Law Climate Change Toolkit

A web application which stands as a toolkit for climate change law assessment.

Prerequisites

Installing the application

Get the source code:

 git clone https://github.com/eaudeweb/lcc-toolkit
 cd lcc-toolkit

Customize the environment files:

 cp docker/postgres.env.example docker/postgres.env
 cp docker/web.env.example docker/web.env
 cp docker/init.sql.example docker/init.sql

Depending on the installation mode, create the docker-compose.override.yml file:

 cp docker-compose.override.[prod|dev].yml docker-compose.override.yml

Start the application stack:

 docker-compose up -d
 docker-compose logs

Attach to the web service:

 docker-compose run web

Create a superuser (for Ansible see https://gist.github.com/elleryq/9c70e08b1b2cecc636d6)

 python manage.py createsuperuser

That's it. If you installed in production mode, you should be able to access the app at http://localhost:8000. If you installed in development mode, follow the next steps.

Development setup

Make sure you have the following settings configured correctly (these are the default values in the .example files, so you shouldn't have to change anything).

  • DEBUG=on in web.env file.

The docker-compose.override.yml file's web section includes this:

build:
  context: .
  args:
    REQFILE: requirements-dev.txt
entrypoint: bash

To create and run the develop stack: docker-compose up -d docker-compose ps docker exec -it lcct.web bash

and run the following:

python manage.py migrate
python manage.py load_fixtures
python manage.py createsuperuser
python manage.py search_index --rebuild
python manage.py runserver 0.0.0.0:8000

Now you should be able to access the app in development mode at http://localhost:8000 By default, there will be no Legislation in the database. In order to generate some fake legislation for testing/debugging purposes, you can run the following command:

python manage.py generate_fake_legislation [N]

Where [N] is the number of Legislation objects to generate.

If you need to make front-end changes, make sure to run:

grunt dev

Testing

Allow the user to create a database: docker exec -it lcct.db bash psql -U postgres ALTER USER demo CREATEDB;

To execute the test suite, attach to the web service and run the following:

python manage.py test

Configuration variables

The application expects configuration via environment variables:

DEBUG Turns on debugging behaviour if set to on. Not secure for use in production.

SECRET_KEY Random secret used for Django browser sessions.

DATABASE_URL Django database connector.

ALLOWED_HOSTS A list of host/domain names that this Django site can serve.

SENTRY_DSN, SENTRY_PUBLIC_DSN URL of Sentry server to report errors. Optional.

GA_TRACKING_ID Google Analytics tracking code. Optional.

POSTGRES_PASSWORD PostgreSQL superuser password.

NODE_ENV Define different environments for nodejs. Possible values are: prod, dev.

Migrating data to a different server

To create a Postgresql dump directly from the running lcct.db container, the following command should be used:

docker exec lcct.db pg_dump -U lcct lcct > <backup-file-name>

The -U lcct part refers to the database user, while the last lcct in the docker exec command refers to the database name.

To restore the backed-up data to a fresh install of the lcc-toolkit stack, the following command should be used:

docker exec lcct.db psql -U lcct lcct < <backup-file-name>The psql parameters are the same as those used for the above pg_dump command.

Migrating Elastic Search from 5.4 to Elastic Search 7.14:

  1. For keeping the volume, you will first need to update to Elastic Search 6.8. The first step is to use the followin elastic search image: docker.elastic.co/elasticsearch/elasticsearch:6.8.18
  2. Add Kibana in the stack: docker.elastic.co/kibana/kibana:6.8.18
  3. Install python packages for Elastic Search version 6.8.18
pip install django-elasticsearch-dsl==6.5.0
pip install elasticsearch-dsl==6.4.0
pip install elasticsearch==6.8.2
  1. Reindex the data
python manage.py search_index --rebuild
  1. Use Kibana Upgrade Assistant to reindex the other nodes that weren't indexed by the python package
  2. Use Kibana console to upgrade the watches index: POST /_xpack/migration/upgrade/.watches
  3. Set elasticsearch to image docker.elastic.co/elasticsearch/elasticsearch:7.14.1 and add the following variable to the environment: discovery.type=single-node
  4. Install the python packages required for ElasticSearch 7.4:
pip install django-elasticsearch-dsl==7.2.0
pip install elasticsearch-dsl==7.3.0
pip install elasticsearch==7.13.3
  1. Reindex one last time from the python shell to check if everything is working as expected:
python manage.py search_index --rebuild

Generate documentation

The application documentation is generated from Sphinx. The following commands should be run on production every time the documentation is changed:

cd docs/
make html
GitHub - eaudeweb/lcc-toolkit: A web application which stands as a toolkit for climate change law assessment.
A web application which stands as a toolkit for climate change law assessment. - GitHub - eaudeweb/lcc-toolkit: A web application which stands as a toolkit for climate change law assessment.
GitHubeaudeweb