wp2pelican

This script import data from a running Wordpress blog and create a static website using Pelican. It uses GraphQL API to connect to Wordpress. The aim is to keep Wordpress as backend but provide a static website as frontend.

What it does

• Imports posts
• Imports pages, but only the main ones and not the child pages. As Pelican does not use child pages, I don't see any use in importing them.
• Import one menu and set it as MENU_ITEMS to be used by Pelican (the name of the menu has to be set in wp2pelicanconf.py). It automatically imports child menus (usefull for drop-down menus).
• Replace URLs as set in the conf file.

Why

• It uses a conf file written in Python: it is not my first choice, but as Pelican works like this I kept it.

Documentation

This is not really a documentation per se, but more notes that I use.

Wordpress and GraphQL

I use wp-graphql plugin in Worpdress to set up the API. It is pretty straightforward as you only need to install the plugin and you're done, but offer no security whatsoever: everyone can connect to the endpoint and then can send request to read/write. I use a HTTP proxy to mitigate that for now, but I'll have to look into that.

Wordpress proxy

If Wordpress (or the GraphQL endpoint) is protected by a HTTP authentification, there are two ways of storing the identifcation:

• In wp2pelicanconf.py with HTTP_AUTH = ("id", "password")
• Use environment variables WP_PROXY_ID and WP_PROXY_PWD: this allow to not write the password in a file and can be use with Gitlab CI.

Installation

git clone git@github.com:shan-x/wp2pelican.git
cp wp2pelican
virtualenv venv
source venv/bin/activate
pip install -r requirements.txt
pyhton3 wp2pelican.py

Setup Pelican

The output of the script will be in $PELICAN_PATH, where the file pelicaconf.py should be located. At the end of this file you have to add:

import sys, os
sys.path.append(os.curdir)
from pelicanvar import *