3.7 - Interpreter Pattern : Behavioral Design Patterns
The Interpreter Pattern is a behavioral design pattern that defines a grammatical representation for a language and provides an interpreter to evaluate sentences in that language. This pattern is used to evaluate simple expressions or sentences that belong to a particular language or grammar. The idea is to build a hierarchical structure of expressions, where each expression implements a common interface.
Key Components of the Interpreter Pattern:
Abstract Expression: Declares the
interpret()
method common to all expressions.Terminal Expression: Implements the
interpret()
method to evaluate individual parts of the language, such as numbers or constants.Non-terminal Expression: Represents operators (such as addition or subtraction) that combine terminal expressions. It also implements the
interpret()
method.Client: Builds the syntax tree representing the sentence to be interpreted and evaluates the sentence by calling
interpret()
on the root expression.
The pattern is typically used in situations where a grammar defines how expressions are evaluated, such as mathematical expressions, programming language interpreters, or Boolean logic evaluators.
Code Example
Here’s a practical example of the Interpreter Pattern using modified class and function names for better clarity.
1. Abstract Expression
The MathExpression
interface declares the evaluate()
method for all expressions.
// Abstract Expression
interface MathExpression {
int evaluate();
}
2. Terminal Expression
The Constant
class is a terminal expression that represents a number. It implements the evaluate()
method, which simply returns the value of the number.
// Terminal Expression
class Constant implements MathExpression {
private int value;
public Constant(int value) {
this.value = value;
}
@Override
public int evaluate() {
return value;
}
}
3. Non-terminal Expression
The Addition
class is a non-terminal expression that represents the addition operation. It takes two MathExpression
objects (which could be either constants or other expressions) and adds their results.
// Non-terminal Expression
class Addition implements MathExpression {
private MathExpression leftExpression;
private MathExpression rightExpression;
public Addition(MathExpression leftExpression, MathExpression rightExpression) {
this.leftExpression = leftExpression;
this.rightExpression = rightExpression;
}
@Override
public int evaluate() {
return leftExpression.evaluate() + rightExpression.evaluate();
}
}
4. Client
The InterpreterDemo
class demonstrates how to build an expression tree and interpret a mathematical expression.
// Client
public class InterpreterDemo {
public static void main(String[] args) {
// (10 + (2 + 3))
MathExpression expression = new Addition(new Constant(10), new Addition(new Constant(2), new Constant(3)));
int result = expression.evaluate();
System.out.println("Result: " + result); // Output: Result: 15
}
}
Explanation:
MathExpression Interface: This interface defines the
evaluate()
method, which all concrete expressions (like numbers or operators) must implement.Constant: The
Constant
class is a terminal expression representing a number. Itsevaluate()
method returns the value of the number stored in the instance.Addition: The
Addition
class is a non-terminal expression that takes two expressions as its operands. When itsevaluate()
method is called, it evaluates the left and right operands and returns their sum.InterpreterDemo (Client): This class constructs an expression tree and interprets the mathematical expression by calling
evaluate()
on the root expression. In this case, the tree represents the expression10 + (2 + 3)
.
Benefits of the Interpreter Pattern:
Easy to Extend: You can easily extend the language by adding new non-terminal expressions (e.g., subtraction, multiplication) without modifying the existing structure.
Modular Structure: The pattern encourages the use of a modular and hierarchical structure, making it easy to understand and maintain.
Reusable Expressions: Common expressions (like numbers or basic operations) can be reused in multiple parts of the syntax tree.
Drawbacks of the Interpreter Pattern:
Complexity: For complex grammars or expressions, this pattern can lead to a large number of classes, making the code harder to maintain.
Performance: The recursive nature of the pattern (where expressions call
interpret()
on their children) can lead to performance issues if the expression tree becomes too large or deep.
Real-World Use Cases:
Mathematical Expression Evaluators: Many calculators and tools use the interpreter pattern to evaluate mathematical expressions like
2 + 3 * (5 - 1)
.Query Interpreters: In systems like SQL engines or search engines, queries are often interpreted using the interpreter pattern.
Configuration Languages: Systems that need to parse and interpret configuration languages, like JSON or XML, can use this pattern.
Compilers and Interpreters: Programming language interpreters use this pattern to evaluate or compile code.
Example with More Operations:
We can easily extend the current example to handle subtraction, multiplication, or division by creating new non-terminal expression classes.
// Non-terminal Expression for Subtraction
class Subtraction implements MathExpression {
private MathExpression leftExpression;
private MathExpression rightExpression;
public Subtraction(MathExpression leftExpression, MathExpression rightExpression) {
this.leftExpression = leftExpression;
this.rightExpression = rightExpression;
}
@Override
public int evaluate() {
return leftExpression.evaluate() - rightExpression.evaluate();
}
}
Now, we can combine different operations:
public class InterpreterDemo {
public static void main(String[] args) {
// (10 + (2 - 3))
MathExpression expression = new Addition(new Constant(10), new Subtraction(new Constant(2), new Constant(3)));
int result = expression.evaluate();
System.out.println("Result: " + result); // Output: Result: 9
}
}
Conclusion:
The Interpreter Pattern provides a structured way to evaluate and interpret expressions or sentences in a specific language or grammar. By defining individual expression types (like constants and operations), this pattern makes it easy to evaluate complex expressions while keeping the structure modular and extensible. Although it might not be the most efficient pattern for large or complex systems, it works well for simple grammars and scenarios where expressions need to be evaluated at runtime.
Subscribe to my newsletter
Read articles from Venu Madhav Emmadi directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by