Preliminary Lessons from a Software Evolution Analysis of Moodle

by Héctor J. Macho (1) and Gregorio Robles (2)

Submitted to the ISELEAR'13

(1) Universidad Rey Juan Carlos (Madrid, Spain); (2) Universidad Rey Juan Carlos (Madrid, Spain)

Methodology

Data Source

We have obtain the sources from the GitHub mirror of the official Moodle repository (https://github.com/moodle/moodle.git). You can obtain the sources from the official Moodle repository too (http://git.moodle.org/gw?p=moodle.git).

You can download the tar.gz used for this paper at these links:

Scripts

In order to obtain the information we have used several scripts.

• LOC (Lines of Code):

  • script_loc.sh (you have to install cloc, ohcount and sloccount previously)

• Similarity:

These scripts would be in the same directory of the tar.gz. You have to execute manytaras.py in order to obtain the sentences that you have to execute to obtain the results (you can redirect standar output of the script to a file and the execute the file as a shell script).

• manytaras.py

• tara.py

• Cyclomatic Complexity:

  We have used PHP Depend to calculate the cyclomatic complexity of Moodle, so you have to install if you want to use the script. Given that only a part of the code of Moodle is written in PHP (the most of it) we aren't going to obtain the total cyclomatic complexity.

  • analyse_PHP.py

• VizGrimoire:

  You need VizGrimoire to obtain the metrics of the repository. We have modified the vg-github.py released by Bitergia under GPLv3.

  • vg-github_modified.py

Results

• LOC (Lines of Code):

  • loc_cloc.txt

  • loc_sloccount.txt

  • loc_ohcount.txt

  • loc_results.ods

• Similarity:

• similarity_results.txt

• similarity_results.ods

• Cyclomatic Complexity:

  • charts.tar.gz

  • xml.tar.gz

  • results_cyclo.ods

• VizGrimoire:

  • scm.ods