ua-2022-map

Map of reception points for refugees near Polish-Ukrainian border

work in progress...

Development

We provide a docker-based dev environment; to spin up the system, run

make
make up
make down

For development and iterating on the code, get into a shell in the container and re-run your scripts from there with

make sh

From inside the shell you can then execute your scripts and iterate, e.g.

/app # ./download_data.py
/app # ./debug-app.py --host 0.0.0.0

If you prefer to run outside of docker instead, read on.

Steps:

1. Create Python virtual env
2. Install dependencies
3. Generate site from templates
4. Run local server

Repeat step 3 after changes.

Commands:

# clone the repo and enter the directory then:
python3 -m venv venv
source ./venv/bin/activate
pip install -r requirements.txt
./debug-app.py

Go to http://localhost:8000/en/ in the browser.

Build

python3 download_data.py
python3 -m http.server 8000 --directory build

Go to http://localhost:8000/en/ in the browser.

Generating sprites

Install NVM then:

nvm use 8
npm install -g @mapbox/spritezero-cli

Generate sprites by running:

nvm use 8
spritezero --ratio 4 ./static/style/sprite ./static/icons/

Deployment to AWS

1. Code is getting deployed using CodePipeline,
2. Static website is deployed to S3 bucket,
3. Reception points are generated on EC2 instance on regular basis,
4. Page is hosted behind CloudFront

Architecture diagram

Process diagram