CSC/ECE 517 Fall 2012/ch1 1w21 aa

Primitive Objects

Introduction

Primitive datatypes are those data types that are not defined in terms of other data types[1]. They are basic or in-built types that a programming language provides for its users. These data types can be used for building other complex and composite data types [2]. In most languages, primitive types usually consist of basic value types for representing the data to be stored. As these data types are built-in, the compilers for these languages provide more support for them by supporting most basic operations. In dynamically typed languages like Ruby, these data types are represented as objects and are referred to as primitive objects. This article discusses the various primitive data types/objects that are present in various objected oriented programming languages and their benefits.

Some of the common data types that are present in most programming languages are

• Integer - used to hold whole numbers. Can be signed or unsigned and include various types like byte, short, int, long, etc.
• Character - used to hold single character, digit or symbol or a sequence of any combination of them that is stored in numeric coding. Usually stored in ASCII encoding. '
• Floating-point - used to represent real numbers with limited precision. They usually comprise of an exponent part and a fractional part. Can be signed or unsigned .
• Fixed-point - similar to floating-point, but has fixed number of digits after the decimal point [3]. They are used in processors or representing base 2 and base 10 fractional values where there is no Floating point unit.
• Boolean - used to represent logical values for comparisons. Typically hold either of 2 values - true/false. Can also be represented as integers. Can be stored in a single bit, but typically stored as a byte.

Primitive data types in C++

C++ is a statically typed, objected oriented programming language. It is widely used on a lot of hardware and software platforms. As C++ just adds object oriented features and a few other enhancements to C, the primitive data types that it provides are same as those provided by C[4]. They are described below [5].

Name	Description	Size	Range
int	Basic numerical data. Can have modifiers that vary size and range.	4 bytes	-2,147,483,648 to 2,147,483,647
char	Can hold one character of data like an alphabet, number, or symbol represented in ASCII	1 byte	-128 to 127
wchar_t	used for storing compiler-defined wide characters and unicode characters [6]	compiler defined	compiler defined
float	single precision floating-point	4 bytes	± 3.402,823,4 * 10^38
double	double precision floating-point	8 bytes	+/–1.7 * 10^308 [7]
bool	represents logical values	1 byte	true/false
void	generic identifier that does not identify type [8]	Does not exist	Does not exist

C++ allows variables of any of these data types to be created and also supports using these data types to construct complex data types like structures and classes. C++ also allows pointers that can be used to store the address of a variable of any simple or complex data type.

Primitive data types in Java

There is a certain group of data types that is used very frequently by the programmer. These types are included in the Java language as the primitive data types.These data types have a fixed size which does not change from one system to another and adds to the portability feature of java. There are 8 primitive data types defined in java which are as follows:[9]

Name	Description	Size(in bits)	Range	Usage	Default Value
byte	Signed two's complement integer	8 bits	-128 to 127	This datatype can be used in arrays where there is a space constraint to save memory.	0
short	Signed two's complement integer	16 bits	-32,768 to 32,767	This datatype can also be used to save memory in large arrrays.	0
int	signed two's complement integer	32 bits	-2,147,483,648 to 2,147,483,647	This datatype is generally the default datatype for all the numbers we use in our program.	0
long	signed two's complement integer	64 bits	-9,223,372,036,854,775,808 to 9,223,373,036,854,775,807	This type is used when we require a value that is outside the range of values provided by int.	0L
float	Single-precision IEEE 754 floating point	32 bits	32-bit IEEE 754 floating-point numbers.	Use this datatype to save memory in large arrays.	0.0f
double	Double-precision 64-bit IEEE 754 floating point	64 bits	64-bit IEEE 754 floating-point numbers.	This data type is generally the default for decimal values.	0.0d
boolean	Boolean	1- bit	false, true	Use this data type for flags that track a true or false condition.	false
char	a char is a single 16-bit character encoded using Unicode	16 bit	Unicode character \u0000(0) through unicode character \uffff(65,535)	This type is used to define single characters.	'\u0000'

Besides these 8 primitive data types java provides support to characterstrings through Java.lang.String class. String is not a primitive datatype but its usage and functionality makes us think of it being so. One of the major differences in Java from the other languages is that,the size of variables of these primitive types do not vary from one system to another, ie, it is not machine/architecture dependent. Also, Java does not allow numeric values to be stored unsigned[10].

Java provides wrapper classes for these primitive types and objects of those classes form primitive objects. For example, for the primitive data type int, the Integer class acts as a wrapper class that provides a lot of utility methods like parseInt() and constants like MAX_VALUE and MIN_VALUE that can be used on an Integer object. The other wrapper classes that Java provides for these primitive classes are Integer, Double, Float, Character, Boolean, Long and Short.

int i = Integer.parseInt("50");
System.out.println(i);
System.out.println(Integer.MAX_VALUE);

Output:

50
2147483647

Primitive data types in C#

C# is a statically-typed object oriented programming language. C# provides all the data types that are available in Java, and adds support for unsigned numerals and a new 128-bit high-precision floating-point type. All primitive data types in C# are objects in the System namespace. The various primitive data types in C# are defined below:[11]

