A tiny flask web app blueprint

This is a basic python3 flask webapp blueprint for quickly adding a web interface to your python scripts. What you get:

• A flask webapp organized as a python3 package following the larger applications template.
• A html template starting point using Bootstrap 3.4.1 to extend for pretty, modern and responsive webpages!
• A Dockerfile to build your very own container and easily deploy it anywhere.

Start here

First, clone this repository git clone https://github.com/cmsirbu/tiny-flask or download a zip file. You don't really need the whole git part as this is the start of an application of your own!

Then, to develop locally, the cleanest way to do it is with a python3 virtualenv. Make sure you have python3 and pip3 installed, then run the commands below (only once, for setup):

# make sure you're in the repository you just downloaded
cd tiny-flask
# create a virtualenv and activate it
python3 -m venv venv
source venv/bin/activate
# update the tools
pip3 install -U pip setuptools
# install the tiny-flask package in the venv with its dependencies
pip3 install -e .

Once you have that set up, you can start the application in development mode anytime using:

# activate the virtualenv
source venv/bin/activate
# start flask's development webserver
FLASK_DEBUG=1 FLASK_APP=tinyflask flask run

The FLASK_DEBUG is absolutely essential for development: project files will be monitored for changes and the app restarted automatically + any errors/exceptions will be shown in page graphically as well as in the console.

You should see a message like this:

 * Serving Flask app "tinyflask"
 * Forcing debug mode on
 * Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)
 * Restarting with stat
 * Debugger is active!

Now open http://127.0.0.1:5000/ in your browser:

Success!

Running the webapp in production

The built-in server offered by flask is only useful for development and serves only one request at a time by default - read about flask deployment options.

Therefore, in order to actually serve multiple clients concurrently, one of the solutions in the above deployment link must be used. Ultimately it's up to you, but I've included an example running gunicorn to serve requests with the gevent async worker (which is more suited to long-running scripts or outgoing requests to other APIs etc.). The number of workers should roughly be around (2 x $num_cores) + 1 as the documentation suggests.

Install in the virtualenv

# activate the virtualenv
source venv/bin/activate
# install gunicorn and gevent
pip3 install gunicorn gevent
# run tinyflask
gunicorn tinyflask:app -b 127.0.0.1:5000 --workers 3 -k gevent
[2018-02-14 21:44:21 +0000] [15227] [INFO] Starting gunicorn 19.7.1
[2018-02-14 21:44:21 +0000] [15227] [INFO] Listening at: http://127.0.0.1:5000 (15227)
[2018-02-14 21:44:21 +0000] [15227] [INFO] Using worker: gevent
[2018-02-14 21:44:21 +0000] [15230] [INFO] Booting worker with pid: 15230
[2018-02-14 21:44:21 +0000] [15231] [INFO] Booting worker with pid: 15231
[2018-02-14 21:44:21 +0000] [15232] [INFO] Booting worker with pid: 15232

Build and run a Docker container

Building a container is a bit more involved, but it provides a very nice minimally packaged version of the webapp that can run on any Docker enabled host.

# build the container
docker build -t tinyflask:0.1 -f tiny-flask/Dockerfile .
# run it locally
docker run -p 5000:80 -d tinyflask:0.1
# open 127.0.0.1:5000 success!

You can save the container image and distribute it locally, publish it to Docker Hub or upload it to a private Docker registry. The sky's the limit!