Skip to content
Codeface is a framework for analysing technical and social aspects of software development
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
codeface
conf
datamodel
docs Move non-essential documentation to a separate folder Jun 25, 2016
experiments
id_service
integration-scripts
log
patches
performance
.Rprofile
.gitattributes
.gitignore
.travis.yml
COPYING
README.md
Vagrantfile
analysis_example.sh
codeface.conf
codeface_testing.conf
nginx.conf
nginx.shinyserver.conf
packages.r
python_requirements.txt
setup.py
shiny-server-pack.sh
shiny-server.config
shiny-server.sh
test_copyright.sh
test_graphs.sh
timeseries.txt
understand.txt

README.md

Installing and using Codeface

Installing Codeface

The recommended way to set up a Codeface instance is via vagrant. Clone the repository and run

vagrant up

to obtain a fully provisioned Codeface machine. Vagrant defaults to Virtualbox as provider, which may cause large performance impacts especially for I/O heavy tasks. You can /alternatively/ use

vagrant up --provider=lxc

if you have the corresponding LXC provider for vagrant installed on your system. To get shell access on the machine in each case, use

vagrant ssh

If vagrant is not yet installed on your system, please consult the corresponding [wiki page] (https://github.com/siemens/codeface/wiki/Runnning-codeface-with-Vagrant).

Analysing Projects

Concept

Conceptually, work with Codeface is split in two stages:

  1. Analyse projects using the batch-mode command line interface. See analysis.md for further details. Note that this process involves substantial amounts of git repo querying and data crunching, and can require several hours for large projects like the Linux kernel.
  2. Inspect the results by visual analysis with the web frontend (see webserver.md for setup details), or by querying the database directly. See file codeface/R/interactive.R for exemplary instructions.

Five Easy Steps to your First Analysis

To perform an analysis of project qemu (a machine emulation software) and inspect the results in the interactive web frontend, run the following steps:

  1. After bringing up the vagrant instance, vagrant ssh into the virtual machine
  2. Start the ID service with /vagrant/id_service/start_id_service.sh&
  3. Run an analysis of qemu with /vagrant/analysis_example.sh (this process may take a while to complete)
  4. Start the webserver with cd vagrant; ./shiny-server.sh
  5. Point your webserver on the host at http://localhost:8081
You can’t perform that action at this time.