pad-bis/pad-bis.py

# Flask application to serve the web pages
# mwclient to interact with the MediaWiki API
# BS to read the html table from the wiki
# os and dotenv to store the mediawiki credentials in a safe place

from flask import Flask, request, redirect, url_for, jsonify, render_template
import mwclient
from bs4 import BeautifulSoup
import os
from dotenv import load_dotenv
from pathlib import Path
import textwrap

# load the mediawiki credentials from the shared folder
dotenv_path = Path("/var/www/.mw-credentials")
load_dotenv(dotenv_path=dotenv_path)

# load the configuration env
load_dotenv()

DEFAULT_PAGE = os.environ.get("DEFAULT_PAGE", '')


# prefix to add /soupboat/padliography to all the routes
class PrefixMiddleware(object):
    def __init__(self, app, prefix=""):
        self.app = app
        self.prefix = prefix

    def __call__(self, environ, start_response):

        if environ["PATH_INFO"].startswith(self.prefix):
            environ["PATH_INFO"] = environ["PATH_INFO"][len(self.prefix):]
            environ["SCRIPT_NAME"] = self.prefix
            return self.app(environ, start_response)
        else:
            start_response("404", [("Content-Type", "text/plain")])
            return ["This url does not belong to the app.".encode()]


# create flask application
app = Flask(__name__)

# Get the URL prefix for the soupboat
# register the middleware to use our base_url as prefix on all the requests
base_url = os.environ.get('BASE_URL', '')
app.wsgi_app = PrefixMiddleware(app.wsgi_app, prefix=base_url)


def add_pad(padliography, link, title, overview, categories, date):
    '''Add a new pad to the wiki page'''

    # 1. Connect to the wiki
    site = mwclient.Site('pzwiki.wdka.nl', path='/mw-mediadesign/')

    # 2. Authenticate using the credential of a bot user registered in the wiki 
    ###  This is necesary the edit the contents of the page
    site.login(
        username=os.environ.get('MW_BOT'),
        password=os.environ.get('MW_KEY')
    )

    # 3. Select the page and get the contents
    # --> prefix the page title with Padliography/
    # so we dont erase eventual pages with the same title
    padliography = f'Padliography/{padliography}'
    page = site.pages[padliography]
    text = page.text()

    # 4. Append the pad as new row in the table of pads
    new_row = f'|-\n| {link} || {title} || {overview} || {categories} || {date} \n|-\n' + '|}'
    text = text.replace('|}', new_row)

    # 5. Apply the edit
    page.edit(text, f'New pad in the {padliography}: {title}')


def get_pads(padliography):
    '''Retrieve pads from the wiki'''

    # 1. Connect to the wiki
    site = mwclient.Site('pzwiki.wdka.nl', path='/mw-mediadesign/')

    # 2. Log in using the credential of a bot user registered in the wiki
    site.login(
        username=os.environ.get('MW_BOT'),
        password=os.environ.get('MW_KEY'),

    )

    # 3. Use the MediaWiki API to get the wikitext contents in HTML
    # Pages in the padliography comes with the Padliography/ prefix
    padliography = f'Padliography/{padliography}'
    html = site.api('parse', prop='text', page=padliography)

    # 4. Parse the HTML with BeautifulSoup to extract data from the table of pads
    table = BeautifulSoup(html['parse']['text']['*'], features="html.parser").find(
        "table", attrs={"class": "padliography"})

    # 5. Find the headers of the table
    headers = [header.text.lower().strip() for header in table.find_all('th')]
    
    # 6. Create a list of pad, using each header as property of the object pad
    pads = [
        {headers[i]: cell.text.rstrip('\n')
         for i, cell in enumerate(row.find_all('td'))}
        for row in table.find_all('tr')]

    #7. Remove empty pads from the list
    pads = [pad for pad in pads if pad != {}]

    return pads


def init_page(padliography, description):
    '''Initialize a new instance of the padliography a the given page'''

    # 1. Connect to the wiki
    site = mwclient.Site('pzwiki.wdka.nl', path='/mw-mediadesign/')

    # 2. Authenticate using the credential of a bot user registered in the wiki 
    #    This is necesary the edit the contents of the page
    site.login(
        username=os.environ.get('MW_BOT'),
        password=os.environ.get('MW_KEY')
    )

    # 3. Select the page and get the contents
    # page in the padliography comes with the Padliography/ prefix
    padliography = f'Padliography/{padliography}'
    page = site.pages[padliography]

    # 4. Insert the table template and a user-provided description

    text = f'''
    {description}

    == Padliography == 

    {{| class = "wikitable sortable padliography"
    |-
    !link !! title !! overview !! categories !! date
    |-
    |}}

    [[Category:Padliography]]

    '''

    # 5. Apply the edit
    page.edit(textwrap.dedent(text), f'New padliographish page created: Pads/{padliography}')


# Routes

@app.route('/')
def home():
    '''Serve the homepage layout'''
    return render_template('home.html', page=DEFAULT_PAGE, base_url=base_url)


@app.route('/<padliography>/')
def page(padliography):
    '''Serve a specific padliography'''
    return render_template('home.html', page=padliography, base_url=base_url)


@app.route('/api/<padliography>/', methods=['GET', 'POST'])
def api(padliography):
    '''Manage the interaction with the MediaWiki API'''

    if request.method == 'POST':
        # Add a new pad
        link = request.json.get('link', None)
        title = request.json.get('title', None)
        overview = request.json.get('overview', '')
        categories = request.json.get('categories', '')
        date = request.json.get('date', None)

        add_pad(padliography, link, title, overview, categories, date)
        redirect(url_for('home'))

    # Return the pad list
    response = jsonify({
        'pads': get_pads(padliography)
    })
    response.headers.add('Access-Control-Allow-Origin', '*')
    return response


@app.route('/api/<padliography>/init', methods=['GET', 'POST'])
def init(padliography):
    if request.method == 'POST':
        description = request.json.get('description', None)
        if padliography is not None:
            init_page(padliography, description)
            return redirect(url_for('home'))
    return 'ok'


# Get the port and mount the app
port = os.environ.get('FLASK_RUN_PORT', '')
debug = os.environ.get('DEBUG', False)
app.run(port=port, debug=debug)