Master: Release 3.4:

Note: the following instructions are for the unstable development version. To install stable releases, please see the readme file for those releases:

Most of the (python) prerequisites for AmCAT are automatically installed using pip (see below). To install the non-python requirements, you can use the following (on Ubuntu 15.10 or 16.04):

sudo apt-get install antiword unrtf rabbitmq-server python3-pip postgresql postgresql-contrib python3-venv git postgresql-server-dev-9.4 python3-dev libxml2-dev libxslt-dev graphviz pspp

If you want to compile lxml and psycopg2 yourself (through pip), you need to install:

sudo apt-get build-dep python3-psycopg2 python3-lxml

You can avoid compiling libraries by installing some dependencies through apt:

sudo apt-get install python3-lxml python3-amqplib python3-psycopg2 python3-requests python3-pygments

It is probably best to install AmCAT in a virtual environment. Run the following commands to setup and activate a virtual environment for AmCAT: (on ubuntu)

pyvenv env
source env/bin/activate

If you use a virtual environment, every time you start working with AmCAT you need to repeat the source line to load the environment. If you don't use a virtual environment, you will need to run most pip command below using sudo.

AmCAT requires a database to store its documents in. The default settings look for a postgres database 'amcat' on localhost. To set up the current user as a superuser in postgres and create the database, use:

sudo -u postgres createuser -s $USER
createdb amcat

AmCAT uses elasticsearch for searching articles. Elasticsearch is provided as a debian package, but it does need some extra plugins to be ready for AmCAT.

wget https://download.elasticsearch.org/elasticsearch/release/org/elasticsearch/distribution/deb/elasticsearch/2.1.1/elasticsearch-2.1.1.deb
sudo dpkg -i elasticsearch-2.1.1.deb
rm elasticsearch-2.1.1.deb

# Install plugins
cd /usr/share/elasticsearch
sudo bin/plugin install mobz/elasticsearch-head
sudo bin/plugin install analysis-icu
sudo bin/plugin install amcat/hitcount

# Enable hitcount as default similarity provider, and enable groovy scripting
cat <<EOT | sudo tee --append /etc/elasticsearch/elasticsearch.yml
index.similarity.default.type: hitcountsimilarity
script.inline: on
script.update: on
EOT

# Restart elastic
sudo systemctl stop elasticsearch
sudo systemctl start elasticsearch

Warning: We enabled a non-sandboxed scripting language (Groovy). Make sure to restrict access to the elastic instance by untrusted parties, as this allows executing arbitrary code as the user elasticsearch.

Now you are ready to install AmCAT. The easiest way to do this is to pip install it direct from github. This is not advised unless you use a virtual environment.

pip install git+https://github.com/amcat/amcat.git

Clone the project from github and pip install the requirements.

git clone -b release-3.4 https://github.com/amcat/amcat.git
pip install -r amcat/requirements.txt

Be sure to add the new directory to the pythonpath and add AMCAT_ES_LEGACY hash to the environment. If you add these lines to amcat-env/bin/activate they will be automatically set when you activate.

export PYTHONPATH=$PYTHONPATH:$HOME/amcat
export AMCAT_ES_LEGACY_HASH=N

AmCAT uses bower to install javascript/CSS libraries. On Ubuntu, you need to install the legacy version of nodejs first, and then install bower by using npm:

sudo apt-get install nodejs-legacy npm
sudo npm install -g bower

Then, in the top-directory of AmCAT itself run:

bower install
cp static/components/highcharts/modules/{heatmap,heatmap-fixed}.js

Whichever way you installed AmCAT, you need to call the migrate command to populate the database and set the elasticsearch mapping:

python -m amcat.manage migrate

You can create a superuser by running:

python -m amcat.manage createsuperuser

For debugging, it is easiest to start amcat using runserver:

python -m amcat.manage runserver

Finally, to use the query screen you need to start a celery worker. In a new terminal, type:

DJANGO_SETTINGS_MODULE=settings celery -A amcat.amcatcelery worker -l info -Q amcat

(if you are using a virtual environment, make sure to activate that first)

The main configuration parameters for AmCAT reside in the settings folder. In many places, these settings are defaults that can be overridden with environment variables.