Flask skeleton using Bootstrap, SCSS, Docker, console and rotating file logging, HTTP basic auth and web and api views with Blueprint.

Summary

This is a simple flask skeleton to quickly start a new web project using:

Bootstrap
SCSS
console and rotating file logging
HTTP basic auth
web and api views with Blueprint

You can use it:

localy with autoreload
dockerized with autoreload
heroku without autoreload

And you can switch between them.

If you’re not confortable with Flask follow:

If you’re not confortable with SASS/SCSS follow:

Getting Started with SASS (with interactive examples)

Usage

To lazy people like me, you can simply use make commands to deploy a new project.

make creates a .deployed file, if you remove it, next time you run make, it will remove your git environment

clone repository to a project name (important):

git clone https://github.com/ahmet2mir/flask-skeleton.git myproject
cd myproject

help message:

$ make || make usage
Tools to create a Flask environment.

When you clone this project, clone into your project name.
Because *make will rename all 'skeleton' modules and params with this name.
Default is 'flaskskeleton'.

    example: git clone https://github.com/ahmet2mir/flask-skeleton.git myproject
                cd myproject
                make local

Allowed targets:

    usage: this help.

    local: create virtualenv, install requirements and run.
        auto-reload: yes.
        ctrl+c: stop process.
        next make local: start main.py.

    docker: build image and run.
        auto-reload: yes.
        ctrl+c: remove container.
        next make docker: start container.

    heroku: create, push, scale and run application.
        auto-reload: no, need to push content.
        ctrl+c: stop log tail but application still running.
        next make herok: start web, show info and tail logs.

/!\ .deployed is a state file, if you remove it, the next time  you will run make, it will remove your git environment.

start locally:

make local
...
* Debugger is active!
* Debugger pin code: 298-561-187

start with docker:

make docker
...
* Debugger is active!
* Debugger pin code: 298-561-187

start with heroku:

make heroku
2015-12-19T11:43:15.165081+00:00 app[web.1]: [2015-12-19 11:43:15 +0000] [3] [INFO] Starting gunicorn 19.4.1
2015-12-19T11:43:15.212135+00:00 app[web.1]: [2015-12-19 11:43:15 +0000] [11] [INFO] Booting worker with pid: 11

to stop press ctrl+c - local: stop process. - docker: remove container. - heroku: stop log tail but application still running.

Next time you use make with target: - make local: start main.py. - make docker: start container. - make heroku: start web, show info and tail logs.

You can, switch between targets without recreating a new project.

Example:

$ make local
...
* Debugger is active!
* Debugger pin code: 298-561-187

ctrl+c
$ make docker
...
* Debugger is active!
* Debugger pin code: 298-561-187

ctrl+c

$ make heroku
...

Test

Visit:

//127.0.0.1:5000/

or get url with heroku:

$ heroku info -s | grep -i web-url | cut -d'=' -f2
https://xxxx-yyyy-0000.herokuapp.com/

API:

curl -XGET -u admin:mypass //127.0.0.1:5000/api/v1.0/health
{
  "health": "Good doctor!"
}

With bad auth:

curl -XGET -u admin:badpass //127.0.0.1:5000/api/v1.0/health
{
  "error": "Unauthorized access"
}

Details

Create new scss files under static/scss/ and make the import in __main__.scss, the main.css is generated in /static/css/main.css
Log file is in root repo named myproject.log
HTTP auth is done by auth.py

Picture by skycaptaintwo under licence CC BY 2.0