This is a placeholder readme file aiming to provide a quick starting point to anyone getting their hands dirty with ecomp.
The readme tries to assume no knowledge of using a Ruby project (gems & bundler etc.) and is likely to be incorrect!
To run the ecomp tools, Ruby 2.1.2 must be available along with RubyGems and Bundler. Java analysis requires a JVM. JavaScript analysis requires nodejs. Ruby analysis requires no other tools.
Ensure you have Ruby 2.1.2 installed and RubyGems and Bundler, then in the directory you checked ecomp out to, run:
bundle
This will download and install the required gems that the tool requires to run.
In the directory you checked out ecomp to, run:
export PATH=$PATH:`pwd`/bin
You should now be able to execute the 'metrics' tool.
The metrics tool requires two arguments, a report output directory (where JSON files will be created) and a file glob string (e.g. **/*.java).
The following examples try to highlight typical usage of ecomp.
Ensuring that run_ecomp.sh is available in your PATH, complexity of a project can be calculated by issueing the following command:
run_ecomp.sh SVN_REPOSITORY PROJECT_NAME '**/*.SOURCE_EXTENSION'
A concrete example could be:
run_ecomp.sh http://myrepo.org/myjavaproject/trunk myjavaproject '**/*.java'
The run_ecomp.sh script will then use git svn to clone the repository to the myjavaproject, correcting the author names as it does so and will then use the metrics tool to start analysing the source of the project.
Once the complexity calculation is complete, the JSON files will then be made available to the public/data/myjavaproject directory, allowing for visualisation of the data to be accessible at http://a.sinatra.host:9292/#myjavaproject.
NOTE: The process is time consuming and involves going through all the previous commits to calculate a play-by-play dataset of complexity changes to each of the source files.
Documenting how to render the datasets so that the data may be visualised.