def process_episode(number):
audiofile = Settings.local_audio_filename(number)
htmlfile = Settings.local_html_filename(number)
# Make sure that we have the html file and (if desired) the mp3 file
if Settings.CACHE_MP3S and not os.path.isfile(htmlfile):
raise Exception("The HTML file for episode {0} is missing".format(number))
if not os.path.isfile(audiofile):
raise Exception("The MP3 file for episode {0} is missing".format(number))
try:
file_contents = open(htmlfile, 'r').read().decode('utf-8')
soup = bs4.BeautifulSoup(file_contents)
except Exception as e:
print "Problem trying to read {0}".format(htmlfile)
raise e
try:
# Get size of mp3 file
# TODO: Come up with some way to get the size of the remote files
filesize = os.path.getsize(audiofile) if Settings.CACHE_MP3S else 28000000
content_div = soup.find("div", {"id" : "content"})
if content_div is None:
raise LookupError("Couldn't find a div named 'content_div'")
acts = get_acts(soup)
# Combine all act text into a single string. *Within* a single act, separate the
# lines by newlines. *Between* acts, separate them by double-newlines
# we might need to stick a '
' after each \n
all_acts_text = '\n\n'.join(['===========================\n' + act['head'] + '\n' + act['body'] for act in acts])
# Start building our item
item = ET.Element('item')
# title tag
title = ET.SubElement(item, 'title')
title.text = get_raw_content(content_div, "h1", {"class" : "node-title"})
description = ET.SubElement(item, 'description')
description.text = get_raw_content(content_div, "div", {"class" : "description"}) + '\n' + all_acts_text
# pubDate tag
# Dates in the html are in the form of "Dec 22, 1995". Parse them to turn them into the RFC format
datestring = get_raw_content(content_div, "div", {"class" : "date"})
dateobj = datetime.strptime(datestring, "%b %d, %Y")
pubDate = ET.SubElement(item, 'pubDate')
pubDate.text = dateobj.strftime("%a, %d %b %Y 00:00:00 +0000")
url = Settings.local_audio_url(number) if Settings.CACHE_MP3S else Settings.remote_audio_url(number)
# link tag
link = ET.SubElement(item, 'link')
link.text = url
# guid tag
guid = ET.SubElement(item, 'guid')
guid.text = url
# enclosure tag (how to actually find the audio clip)
enclosure = ET.SubElement(item, 'enclosure')
enclosure.set('url',url)
enclosure.set('length',str(filesize))
enclosure.set('type','audio/mpeg')
# itunes:summary tag (this shows where the liner-notes or lyrics normally go)
# summary = ET.SubElement(item, 'itunes:summary')
# summary.text = all_acts_text
# subtitle = ET.SubElement(item, 'itunes:subtitle')
# subtitle.text = all_acts_text
# resultset = soup.find_all("div", {"class", "act-body"})
# print "Acts: {0}".format(len(resultset))
return item
except ValueError as e:
print "Caught an error when trying to process episode {0}".format(number)
raise Exception("Problem processing episode {0}".format(number))
def fetch_audio(number): fetch_to_file(Settings.remote_audio_url(number), Settings.local_audio_filename(number)) return Settings.local_audio_filename(number)