Colab, a Software for Communities

What is Colab?

Application that integrates existing systems to represent the contributions of the members through:

• The amendments to the Wiki trac system.
• Changes to the trac system code.
• Discussions at the mailman list.
• And other systems in the community.

Features

• Developed by Interlegis Communities http://colab.interlegis.leg.br/
• Written with Python http://python.org/
• Built in Django Web Framework https://www.djangoproject.com/
• Search engine with Solr https://lucene.apache.org/solr/

Colab and Solr

This software uses Apache Solr as search platform based on Apache Lucene.

With Solr generates the REST style API with which you can make HTTP requests to get results: natively in XML or JSON, PHP, Ruby and Python and then treatment.

Installation (Development Environment)

A detailed installation steps can be found in:

Detailed full environment installation

Here we'll cover how to setup a development environment using a Vagrant virtual machine.

Before getting started you should install the following softwares:

• Vagrant (tested with version 1.2.7)
• Virtualbox (version >= 4.0)
• fabric (tested with version 1.7.0)
• Git

Getting started with the Virtual Machine

First you will need to clone the repository:

git clone [email protected]:colab-community/colab.git

NOTE:

Here we are assuming you have ssh permissions to clone the repo using ssh. If not fork it and clone your own fork (or use https instead of ssh).

Enter in the repository you've just cloned. To start working all you need is to turn the virtual machine on with the command:

vagrant up

NOTE:

BE PATIENT!

This will take a while. The vagrant up will download a full vm (virtualbox) running a Ubuntu 12.04 64bits. After the vm is up and running the command will also configure it (using puppet) and that will also take a bit.

Running Colab

Now that you have a vm running we have two options to run Colab:

• Django development server (runserver)
• Gunicorn + supervisor + Nginx

Django development server (runserver)

This option is advised for developers working in new features for Colab. The code used to run Colab will be the same code placed on your machine, that means that if you change the code in your own computer the code on the vm will also change.

Make sure you have a local_settings.py file placed in your repository. It should be located in src/colab/.

To get started you can copy the example file as follow:

cp src/colab/local_settings-dev.py src/colab/local_settings.py

Now we are ready to run:

fab runserver

Note

As this is the first time you run this command it will install all requirements from requirements.txt into a virtualenv. To update those requirements you should run fab runserver:update.

The fab runserver command will open the django builtin development server on the port 7000 but due to vagrant magic you will be able to access it on http://localhost:8000/.

Gunicorn + supervisor + Nginx

This option will run Colab in a way very similar to the production environment. This should be used to test puppet manifests and also the configuration of each one of the services running.

First of all we need to clone the repo and configure your local_settings.py. That is done by calling the command:

fab install:path/to/your/local_settings.py

Now you need to deploy the code using the command:

fab deploy

For the next deploy you can just run fab deploy and in case your requirements.txt changes fab deploy:update.

The deployed code will be accessible on http://localhost:8080.