⚠️ Important Warning ⚠️

WARNING: This repository is currently under development.

Please note:

It includes experimental code that may be abandoned after evaluation.
It has code for tools that may be removed. Some of the tools currently in the repo are mutually exclusive.
Many commits include incomplete code. (These are typically, but not necessarily, identified by having "WIP" in the comment.
It is a custom application with processsing rules that are specific to the history of the Bioconductor project. We have no intention, and do not recommend, attempting to use this system in other contexts.
Much of the text below is a vestige of the original cookiecutter-flask code base. Do not rely on it.

bioc_webstats

Public accss to Bioconductor package download history

Portions of this code base are derived from cookie-cutter-flask and used with permission granted under the MIT license.

Docker Quickstart

This app can be run completely using Docker and docker-compose. Using Docker is recommended, as it guarantees the application is run using compatible versions of Python and Node.

There are three main services:

To run the development version of the app

docker-compose up flask-dev

To run the production version of the app

docker-compose up flask-prod

The list of environment: variables in the docker-compose.yml file takes precedence over any variables specified in .env.

To run any commands using the Flask CLI

docker-compose run --rm manage <<COMMAND>>

Therefore, to initialize a database you would run

docker-compose run --rm manage db init
docker-compose run --rm manage db migrate
docker-compose run --rm manage db upgrade

A docker volume node-modules is created to store NPM packages and is reused across the dev and prod versions of the application. For the purposes of DB testing with sqlite, the file dev.db is mounted to all containers. This volume mount should be removed from docker-compose.yml if a production DB server is used.

Go to http://localhost:8080. You will see a pretty welcome screen.

Running locally

Run the following commands to bootstrap your environment if you are unable to run the application using Docker

cd bioc_webstats
pip install -r requirements/dev.txt
npm install
npm run-script build
npm start  # run the webpack dev server and flask server using concurrently

Go to http://localhost:5000. You will see a pretty welcome screen.

Database Initialization (locally)

Once you have installed your DBMS, run the following to create your app's database tables and perform the initial migration

flask db init
flask db migrate
flask db upgrade

Deployment

When using Docker, reasonable production defaults are set in docker-compose.yml

FLASK_ENV=production
FLASK_DEBUG=0

Therefore, starting the app in "production" mode is as simple as

docker-compose up flask-prod

If running without Docker

export FLASK_ENV=production
export FLASK_DEBUG=0
export DATABASE_URL="<YOUR DATABASE URL>"
npm run build   # build assets with webpack
flask run       # start the flask server

Shell

To open the interactive shell, run

docker-compose run --rm manage shell
flask shell # If running locally without Docker

By default, you will have access to the flask app.

Running Tests/Linter

To run all tests, run

docker-compose run --rm manage test
flask test # If running locally without Docker

To run the linter, run

docker-compose run --rm manage lint
flask lint # If running locally without Docker

The lint command will attempt to fix any linting/style errors in the code. If you only want to know if the code will pass CI and do not wish for the linter to make changes, add the --check argument.

Migrations

Whenever a database migration needs to be made. Run the following commands

docker-compose run --rm manage db migrate
flask db migrate # If running locally without Docker

This will generate a new migration script. Then run

docker-compose run --rm manage db upgrade
flask db upgrade # If running locally without Docker

To apply the migration.

For a full migration command reference, run docker-compose run --rm manage db --help.

If you will deploy your application remotely (e.g on Heroku) you should add the migrations folder to version control. You can do this after flask db migrate by running the following commands

git add migrations/*
git commit -m "Add migrations"

Make sure folder migrations/versions is not empty.

Asset Management

Files placed inside the assets directory and its subdirectories (excluding js and css) will be copied by webpack's file-loader into the static/build directory. In production, the plugin Flask-Static-Digest zips the webpack content and tags them with a MD5 hash. As a result, you must use the static_url_for function when including static content, as it resolves the correct file name, including the MD5 hash. For example

<link rel="shortcut icon" href="{{static_url_for('static', filename='build/favicon.ico') }}">

If all of your static files are managed this way, then their filenames will change whenever their contents do, and you can ask Flask to tell web browsers that they should cache all your assets forever by including the following line in .env:

SEND_FILE_MAX_AGE_DEFAULT=31556926  # one year

Bioconductor/bio-web-stats