src/gpodder/youtube.py

   1 # -*- coding: utf-8 -*-
   2 #
   3 # gPodder - A media aggregator and podcast client
   4 # Copyright (c) 2005-2011 Thomas Perl and the gPodder Team
   5 #
   6 # gPodder is free software; you can redistribute it and/or modify
   7 # it under the terms of the GNU General Public License as published by
   8 # the Free Software Foundation; either version 3 of the License, or
   9 # (at your option) any later version.
  10 #
  11 # gPodder is distributed in the hope that it will be useful,
  12 # but WITHOUT ANY WARRANTY; without even the implied warranty of
  13 # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
  14 # GNU General Public License for more details.
  15 #
  16 # You should have received a copy of the GNU General Public License
  17 # along with this program.  If not, see <http://www.gnu.org/licenses/>.
  18 #
  19 #  gpodder.youtube - YouTube and related magic
  20 #  Justin Forest <justin.forest@gmail.com> 2008-10-13
  21 #
  22
  23
  24 import gpodder
  25
  26 from gpodder import util
  27 from gpodder.liblogger import log
  28
  29 try:
  30     import simplejson as json
  31 except ImportError:
  32     import json
  33
  34 import re
  35 import urllib
  36
  37 # See http://en.wikipedia.org/wiki/YouTube#Quality_and_codecs
  38 # Currently missing: the WebM 480p and 720 formats; 3GP profile
  39 supported_formats = [
  40     (37, '37/1920x1080/9/0/115', '1920x1080 (HD)'),
  41     (22, '22/1280x720/9/0/115', '1280x720 (HD)'),
  42     (35, '35/854x480/9/0/115', '854x480'),
  43     (34, '34/640x360/9/0/115', '640x360'),
  44     (18, '18/640x360/9/0/115', '640x360 (iPod)'),
  45     (18, '18/480x360/9/0/115', '480x360 (iPod)'),
  46     (5, '5/320x240/7/0/0', '320x240 (FLV)'),
  47 ]
  48
  49 class YouTubeError(Exception): pass
  50
  51 def get_real_download_url(url, preferred_fmt_id=None):
  52     # Default fmt_id when none preferred
  53     if preferred_fmt_id is None:
  54         preferred_fmt_id = 18
  55
  56     # For Maemo 5, we force fmt_id 5 for performance reasons
  57     if gpodder.ui.fremantle and not gpodder.ui.fermintle:
  58         preferred_fmt_id = 5
  59
  60     vid = get_youtube_id(url)
  61     if vid is not None:
  62         page = None
  63         url = 'http://www.youtube.com/watch?v=' + vid
  64
  65         while page is None:
  66             req = util.http_request(url, method='GET')
  67             if 'location' in req.msg:
  68                 url = req.msg['location']
  69             else:
  70                 page = req.read()
  71
  72         # Try to find the best video format available for this video
  73         # (http://forum.videohelp.com/topic336882-1800.html#1912972)
  74         def find_urls(page):
  75             r4 = re.search('.*"fmt_url_map"\:\s+"([^"]+)".*', page)
  76             if r4 is not None:
  77                 fmt_url_map = r4.group(1)
  78                 for fmt_url_encoded in fmt_url_map.split(','):
  79                     fmt_url = urllib.unquote(fmt_url_encoded)
  80                     fmt_url = fmt_url.replace('\\/', '/').replace('\u0026', '&')
  81                     fmt_id, url = fmt_url.split('|', 2)
  82                     yield int(fmt_id), url
  83
  84         fmt_id_url_map = sorted(find_urls(page), reverse=True)
  85         # Default to the highest fmt_id if we don't find a match below
  86         if fmt_id_url_map:
  87             default_fmt_id, default_url = fmt_id_url_map[0]
  88         else:
  89             raise YouTubeError('fmt_url_map not found for video ID "%s"' % vid)
  90
  91         formats_available = set(fmt_id for fmt_id, url in fmt_id_url_map)
  92         fmt_id_url_map = dict(fmt_id_url_map)
  93
  94         if gpodder.ui.fremantle:
  95             # This provides good quality video, seems to be always available
  96             # and is playable fluently in Media Player
  97             if preferred_fmt_id == 5:
  98                 fmt_id = 5
  99             else:
 100                 fmt_id = 18
 101         else:
 102             # As a fallback, use fmt_id 18 (seems to be always available)
 103             fmt_id = 18
 104
 105             # This will be set to True if the search below has already "seen"
 106             # our preferred format, but has not yet found a suitable available
 107             # format for the given video.
