Skip to content

xofbd/altair-flask-demo

Repository files navigation

Python License: GPL-3.0 CI codecov

Altair Flask Demo

This project demonstrates how to serve an Altair visualization using Flask. The application shows potential sites to implement a geothermal energy technology by re-purposing old wells. The user specifies the criteria for the wells and a plot is generated showing the location of those wells. One can deploy this app using a service like Heroku or just locally.

Prerequisites

Python environment

All required Python packages can be found in the requirements.txt file. Since the project makes use of GNU Make, there is no need to directly create the environment. The creation and activation of the environment is handled by running the Make rules that are shown in later sections. If you do go the Make route, make sure you have Python Poetry installed.

SQL database

The application uses an SQL database and makes queries against it to obtain the wells to visualize. Two options are presented: SQLite for only local deployment or PostgreSQL for either local or deployment to Heroku. Instructions for both are found below. Note, you can skip the last step of each and jump to deployment by running make all, as running make all takes care of creating the database once .env has been configured properly. By default, create-db truncates the records in the database to comply with Heroku's free tier limit. However, you can prevent truncation by setting the environmental variable TRUNCATE to an empty string. E.g., make TRUNCATE="" all.

SQLite

  1. Create your own .env: cp .env.template .env
  2. Assign URI_DB to sqlite:///data/wells.db
  3. Run make create-db or make TRUNCATE="" create-db (no truncation)

Heroku managed PostgreSQL

  1. Create a Heroku account
  2. Install the Heroku CLI
  3. Create your own .env: cp .env.template .env
  4. Create a Heroku app: heroku create
  5. In your .env, set APP_NAME to your application's name form the previous step
  6. Run make create-db

Other databases

You can use a database that is not managed by Heroku. For example, your own Postgres server you have setup yourself or a fully managed one by a different provider. If you want to go the latter route, consider ElephantSQL. The free tier lets you have a database up to 20 MB. Whatever you decide, the instructions are similiar, run make create-db. However, before you run make create-db, make sure you add export OTHER_DB=true to your .env.

Two things to consider:

  1. If you are deploying to Heroku, make sure you set APP_NAME as make create-db will set the URI_DB environmental variable on Heroku. If you don't set APP_NAME, you will need to make sure you set URI_DB on Heroku yourself.

  2. If you are using a different RDBMS, e.g., MySQL, you'll need to make sure you have the appropriate Python driver installed in your virtual environment.

Deployment

Now that you have the SQL database has been created and .env configured properly, you are ready to deploy locally or to Heroku. To deploy the app locally, simply run make deploy. It will use whatever database is specified in URI_DB in .env.

To deploy the app to Heroku:

  1. bin/set_secret_key
  2. git push heroku master
  3. heroku open

Deployment with Docker

You can deploy and run the application locally using Docker. After creating a database, you can run make docker-run or:

  1. docker build -t flask_app .
  2. docker run --init --rm -d --publish 127.0.0.1:5000:5000 flask_app

Once running, visit localhost:5000 on your web browser. As with other methods, you need to make sure you've created your database and properly configured your .env.

License

This project is distributed under the GNU General Purpose License. Please see LICENSE for more information.

About

Demonstration of using Flask and Altair

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published