Repository for my main website written in Python 3.8 and goin' fast using Sanic.
As a security measure, a separate user is created that runs the webserver and
handles ownership of all files within the project. This allows isolation of
files and provides an additional barrier between the system and the
application. Due to this, some steps here will mention a sanic user.
Services and the application they spawn run under this user. As such, setup
configuration should also be done under this user.
The first thing you will want to do is create a new virtual environment to install dependencies to so that they do not interfere with other projects. Make sure you have virtualenv installed beforehand.
$ pip install virtualenv ; python3 -m venv venv
This will create a new directory and prepare configuration files. To enter the new environment simply source the activation binary.
$ source venv/bin/activate
You can always leave the environment by calling the deactivate command.
From here on you can run everything you normally would and it should remain in the closed system you just created.
To get the server up and running you will have to first install all dependencies. These are listed in requirements.txt which pip can read and install for you.
$ sudo -u sanic pip3 install -r requirements.txt --no-cache-dir
The main server that takes care of all requests, routing and main functionality is Sanic. For the database SQLite3 with the help of SQLAlchemy is used. To setup the main database you'll want to run the setup.py script in the root directory. This also yields the first launch administrator password which you will need to get into the admin panel of the site. asdasd $ python3 setup.py
The site uses a custom build system for handling its resources. For style sheets it uses SCSS through libsass to compile these for the static directory of the site. JavaScript is minified through jsmin and also sent to the static directory from where all assets are served. Assets are compiled each time the server is launched. If you want to have instant feedback while developing then executing assets.py is the solution. It will track the assets directory using watchdog and compile any assets as they are changed and move them to the static directory.
Running the server after all prerequisites have been met is rather simple. All you have to do is execute the app.py script from Python.
$ python3 app.py
There's also an nginx setup already included in this repository's system/nginx directory which should get things started rather fast. To close things off you should change the configuration at the end of the app.py file to suit your needs. As it is a part of my website in a sense, my custom Nextcloud setup configuration of Nekocloud with my LibreOffice Online nginx and base configuration Nekodocs is also included. You can find the loolwsd config in system/loowsd.
For the server side handling of the process a systemd service script is also included within the system/systemd directory.
The site makes use of markdown for content rendering and editing of dynamic content. Custom pages and blog entries are published using markdown. The server renders the markdown and saves it to the database to avoid recompiling assets each page visit. This means that as a trade-off, more disc space is used to improve the time of serving content.
This repository is released under the MIT license. For more information please refer to LICENSE