 108             seen_preferred = False
 109
 110             for id, wanted, description in supported_formats:
 111                 # If we see our preferred format, accept formats below
 112                 if id == preferred_fmt_id:
 113                     seen_preferred = True
 114
 115                 # If the format is available and preferred (or lower),
 116                 # use the given format for our fmt_id
 117                 if id in formats_available and seen_preferred:
 118                     log('Found available YouTube format: %s (fmt_id=%d)', \
 119                             description, id)
 120                     fmt_id = id
 121                     break
 122
 123         url = fmt_id_url_map.get(fmt_id, None)
 124         if url is None:
 125             url = default_url
 126
 127     return url
 128
 129 def get_youtube_id(url):
 130     r = re.compile('http://(?:[a-z]+\.)?youtube\.com/v/(.*)\.swf', re.IGNORECASE).match(url)
 131     if r is not None:
 132         return r.group(1)
 133
 134     r = re.compile('http://(?:[a-z]+\.)?youtube\.com/watch\?v=([^&]*)', re.IGNORECASE).match(url)
 135     if r is not None:
 136         return r.group(1)
 137
 138     return None
 139
 140 def is_video_link(url):
 141     return (get_youtube_id(url) is not None)
 142
 143 def get_real_channel_url(url):
 144     r = re.compile('http://(?:[a-z]+\.)?youtube\.com/user/([a-z0-9]+)', re.IGNORECASE)
 145     m = r.match(url)
 146
 147     if m is not None:
 148         next = 'http://www.youtube.com/rss/user/'+ m.group(1) +'/videos.rss'
 149         log('YouTube link resolved: %s => %s', url, next)
 150         return next
 151
 152     r = re.compile('http://(?:[a-z]+\.)?youtube\.com/profile?user=([a-z0-9]+)', re.IGNORECASE)
 153     m = r.match(url)
 154
 155     if m is not None:
 156         next = 'http://www.youtube.com/rss/user/'+ m.group(1) +'/videos.rss'
 157         log('YouTube link resolved: %s => %s', url, next)
 158         return next
 159
 160     return url
 161
 162 def get_real_cover(url):
 163     r = re.compile('http://www\.youtube\.com/rss/user/([^/]+)/videos\.rss', \
 164             re.IGNORECASE)
 165     m = r.match(url)
 166
 167     if m is not None:
 168         username = m.group(1)
 169         api_url = 'http://gdata.youtube.com/feeds/api/users/%s?v=2' % username
 170         data = util.urlopen(api_url).read()
 171         match = re.search('<media:thumbnail url=[\'"]([^\'"]+)[\'"]/>', data)
 172         if match is not None:
 173             log('YouTube userpic for %s is: %s', url, match.group(1))
 174             return match.group(1)
 175
 176     return None
 177
 178 def find_youtube_channels(string):
 179     url = 'http://gdata.youtube.com/feeds/api/videos?alt=json&q=%s' % urllib.quote(string, '')
 180     data = json.load(util.urlopen(url))
 181
 182     class FakeImporter(object):
 183         def __init__(self):
 184             self.items = []
 185
 186     result = FakeImporter()
 187
 188     seen_users = set()
 189     for entry in data['feed']['entry']:
 190         user = entry['author'][0]['name']['$t']
 191         title = entry['title']['$t']
 192         url = 'http://www.youtube.com/rss/user/%s/videos.rss' % user
 193         if user not in seen_users:
 194             result.items.append({
 195                 'title': user,
 196                 'url': url,
 197                 'description': title
 198             })
 199             seen_users.add(user)
 200
 201     return result
 202