J-MIndex (Java Maintainability Index Analyzer)

A professional command-line tool for analyzing Java source code to calculate its Maintainability Index (M-Index) at multiple levels: method, class, package, and project. The tool uses JavaParser to parse source files and computes metrics such as Halstead Volume, Cyclomatic Complexity, and Lines of Code (LOC), then calculates a normalized Maintainability Index (0–100).

Features

• Analyze all .java files in a directory (recursively, excluding test folders)
• Calculate Maintainability Index (M-Index) for each method
• Aggregate and display M-Index at class, package, and project levels
• Export results to CSV (method, class, package, project summary)
• Uses normalized MI formula (0–100)
• Skips files in any test directory
• Extensible, testable, and ready for research or production

Requirements

• Java 11 or higher
• Maven

Build

mvn clean package

• The output JAR will be in target/jmi-<version>.jar (e.g., jmi-0.0.1.jar)

Run

Analyze and print results:

java -jar target/jmi-0.0.1.jar -project <source-directory>

Export results to CSV:

java -jar target/jmi-0.0.1.jar -project <source-directory> -out result.csv

• Replace <source-directory> with the path to your Java source code folder.

Example Output (Console)

--- Maintainability Index Results ---
Class: ExampleClass
  Method: foo | MI: 85.23 | CC: 2 | HV: 23.45 | LOC: 10
  [Class Avg MI: 85.23]

[Package Avg MI]
  Package: com.example | Avg MI: 85.23

[Project Avg MI: 85.23]

Example Output (CSV)

Class,Method,MI,CyclomaticComplexity,HalsteadVolume,LOC
ExampleClass,foo,85.23,2,23.45,10
...

Package,AvgMI
com.example,85.23

ProjectAvgMI,AvgMI
ProjectAvgMI,85.23

Metrics Used

• Maintainability Index (MI):
  • MI = MAX(0, (171 - 5.2 * ln(Halstead Volume) - 0.23 * Cyclomatic Complexity - 16.2 * ln(LOC)) * 100 / 171)
  • Normalized to 0–100 (higher is better)
• Cyclomatic Complexity (CC):
  • McCabe's formula: CC = E - N + 2P (E=edges, N=nodes, P=connected components)
• Halstead Volume (HV):
  • V = N * log2(n) (N=total operators+operands, n=distinct operators+operands)
• Lines of Code (LOC):
  • Number of lines in each method

How to Test

• Run mvn test to execute unit tests for metric calculations
• Try with open source Java projects and compare with SonarQube, JHawk, or Understand for validation

Extending/Customizing

• Add new exporters (e.g., JSON) in ResultExporter
• Add new metrics in MetricCalculator
• Adjust exclusion logic for other folders if needed

License

See LICENSE