dhammatalks_podcast.py

import bs4
import dateparser
import re

with open('/home/imartinezvazquez/rss/mp3_long.php') as fd:
    doc = bs4.BeautifulSoup(fd.read())

links = doc.find_all('a')
links = [l for l in links if l.attrs['href'].endswith('.mp3')]

l.parent.parent

link.parent.parent.find('span', class_='duration').text.partition(':')

import pytz

def parse_link(link):
        datestr = link.parent.parent.find('span', class_='date').text
        date = dateparser.parse(datestr+' EST')
        duration = link.parent.parent.find('span', class_='duration').text
        minutes = int(re.match(r'Duration: (\d+) min', duration).group(1))
        duration = datetime.timedelta(minutes=minutes)
        title = link.text.strip()
        href = link.attrs['href']
        return title, date, duration, href

from podgen import Podcast, Episode, Media

p = Podcast(
       name="Dhammatalks.org Lectures",
       description="Dhammatalks.org Lectures",
       website="https://www.dhammatalks.org/mp3_long.php"
    )

for link in links:
    title, date, duration, href = parse_link(link)
    episode = Episode(
        title=title, 
        publication_date=date,
        media=Media.create_from_server_response(href, duration=duration)
    )
    p.episodes.append(episode)

p.explicit = False
import codecs
with codecs.open('/tmp/lectures.rss', 'w', encoding='utf8') as fd:
    fd.write(unicode(p))