User-Defined Types (Cambridge (CIE) A Level Computer Science): Revision Note

What are user-defined data types?

• A user-defined data type is a custom data type created by the programmer, built from existing types to match the specific needs of a program

• It is built using primitive data types such as INTEGER, STRING, BOOLEAN, etc

Why are user-defined types necessary?

• Primitive types alone cannot model many real-world entities (e.g. a customer, a date, a linked list)

• User-defined types let you:

  • Group related data under a single identifier (record)

  • Restrict a variable to a fixed set of valid values (enumerated)

  • Build dynamic data structures such as linked lists (pointer)

  • Make code more readable, reusable and easier to validate

Non-composite

• A non-composite data type is defined without referencing any other data type

• It can be a primitive type available in the language, or a user-defined type (enumerated or pointer)

What is an enumerated data type?

• An enumerated data type is a non-composite, user-defined type consisting of a list of all possible values the variable can take

• The values are ordered: they have an implied sequence based on the order they are declared, so comparisons work (e.g. Monday comes before Tuesday)

• Use an enumerated type to restrict a variable to a known, fixed list, improving readability and reducing invalid values

Example code

TYPE DayType = (Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, Sunday)

DECLARE Today : DayType
Today ← Monday

IF Today = Saturday OR Today = Sunday THEN
    OUTPUT "It's the weekend!"
ELSE
    OUTPUT "It's a weekday."
ENDIF
 

• Today can only hold one of the seven ordered values in DayType

What is a pointer data type?

• A pointer is a non-composite, user-defined type that stores a memory address only, rather than the data itself

• Its definition also indicates the type of data stored at that memory location

• Pointers are used to build dynamic data structures such as linked lists

• The ^ shows the type is a pointer

• <data type> is the type of data stored at the address

Example code

TYPE Node
    DECLARE Data : INTEGER
    DECLARE Next : ^Node
ENDTYPE

DECLARE StartPtr : ^Node
DECLARE NewNode : ^Node

NEW(NewNode)          // allocate memory for a new node
NewNode^.Data ← 10    // ^ accesses the contents at the address
NewNode^.Next ← NULL  // NULL marks the end of the list
StartPtr ← NewNode
 

Composite

• A composite data type is a collection of data made up of multiple elements, of the same or different data types, grouped under a single identifier

• It references other data types in its definition (primitive or other user-defined types)

• The three composite user-defined types you need are record, set and class

What is a record?

• A record is a composite type that groups together a fixed number of related items (fields), which may be of different data types, to form a single new type

• Access a field using dot notation: RecordName.FieldName

TYPE <RecordName>    
  DECLARE <Field1> : <DataType1>    
  DECLARE <Field2> : <DataType2>
ENDTYPE

Example code

TYPE Order
    DECLARE AccountNumber : INTEGER
    DECLARE OrderNumber   : INTEGER
    DECLARE OrderPrice    : REAL
    DECLARE OrderDate     : DATE
ENDTYPE

DECLARE ThisOrder : Order
ThisOrder.AccountNumber ← 4521
ThisOrder.OrderPrice ← 19.99
 

What is a set?

• A set is a composite type holding a list of unordered elements

• All elements must be of the same data type, and the set definition includes the data type it uses

• Set theory operations such as union and intersection can be applied to the elements

TYPE <identifier> = SET OF <data type>
DEFINE <SetName> (value1, value2, … ) : <identifier>

Example code

TYPE SymbolSet = SET OF CHAR
DEFINE Operators ('+', '-', '*', '/', '^') : SymbolSet

IF '+' ISIN Operators THEN
    OUTPUT "Plus is in the set"
ENDIF
 

What is a class?

• A class is a composite type that groups together both data (attributes) and the methods that operate on that data, under a single identifier

• It is used as a template to create objects in object-oriented programming

• Full detail on defining and using classes is covered in Object-Oriented Programming

Choosing a user-defined data type

• Exam questions may ask you to choose and design an appropriate type for a problem

• Match the type to what the data looks like: