Tiny webapp to convert AGDQ 2016 schedule page to CSV for importing into Google calendar

agdq-gcal-export.py

#!/bin/env python2
# -*- encoding: utf-8 -*-

import csv
import datetime
import pytz

import flask
import bs4
import requests


AGDQ_SCHEDULE_URL = 'https://gamesdonequick.com/schedule'
TIMESTAMP_FMT = '%Y-%m-%dT%H:%M:%SZ'
CSV_DATE_FMT = '%m/%d/%Y'
CSV_TIME_FMT = '%I:%M %p'
CSV_KEYS = ['Subject', 'Start Date', 'Start Time', 'End Date', 'End Time', \
    'All Day Event', 'Description']

app = flask.Flask(__name__)

@app.route('/')
def export_schedule_csv():
    schedule_resp = get_stream_url(AGDQ_SCHEDULE_URL)
    parsed_table = parse_schedule_page(schedule_resp)

    return flask.Response(generate_schedule_csv(parsed_table), mimetype='text/csv')


def get_stream_url(url):
    resp = requests.get(url, stream=True)
    return resp.content

def parse_schedule_page(page_content):
    iter_cells = lambda row: row.find_all(name='td', recursive=False)

    soup = bs4.BeautifulSoup(page_content, 'html.parser')
    table = soup.find(id='runTable').parent
    keys = [cell.text.lower().replace(' ', '_') \
        for cell in iter_cells(table.thead.tr)]

    table_data = [dict(zip(keys, (cell.text.encode('utf-8') for cell in iter_cells(row)))) \
        for row in table.tbody.find_all(name='tr', recursive=False)]

    return table_data

def generate_schedule_csv(table_data, local_timezone='America/Detroit'):
    quote = lambda s: '"{}"'.format(s.replace('"',r'\"'))
    fmt_time = lambda dt: dt.strftime(CSV_TIME_FMT)
    fmt_date = lambda dt: dt.strftime(CSV_DATE_FMT)
    parse_timedelta = lambda s: datetime.timedelta( \
        **dict(zip( \
            ['seconds', 'minutes', 'hours'], \
            (int(i) for i in s.split(':')[::-1])) \
        ))
    parse_timestamp = lambda ts: pytz.utc.localize(datetime.datetime.strptime( \
        ts, TIMESTAMP_FMT)).astimezone(pytz.timezone(local_timezone))
    csv_fmt_header = lambda: ','.join(CSV_KEYS) + '\n'

    def csv_fmt_data_row(row):
        start_dt = parse_timestamp(row['start_time'])
        end_dt = start_dt + parse_timedelta(row['setup_time']) + \
            parse_timedelta(row['run_time'])
        return ','.join(quote(i) for i in [
            row['name'],
            fmt_date(start_dt),
            fmt_time(start_dt),
            fmt_date(end_dt),
            fmt_time(end_dt),
            'False',
            ' '.join([row['category'], row['description']]),
        ]) + '\n'

    yield csv_fmt_header()
    for row in table_data:
        # skip the last row because it doesn't have as many cells as the rest
        # (easier than an actual fix)
        if row['name'] != 'Finale!':
            yield csv_fmt_data_row(row)

if __name__ == '__main__':
    app.run(debug=True)

requirements.txt

beautifulsoup4==4.4.1
Flask==0.10.1
itsdangerous==0.24
Jinja2==2.8
MarkupSafe==0.23
pytz==2015.7
requests==2.9.1
Werkzeug==0.11.3