BooksTools

Code Metrics and Dependencies for Eclipse

Frank Sauer created the fancy Eclipse plugin simply called Metrics.

With this plugin you can solve two QA tasks at once…

  1. (OO) Code Metrics such as McCabe cyclomatic complexity and the dependency metrics by Robert Martin.

    • Others include most common LOC (line-of-code), NOI (number of interfaces), PAR (number of parameters) and other common software-engineering metrics.


    • Metrics can now trigger warnings that show up in the task view as well as the editors, indicating methods and types for which metrics safe ranges are being violated
    • Metrics can be exported to an XML file to be processed with XSL into any kind of report you want.
    • Metrics can even be generated in a nightly build in ANT
    • the author recommends these books as basis and was inspired himself by them




      with metrics like LOC and McCabe Cyclomatic Complexity

      - that counts the number of flows through a piece of code. Each time a branch occurs (if, for, while, do, case, catch and the ?: ternary operator, as well as the && and || conditional logic operators in expressions) this metric is incremented by one. See also


    • and



      with metrics like Efferent Coupling (Ce) – The number of classes inside a package that depend on classes outside the package.


  2. a graphical dependency analysis tool based on the dynamic hyperbolic package TouchGraph

    • graph can be zoomed and rotated
    • a Tangle popup can be used to see why all nodes are connected to each other and how and a “Find Shortest Path” feature

Download it at the sourceforge Metrics repository

Average rating
(0 votes)

Comments

Code Metrics and Dependencies for Eclipse

Measuring quality is a good idea. And most importantly: define a reasonable scale. It's one thing to know you're driving your car with 50 km/h - but it does make a difference whether you're in a 30 km/h zone or on the Autobahn.

Lesson learned: metrics are useful within a defined context or ->standard. And once you collect metrics, you'll want to make sure they stay within a certain range (search Google on "Statistical Quality Assurance" or email me :) ) and whether your organization makes improvements over time, etc, etc.

For Java projects, you get all this and much more - automatically. Quality at the source: www.qido.at

Similar entries