DHIS2-IMIS-ETL-Portal

IMIS public data portal using Python, Flask, Jinja templates, Bulma CSS, Docker, and Highcharts.

Installation

• Install on a VM running Ubuntu 20.04 (512MB RAM, 1CPU, 10GB is fine)
• Make sure only Web Ports are open (SSH, HTTP, HTTPS)
• On VM, install Docker

ssh <YOUR.IP.ADDRESS>
git clone <this-repo>
cd <this-repo>

then set up permissions and test with hello-world

sudo groupadd docker
sudo usermod -aG docker $USER
newgrp docker
docker run hello-world

then create and fill out a .env file

DHIS2BASEURL=
DHIS2USERNAME=
DHIS2PASSWORD=

and chmod 0600 .env to restrict access

Warning Keep the .env file in a secure location

build the image and run it

docker image build -t imis_portal . 
docker run -it -p 80:80 -d imis_portal

This will automatically read the .env file, start the container, and returns a container hash, e.g. 1efc3c3523dc12....

To watch logs:

docker logs <containerhash> -f

... then you should see the website running at http://YOUR.IP.ADDRESS (or at http://yourwebsite.com)

Development

Requirements: Docker, Python 3

Create .env file

DHIS2BASEURL=
DHIS2USERNAME=
DHIS2PASSWORD=

Install and run:

python3 -m virtualenv .venv
source .venv/bin/activate
pip install -r requirements.txt
python main.py

To download data:

python background.py

Adding more charts

1. Create a new chart in DHIS2
2. Make sure the user DHIS2USERNAME has access to the chart
3. Identify the /api/analytics DHIS2 API request used to fetch data for that chart
4. Add it to charts in background.py
5. Find an equivalent visualization at highcharts.com
6. Add it to the templates, e.g. in index.html and possibly custom CSS into the static folder
7. Make it dynamic by parsing the data downloaded via background.py into the data folder and pass it to the template.