x * y	5	7	35	7
/	Division	x = x / y	35	7	5	7
%	Modulus (remainder)	x = x % y	21	5	1	5	(5 * 4) = 20
+	Addition	x = x + y	12	27	39	27
-	Subtraction	x = x - y	39	12	27	12

One use for Modulus is to determine whether aNumber is odd or even: calculate aNumber modulus 2. A result of zero means aNumber is even.

Shift Operators

Shifting operations move the bit pattern (binary representation) of integral (int or long) data types left or right.

This is commonly done in control systems to align bits that are read from, or to be written to, I/O ports. Tricky programmers use it to efficiently perform integer multiplication or division by powers of two (doubling values).

Shifting can be used with confidence in Java because the bit-level representation of all types is defined and platform-independent.

Operator	Description	Sample Coding	Result if x=6, y=5	Result if x=5, y=5
<<	shift left (multiply by power of 2)	.	.	.	Trailing bits are discarded. Java does not have the Assembly languages rotate operation to define the value of such bits.
>>	shift right (divide by power of 2)	.	.	.	Leading bits are padded with zero.
>>>	unsigned (logical) shift right	.	.	.	Java only because (unlike C/C++) Java numeric types (except char) are signed. Don't use for data typed as byte because promotion occurs before the operator is applied.

Comparative Operators

Comparative Operators yield a result of boolean value True or False.

Operator	Description	Sample Coding	Result if x=6, y=5	Result if x=5, y=5
<	Less than	x < y	false	false
<=	Less than or equal to	x <= y	False	True
>	Greater than	x > y	True	False
>=	Greater than or equal to	x >= y	True	True
.	instanceof	.	.	.
= =	Equals	x = = y	False	True
!=	Not equals	x != y	True	False

To compare the values referenced by two objects rather than the existance of the objects, use the Boolean.equals() method or the String.equals() method.

Where contains text operator is used by CGI

For server-side CGI scripts, the value of CONTENT_TYPE has changed from HTTP 1.0 value of
Content-Type: application/x-www-form-urlencoded

to the HTTP 1.1 value of:

Content-Type: application/x-www-form-urlencoded; charset=ISO-8859-1

So defensive programmers should avoid using an equal operator which expects a precise match to check for “application/x-www-form-urlencoded”. Programmers should instead check whether the value is included in the CONTENT_TYPE string. In Perl, the code is:

if ($ENV{'CONTENT_TYPE'} =~ /application\/x-www-form-urlencoded/) {
    print " Here's your page back.\n";
}

This script would work for Internet Explorer 5.0, Navigator 4.x, and Navigator 5.0 because it allows for the inclusion of a charset specification in the CONTENT_TYPE string.

— from Eric Krock's Updating Your JavaScript and CGI Scripts for Version 5 Browsers

Bit-wise operator usage

Column	Value
1st	1
2nd	2
3rd	4
4th	8
5th	16
6th	32
7th	64
8th	128
9th	256
10th	512
11th	1028
Sum	1184

With SQL 7, the IF statement below following is a bit-wise comparison:

CREATE TRIGGER trigMine ON tableX FOR UPDATE AS
IF COLUMNS_UPDATED() & 14 <> 0

A bit mask value of 14 tests whether the 2nd or 3rd or 4th columns of tableX have been updated. (values 2 + 4 + 8 = 14).

The bitmask to verify if columns 6, 8, or 11 was modified can be determined by the following: power(2,(6-1)) + power(2,(8-1)) + power(2,(11-1)) = 1184.

To detect if any rows have been updated:

If (COLUMNS_UPDATED() & 1184) > 0

To verify that all the rows in question have been updated:

If (COLUMNS_UPDATED() & 1184) = 1184

IP Address Subnetting

Question: What bitwise operation was calculated if 5=0101 and 4=0100, return a result of 5?

Answer: AND, XOR, or OR

Logical Operators

These are rarely used to combine boolean expressions. They yield a result of value True or False.

Operator	Description	Sample Coding	Condition x	Condition y	Result z
&	AND	z = x & y	true	false	false
^	XOR (eXclusive OR)	z = x ^ y	true	true	false
|	OR	z = x | y	true	false	true
&&	Optimized AND	z = x && y	true	false	false
||	Optimized OR	z = x || y	true	false	true

For example: The b increment operation may not be performed if a is less than or equal to 0.

if((a > 0) && (b++ < 1)) {
    // do something
}

Ternary Operator

Operator	Description	Sample Coding	Expanded Equivalent Coding	Before x,y	After x,y	Result z
? :	If-then-else	z = x>y ? 2*x : 2*y	if (x > y) { z = 2*x; } else { z = 2*y; }	10, 30	10, 30	60

Logical Assignment Operators

Operator	Description	Sample Coding	Expanded Equivalent Coding	Before x,y	After x,y
&=	AND Assignment	x &= y	x = x & y;	true, false	false, false
|=	OR Assignment	x |= y	x = x | y;	true, false	true, false
^=	XOR Assignment	x ^= y	x = x ^ y;	true, true	false, true

A	B	AND	XOR	OR
0	0	0	0	0
0	1	0	1	1
1	0	0	1	1
1	1	1	0	1

In boolean logic, 1 = True, 0 = False.

0 OR 0 produces 0. Any 1 produces 1.
1 AND 1 produces 1. Unless all are 1, it's 0.
0 XOR 1 produces 1, as does
1 XOR 0 produces 1. All others produce 0.

Arithmetic Assignment Operators

Operator	Description	Sample Coding	Expanded Equivalent Coding	Before		After
				x	y	x	y
+=	Addition Assignment	x += y;	x = x + y;	12	27	39	27
-=	Subtraction Assignment	x -= y;	x = x - y;	39	12	27	12
*=	Multiplication Assignment	x *= y;	x = x * y;	5	7	35	7
/=	Division Assignment	x /= y;	x = x / y;	35	7	5	7
%=	Modulus (Remainder) Assignment	x %= y;	x = x % y;	21	5	1	5

Classification of Anomalies

Beizer (in his "Software Testing Techniques") and Copeland (in his "A Practitioners Guide to Software Test Design") in the "Technical Testing" field reference those in the "Software Engineering" field who use these annotations:

d = Declaration (creation, initialization) of data in variables (on the left hand side of a computation)
u = Used for something (=c and p)
c - Right hand side of computation, pointer (calculation)
p = Used in a predicate (or as control variable of loop)

k = kill (deallocate)
- = nothing

These annotations are combined in time-sequence pairs to define anomalies (an illogical or useless sequence of data object state change) usually caught by compilers:

dd = Data defined twice.
dk = Data defined and then killed without being used.
kk = Killed twice.
ku = Killed and then used.
-k = variable not defined but killed
-u = used before defined
d- = defined but never used

In compiler design dataflow analysis:

• An UD chain is a list of all definitions that can reach a given use of a variable.
• An DU chain is a list of all uses that can be reached by a given definition of a variable.