ck
A Java library for calculating class-level and method-level code metrics via static analysis, without requiring compiled code.
What is ck?
CK is a Java library and command-line tool that calculates a wide range of code metrics through static analysis of source code. It helps developers and researchers measure software quality attributes like coupling, cohesion, complexity, and size without needing to compile the code.
Java developers, software engineering researchers, and teams focused on code quality analysis, refactoring, and technical debt management.
CK provides a comprehensive, extensible, and accurate metric suite using pure static analysis, avoiding the overhead and dependencies of bytecode analysis, and is easily integrable into build pipelines or research workflows.
Overview
Code metrics for Java code by means of static analysis
Use Cases
Best For
- Analyzing code quality and technical debt in Java projects
- Research studies on software metrics and their correlations
- Integrating code metric collection into CI/CD pipelines
- Identifying refactoring candidates based on coupling and cohesion
- Educational purposes for teaching software measurement
- Building custom static analysis tools on top of its extensible framework
Not Ideal For
- Teams requiring real-time, interactive code quality feedback within an IDE, as CK operates via command-line or library integration without live analysis features.
- Projects in languages other than Java, since CK is specifically designed for static analysis of Java source code only.
- Environments with strict memory constraints, as CK's AST parsing can be memory-intensive for large codebases, requiring tuning of 'max files per partition'.
- Scenarios where bytecode-level accuracy is essential, as CK's static analysis may miss dynamic behaviors or fail on unresolved method overloading.
Pros & Cons
Pros
Calculates over 30 metrics including CBO, DIT, LCOM, and unique word counts, providing extensive insights into code quality without needing compilation, as detailed in the README.
Analyzes Java source code directly using Eclipse JDT, eliminating the dependency on compiled bytecode and simplifying setup, as it operates solely on .java files.
Uses Visitor and Factory patterns, allowing developers to easily add custom metrics by implementing ClassLevelMetric or MethodLevelMetric interfaces, making it adaptable for research or custom tools.
Can be run as a standalone command-line tool or integrated as a Maven dependency, catering to both ad-hoc analysis and automated CI/CD pipelines, with examples provided in the README.
Cons
The README notes that users may need to tune 'max files per partition' to avoid out-of-memory errors, indicating scalability issues with large projects and complex AST parsing.
As admitted in the README, some metrics like LCOM are not reliable, and others like RFC can fail in cases of method overloading, limiting accuracy for certain code patterns.
CK is limited to Java projects, so it cannot be used for multi-language codebases or other programming languages, restricting its applicability in polyglot environments.
Using JAR files for better type resolution requires manual configuration, and the tool may not handle all dependency scenarios seamlessly, as noted in the 'use jars' parameter description.
Frequently Asked Questions
Related Projects
PHP ParserA PHP parser written in PHP
TypeScript ESLint:sparkles: Monorepo for all the tooling which enables ESLint to support TypeScript
pyrightStatic Type Checker for Python
Reviewdog🐶 Automated code review tool integrated with any code analysis tools regardless of programming language
Found a gem we're missing?
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.