get_podcasts/get_podcasts.py

#!/usr/bin/env python3

#############################
###### USER VARIABLES #######
#############################

# set download root dir
download_root_dir = '/home/bryan/Downloads/Podcasts'

# set handheld directory
handheld_root_dir = '/run/media/bryan/D5E7-1B94/Podcasts'

# feeds (python list)
feedurls = ['http://linuxactionnews.com/rss', \
            'http://wakingup.libsyn.com/rss', \
            'http://feeds.soundcloud.com/users/soundcloud:users:146429914/sounds.rss', \
            'http://billburr.libsyn.com/rss', \
            'http://feeds.99percentinvisible.org/99percentinvisible', \
            'http://rss.art19.com/tim-ferriss-show']

# number of old episodes to keep
old_episodes_keep = 3

# include episode name in filename (1 = on, 0 = off)
episode_name_in_filename = 0

# enable debug output (1 = on, 0 = off)
debug = 0

#################################
###### END USER VARIABLES #######
#################################

import podcastparser
import urllib.request
import os
import time
import datetime
import glob
import shutil

# spoof headers for certain rss feeds
opener=urllib.request.build_opener()
opener.addheaders=[('User-Agent','Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/36.0.1941.0 Safari/537.36')]
urllib.request.install_opener(opener)

# create new timestamp
timestamp_file = os.path.join(download_root_dir, '.last_sync')
new_ts = int(time.time())

# get old timestamp
if os.path.exists(timestamp_file):
  with open(timestamp_file) as f:
    old_ts = int(f.readline())
else:
  old_ts = 0

#################################
######### GET EPISODES ##########
#################################

# parse feeds
for feed in feedurls:

  # parsed is a dict
  parsed = podcastparser.parse(feed, urllib.request.urlopen(feed), old_episodes_keep)
  
  # debug
  if debug == 1:
    import pprint
    pprint.pprint(parsed)
  
  # get podcast title
  podcast_title = parsed['title']
  print('Checking ' + podcast_title + ' for new episdoes...')
  
  # check if download dir exists and if not, make it
  podcast_download_dir = os.path.join(download_root_dir, podcast_title)
  if not os.path.exists(podcast_download_dir):
    os.mkdir(podcast_download_dir)

  # parse episodes
  for episode in parsed['episodes']:
    # get release time
    release_time = int(episode['published'])
    
    # debug
    if debug == 1:
      print(old_ts)
      print(release_time)
    
    if release_time > old_ts:
      # format release time to date format
      episode_date = datetime.datetime.fromtimestamp(release_time)
      episode_date = episode_date.strftime("%y-%m-%d")
      
      # debug
      if debug == 1:
        print(episode_date)
        
      # create filename based on episode date and/or episode title
      if episode_name_in_filename == 0:
        episode_title = episode_date
      else:
        episode_title = episode_date + " " + episode['title']
        
      full_episode_path = os.path.splitext(os.path.join(podcast_download_dir, episode_title))[0]
    
      for enclosures in episode['enclosures']:
         # get download url
         url = enclosures['url']
         if enclosures['mime_type'] == 'audio/mpeg':
           full_episode_path += '.mp3'
      # download file and save to episode title
      if not os.path.exists(full_episode_path):
        try:
          print('Downloading ' + episode['title'])
          urllib.request.urlretrieve(url, full_episode_path)
        except urllib.error.HTTPError as e:
          print('HTTPError' + ': Could not download ' + podcast_title + '\n' + 'Reason: ' + e.reason + '\n' + 'URL: ' + url)
        except urllib.error.URLError:
          print('URLError')
        except urllib.error.ContentTooShortError:
          print('Download failed, file corrupt!')

  # remove old episodes
  old_files = glob.glob(os.path.join(podcast_download_dir, '*'))
  old_files.sort()
  for file in old_files[:-old_episodes_keep]:
    os.unlink(file)
          
# write new timestamp file
with open(timestamp_file, 'w') as outf:
  outf.write(str(new_ts))
        
#################################
######## MANAGE HANDHELD ########
#################################        

if os.path.exists(handheld_root_dir):

  # define copy to handheld function
  def copy_files(src_file, dest_dir):
    if not os.path.isdir(dest_dir):
      os.makedir(dest_dir)
    filename = os.path.basename(src_file)
    dest_file = os.path.join(dest_dir, filename)
    if not os.path.exists(dest_file):
      print('Copying ' + src_file + ' to ' + dest_file)
      shutil.copyfile(src_file, dest_file)          

      
  # copy to handheld
  podcast_dirs = glob.glob(os.path.join(download_root_dir, '*'))
  for podcast_dir in podcast_dirs:
    podcast_name = os.path.basename(podcast_dir)
    handheld_target_dir = os.path.join(handheld_root_dir, podcast_name)
    files_to_copy = glob.glob(os.path.join(podcast_dir, '*'))
    for file in files_to_copy:
      if debug == 1:
        print(file + ' ' + handheld_target_dir)
      copy_files(file, handheld_target_dir)

  # remove old files from handheld
  podcast_dirs = glob.glob(os.path.join(handheld_root_dir, '*'))
  for podcast_dir in podcast_dirs:
    old_files = glob.glob(os.path.join(podcast_dir, '*'))
    old_files.sort()
    for file in old_files[:-old_episodes_keep]:
      os.unlink(file)

else:
  print('Handheld not detected, skipping...')