try:
new_url = self._test_redirect(url)
if new_url:
- return [self.url_result(new_url)]
+ return self.url_result(new_url)
except compat_urllib_error.HTTPError:
# This may be a stupid server that doesn't like HEAD, our UA, or so
pass
raise ExtractorError(u'Failed to download URL: %s' % url)
self.report_extraction(video_id)
+
+ # it's tempting to parse this further, but you would
+ # have to take into account all the variations like
+ # Video Title - Site Name
+ # Site Name | Video Title
+ # Video Title - Tagline | Site Name
+ # and so on and so forth; it's just not practical
+ video_title = self._html_search_regex(
+ r'(?s)<title>(.*?)</title>', webpage, u'video title',
+ default=u'video')
+
+ # video uploader is domain name
+ video_uploader = self._search_regex(
+ r'^(?:https?://)?([^/]*)/.*', url, u'video uploader')
+
# Look for BrightCove:
bc_url = BrightcoveIE._extract_brightcove_url(webpage)
if bc_url is not None:
return self.url_result(surl, 'Vimeo')
# Look for embedded YouTube player
- mobj = re.search(
- r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?youtube.com/embed/.+?)\1', webpage)
+ matches = re.findall(
+ r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?youtube\.com/embed/.+?)\1', webpage)
+ if matches:
+ urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Youtube')
+ for tuppl in matches]
+ return self.playlist_result(
+ urlrs, playlist_id=video_id, playlist_title=video_title)
+
+ # Look for embedded Dailymotion player
+ matches = re.findall(
+ r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/embed/video/.+?)\1', webpage)
+ if matches:
+ urlrs = [self.url_result(unescapeHTML(tuppl[1]), 'Dailymotion')
+ for tuppl in matches]
+ return self.playlist_result(
+ urlrs, playlist_id=video_id, playlist_title=video_title)
+
+ # Look for embedded Wistia player
+ match = re.search(
+ r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:fast\.)?wistia\.net/embed/iframe/.+?)\1', webpage)
+ if match:
+ return {
+ '_type': 'url_transparent',
+ 'url': unescapeHTML(match.group('url')),
+ 'ie_key': 'Wistia',
+ 'uploader': video_uploader,
+ 'title': video_title,
+ 'id': video_id,
+ }
+
+ # Look for embedded blip.tv player
+ mobj = re.search(r'<meta\s[^>]*https?://api.blip.tv/\w+/redirect/\w+/(\d+)', webpage)
+ if mobj:
+ return self.url_result('http://blip.tv/seo/-'+mobj.group(1), 'BlipTV')
+ mobj = re.search(r'<(?:iframe|embed|object)\s[^>]*https?://(?:\w+\.)?blip.tv/(?:play/|api\.swf#)([a-zA-Z0-9]+)', webpage)
if mobj:
- surl = unescapeHTML(mobj.group(u'url'))
- return self.url_result(surl, 'Youtube')
+ player_url = 'http://blip.tv/play/%s.x?p=1' % mobj.group(1)
+ player_page = self._download_webpage(player_url, mobj.group(1))
+ blip_video_id = self._search_regex(r'data-episode-id="(\d+)', player_page, u'blip_video_id', fatal=False)
+ if blip_video_id:
+ return self.url_result('http://blip.tv/seo/-'+blip_video_id, 'BlipTV')
# Look for Bandcamp pages with custom domain
mobj = re.search(r'<meta property="og:url"[^>]*?content="(.*?bandcamp\.com.*?)"', webpage)
if mobj is not None:
burl = unescapeHTML(mobj.group(1))
- return self.url_result(burl, 'Bandcamp')
+ # Don't set the extractor because it can be a track url or an album
+ return self.url_result(burl)
+
+ # Look for embedded Vevo player
+ mobj = re.search(
+ r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:cache\.)?vevo\.com/.+?)\1', webpage)
+ if mobj is not None:
+ return self.url_result(mobj.group('url'))
# Start with something easy: JW Player in SWFObject
mobj = re.search(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
mobj = re.search(r'[^A-Za-z0-9]?(?:file|source)=(http[^\'"&]*)', webpage)
if mobj is None:
# Broaden the search a little bit: JWPlayer JS loader
- mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"&]*)', webpage)
+ mobj = re.search(r'[^A-Za-z0-9]?file["\']?:\s*["\'](http[^\'"]*)', webpage)
if mobj is None:
# Try to find twitter cards info
mobj = re.search(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
video_id = compat_urllib_parse.unquote(os.path.basename(video_url))
# here's a fun little line of code for you:
- video_extension = os.path.splitext(video_id)[1][1:]
video_id = os.path.splitext(video_id)[0]
- # it's tempting to parse this further, but you would
- # have to take into account all the variations like
- # Video Title - Site Name
- # Site Name | Video Title
- # Video Title - Tagline | Site Name
- # and so on and so forth; it's just not practical
- video_title = self._html_search_regex(r'<title>(.*)</title>',
- webpage, u'video title', default=u'video', flags=re.DOTALL)
-
- # video uploader is domain name
- video_uploader = self._search_regex(r'(?:https?://)?([^/]*)/.*',
- url, u'video uploader')
-
- return [{
+ return {
'id': video_id,
'url': video_url,
'uploader': video_uploader,
- 'upload_date': None,
'title': video_title,
- 'ext': video_extension,
- }]
+ }
projects / youtube-dl / blobdiff