Name	.NET Class	Size	Description	Range
byte	System.Byte	8 bits	8-bit unsigned integral type.	0 to 255
sbyte	System.SByte	8 bits	signed two's complement integer	-128 to 127
short	System.Int16	16 bits	16-bit signed integral type.	-32,768 to 32,767
ushort	System.UInt16	16 bits	16-bit unsigned integral type.	0 to 65,535
int	Int32	32 bits	signed two's complement integer	-2,147,483,648 to 2,147,483,647
uint	System.Int32	32 bits	32-bit signed integral type.	0 to 4,294,967,295
long	System.Int64	64 bits	64-bit signed integral type.	-9,223,372,036,854,775,808 to 9,223,373,036,854,775,807
ulong	System.UIint64	64 bits	64-bit unsigned integral type.	0 to 18,446,744,073,709,551,615
float	System.Single	32 bits	Single-precision floating-point type.	-3.402823e38 to 3.02823e38
double	System.Double	64 bits	Double-precision floating-point type.	-1.79769313486232e308 to 1.79769313486232e308
bool	System.Boolean	1 bit	Logical Boolean type	false, true
char	System.Char	16 bits	A 16-bit Unicode character.	Unicode character \u0000 through unicode character \uffff
object	System.Object	N/A	Ultimate base type of all other types.	N/A
string	System.String	N/A	A sequence of Unicode characters.	N/A
decimal	System.Decimal	128	Precise decimal with 28 significant digits.	±1.0 × 10e−28 to ±7.9 × 10e28

Primitive data types/objects in Ruby

Ruby is a multi-paradigm programming language. It is dynamically typed, and allows to be programmed procedural, object-oriented or functional[12]. As it is also a pure object oriented programming language, even the basic inbuilt types like integer and string and even constants are represented as objects. There are a lot of inbuilt classes that are provided by Ruby[13]. These classes differ in some extent to their counterparts in other languages. Having all the basic types as classes allows a lot of methods to be executed on them that reduces the amount of code that needs to be newly written. Some of the basic classes are:

Name	Description	Range
Integer	base class for FixNum and BigNum [14]	depends on FixNum and BigNum
FixNum	store integer values that fit in a native machine word [15]	± 2^30 (for 32-bit systems)
BigNum	store Integer values larger than FixNum[16]	values > 2^30 (for 32-bit systems)
String	sequence of characters/bytes [17]	Does not exist.
Float	real numbers with double precision floating point precision[18]	± 1.7976 * 10^308 (for 32-bit systems)
TrueClass	represents logical true value[19]	true
FalseClass	represents logical false value[20]	false

There are a various other inbuilt classes that are provided by Ruby. Those classes can be constructed by combining one or more of the above given primitive classes. As these are existing classes, there are a lot of helper methods that allow easy calculations and operations on the objects of these classes. Some of these classes also have inbuilt iterator methods that allow the values to be modified automatically for using in loops. As with many other programming languages, some of these objects take up different size (and hence range) on different architectures. These existing classes are also editable and allows to override or add new functionality to them for custom use in the application.

Advantages of using primitive data types

The various advantages of using primitive data types are as follows:

Ease of use

Using primitive data types is much more efficient way of programming than programming using objects. For example, in Java when we use boxing in a loop it causes certain performance issues and is inefficient compared to using the primitive types directly. For more details and a sample program see [21].

Simplicity

Working with primitive data types is more intuitive.

Example: 1. As a counter in a loop.

(int i=0;i<10;i++)

2. In conditional statements.

if(x > max)
    max = x;

Disadvantages of using primitive data types

The various disadvantages of using the primitive data types are as follows:

Primitive objects cannot be extended

Most of the object oriented programming languages do not allow the behavior or capabilities of primitive data types to be modified by programs.For example, the primitive data types in languages such as Java cannot be inherited to create further subtypes. An exception to this includes Smalltalk. It allows all data types to be extended within a program. It allows the programmer to modify the operations that can be performed on them and also to redefine the built in operations.

Do not have utility functions associated with them

Most of the object oriented languages have certain utility functions associated with non primitive data types that eases their use as compared to primitive data types that do not have such functions.

Example:

The complex data types like arrays have functions associated with them.

array.length();

However, this is not the case in pure objected oriented languages like Ruby, which provide a utility functions to primitive types as well.

Conclusion

Primitive objects that are provided in object oriented programming languages have their own advantages and disadvantages. Though using primitive data types in most languages reduces the degree of customization possible, they provide a better performance and ease of use to compensate. The extent to which a primitive data type can be used efficiently depends on the application and the programming language used for the application.

References

1. http://ece.uprm.edu/~ahchinaei/courses/2010jan/icom4036/slides/11icom4036DataTypes1.pdf
2. http://en.wikipedia.org/wiki/Primitive_data_type
3. http://en.wikipedia.org/wiki/Fixed-point_arithmetic
4. http://en.wikipedia.org/wiki/C%2B%2B
5. http://en.cppreference.com/w/cpp/language/types
6. http://en.wikipedia.org/wiki/Wide_character
7. http://msdn.microsoft.com/en-us/library/e02ya398(v=vs.80).aspx
8. http://sparkcharts.sparknotes.com/cs/cplusplus/section2.php
9. http://docs.oracle.com/javase/tutorial/java/nutsandbolts/datatypes.html
10. http://msdn.microsoft.com/en-us/library/ms228360(v=vs.80).aspx
11. http://www.cse.iitb.ac.in/~cs701/old/beamer/Closures.pdf
12. http://ruby-doc.org/docs/ProgrammingRuby/html/builtins.html
13. http://ruby-doc.org/docs/ProgrammingRuby/html/ref_c_integer.html
14. http://ruby-doc.org/docs/ProgrammingRuby/html/ref_c_bignum.html
15. http://ruby-doc.org/docs/ProgrammingRuby/html/ref_c_string.html
16. http://ruby-doc.org/docs/ProgrammingRuby/html/ref_c_float.html
17. http://ruby-doc.org/docs/ProgrammingRuby/html/ref_c_trueclass.html
18. http://ruby-doc.org/docs/ProgrammingRuby/html/ref_c_falseclass.html
19. http://today.java.net/pub/a/today/2005/03/24/autoboxing.html#performance%5Fissue