feliciousx / static-code-analyzer Goto Github PK

This repository contains the code used to analyze C++ code for an assignment.

The repository being analyzed was nodejs

1. Clone nodejs
2. cd to the nodejs repository
3. Copy my static-code-analyzer files here
4. Run git for-each-ref --sort='*authordate' --format='%(tag)' refs/tags > versions to get all the releases sorted by date into a file called versions.
5. Run git for-each-ref --sort='*authordate' --format='%(tag) %(*authordate:iso8601)' refs/tags > nodejs-dates to get all the releases with the date as well.
6. Run analyze.sh versions which takes the first argument as the file. It will read the file and go through everyline while git checkout to that specific version in the nodejs repository and run cccc static code analyzer on it, creating folders in analyze/%version for every checkout.
7. Move all the data to another folder for analyzing (including the static-code-analyzer files.
8. Run node main.js versions > data.csv that will parse the XML data from the analyze folders into a .csv format
9. Run node date.js nodejs-dates > age.csv that will sort the date by date with the age as well.
10. Use R to plot graph using data.csv and age.csv

TODO:-

• Write the R code for plotting graph

Data obtained is the total of:-

• Number of modules
• Lines of code
• Cyclomatic complexity
• Number of methods
• Number of private methods

• nodejs
• npm