(?:https?://)? # http(s):// (optional)
(?:youtu\.be/|(?:\w+\.)?youtube(?:-nocookie)?\.com/|
tube\.majestyc\.net/) # the various hostnames, with wildcard subdomains
+ (?:.*?\#/)? # handle anchor (#/) redirect urls
(?!view_play_list|my_playlists|artist|playlist) # ignore playlist URLs
(?: # the various things that can precede the ID:
(?:(?:v|embed|e)/) # v/ or embed/ or e/
except Trouble as trouble:
self._downloader.trouble(trouble[0])
+ if 'length_seconds' not in video_info:
+ self._downloader.trouble(u'WARNING: unable to extract video duration')
+ video_duration = ''
+ else:
+ video_duration = urllib.unquote_plus(video_info['length_seconds'][0])
+
# token
video_token = urllib.unquote_plus(video_info['token'][0])
'thumbnail': video_thumbnail.decode('utf-8'),
'description': video_description,
'player_url': player_url,
- 'subtitles': video_subtitles
+ 'subtitles': video_subtitles,
+ 'duration': video_duration
})
return results
self._downloader.trouble(u'ERROR: unable to extract media URL')
return
flashvars = urllib.unquote(mobj.group(1))
- if 'hqURL' in flashvars: max_quality = 'hqURL'
- elif 'sdURL' in flashvars: max_quality = 'sdURL'
- else: max_quality = 'ldURL'
+
+ for key in ['hd1080URL', 'hd720URL', 'hqURL', 'sdURL', 'ldURL', 'video_url']:
+ if key in flashvars:
+ max_quality = key
+ self._downloader.to_screen(u'[dailymotion] Using %s' % key)
+ break
+ else:
+ self._downloader.trouble(u'ERROR: unable to extract video URL')
+ return
+
mobj = re.search(r'"' + max_quality + r'":"(.+?)"', flashvars)
if mobj is None:
- mobj = re.search(r'"video_url":"(.*?)",', flashvars)
- if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract media URL')
+ self._downloader.trouble(u'ERROR: unable to extract video URL')
return
+
video_url = urllib.unquote(mobj.group(1)).replace('\\/', '/')
# TODO: support choosing qualities
return
video_title = unescapeHTML(mobj.group('title').decode('utf-8'))
- mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a></span>', webpage)
+ video_uploader = u'NA'
+ mobj = re.search(r'(?im)<span class="owner[^\"]+?">[^<]+?<a [^>]+?>([^<]+?)</a>', webpage)
if mobj is None:
- self._downloader.trouble(u'ERROR: unable to extract uploader nickname')
- return
- video_uploader = mobj.group(1)
+ self._downloader.trouble(u'WARNING: unable to extract uploader nickname')
+ else:
+ video_uploader = mobj.group(1)
video_upload_date = u'NA'
mobj = re.search(r'<div class="[^"]*uploaded_cont[^"]*" title="[^"]*">([0-9]{2})-([0-9]{2})-([0-9]{4})</div>', webpage)
class YoutubePlaylistIE(InfoExtractor):
"""Information Extractor for YouTube playlists."""
- _VALID_URL = r'(?:https?://)?(?:\w+\.)?youtube\.com/(?:(?:course|view_play_list|my_playlists|artist|playlist)\?.*?(p|a|list)=|user/.*?/user/|p/|user/.*?#[pg]/c/)(?:PL)?([0-9A-Za-z-_]+)(?:/.*?/([0-9A-Za-z_-]+))?.*'
+ _VALID_URL = r'(?:(?:https?://)?(?:\w+\.)?youtube\.com/(?:(?:course|view_play_list|my_playlists|artist|playlist)\?.*?(p|a|list)=|user/.*?/user/|p/|user/.*?#[pg]/c/)(?:PL|EC)?|PL|EC)([0-9A-Za-z-_]+)(?:/.*?/([0-9A-Za-z_-]+))?.*'
_TEMPLATE_URL = 'http://www.youtube.com/%s?%s=%s&page=%s&gl=US&hl=en'
- _VIDEO_INDICATOR_TEMPLATE = r'/watch\?v=(.+?)&list=.*?%s'
+ _VIDEO_INDICATOR_TEMPLATE = r'/watch\?v=(.+?)&([^&"]+&)*list=.*?%s'
_MORE_PAGES_INDICATOR = r'yt-uix-pager-next'
IE_NAME = u'youtube:playlist'
return
+class YoutubeChannelIE(InfoExtractor):
+ """Information Extractor for YouTube channels."""
+
+ _VALID_URL = r"^(?:https?://)?(?:youtu\.be|(?:\w+\.)?youtube(?:-nocookie)?\.com)/channel/([0-9A-Za-z_-]+)(?:/.*)?$"
+ _TEMPLATE_URL = 'http://www.youtube.com/channel/%s/videos?sort=da&flow=list&view=0&page=%s&gl=US&hl=en'
+ _MORE_PAGES_INDICATOR = r'yt-uix-button-content">Next' # TODO
+ IE_NAME = u'youtube:channel'
+
+ def report_download_page(self, channel_id, pagenum):
+ """Report attempt to download channel page with given number."""
+ self._downloader.to_screen(u'[youtube] Channel %s: Downloading page #%s' % (channel_id, pagenum))
+
+ def _real_extract(self, url):
+ # Extract channel id
+ mobj = re.match(self._VALID_URL, url)
+ if mobj is None:
+ self._downloader.trouble(u'ERROR: invalid url: %s' % url)
+ return
+
+ # Download channel pages
+ channel_id = mobj.group(1)
+ video_ids = []
+ pagenum = 1
+
+ while True:
+ self.report_download_page(channel_id, pagenum)
+ url = self._TEMPLATE_URL % (channel_id, pagenum)
+ request = urllib2.Request(url)
+ try:
+ page = urllib2.urlopen(request).read()
+ except (urllib2.URLError, httplib.HTTPException, socket.error), err:
+ self._downloader.trouble(u'ERROR: unable to download webpage: %s' % str(err))
+ return
+
+ # Extract video identifiers
+ ids_in_page = []
+ for mobj in re.finditer(r'href="/watch\?v=([0-9A-Za-z_-]+)&', page):
+ if mobj.group(1) not in ids_in_page:
+ ids_in_page.append(mobj.group(1))
+ video_ids.extend(ids_in_page)
+
+ if re.search(self._MORE_PAGES_INDICATOR, page) is None:
+ break
+ pagenum = pagenum + 1
+
+ for id in video_ids:
+ self._downloader.download(['http://www.youtube.com/watch?v=%s' % id])
+ return
+
+
class YoutubeUserIE(InfoExtractor):
"""Information Extractor for YouTube users."""