from .common import InfoExtractor
from .youtube import YoutubeIE
from ..utils import (
- compat_urllib_error,
compat_urllib_parse,
- compat_urllib_request,
compat_urlparse,
compat_xml_parse_error,
+ determine_ext,
ExtractorError,
float_or_none,
HEADRequest,
'uploader': 'www.handjobhub.com',
'title': 'Busty Blonde Siri Tit Fuck While Wank at Handjob Hub',
}
- }
+ },
+ # RSS feed
+ {
+ 'url': 'http://phihag.de/2014/youtube-dl/rss2.xml',
+ 'info_dict': {
+ 'id': 'http://phihag.de/2014/youtube-dl/rss2.xml',
+ 'title': 'Zero Punctuation',
+ 'description': 're:'
+ },
+ 'playlist_mincount': 11,
+ },
+ # Multiple brightcove videos
+ # https://github.com/rg3/youtube-dl/issues/2283
+ {
+ 'url': 'http://www.newyorker.com/online/blogs/newsdesk/2014/01/always-never-nuclear-command-and-control.html',
+ 'info_dict': {
+ 'id': 'always-never',
+ 'title': 'Always / Never - The New Yorker',
+ },
+ 'playlist_count': 3,
+ 'params': {
+ 'extract_flat': False,
+ 'skip_download': True,
+ }
+ },
+ # MLB embed
+ {
+ 'url': 'http://umpire-empire.com/index.php/topic/58125-laz-decides-no-thats-low/',
+ 'md5': '96f09a37e44da40dd083e12d9a683327',
+ 'info_dict': {
+ 'id': '33322633',
+ 'ext': 'mp4',
+ 'title': 'Ump changes call to ball',
+ 'description': 'md5:71c11215384298a172a6dcb4c2e20685',
+ 'duration': 48,
+ 'timestamp': 1401537900,
+ 'upload_date': '20140531',
+ 'thumbnail': 're:^https?://.*\.jpg$',
+ },
+ },
+ # Wistia embed
+ {
+ 'url': 'http://education-portal.com/academy/lesson/north-american-exploration-failed-colonies-of-spain-france-england.html#lesson',
+ 'md5': '8788b683c777a5cf25621eaf286d0c23',
+ 'info_dict': {
+ 'id': '1cfaf6b7ea',
+ 'ext': 'mov',
+ 'title': 'md5:51364a8d3d009997ba99656004b5e20d',
+ 'duration': 643.0,
+ 'filesize': 182808282,
+ 'uploader': 'education-portal.com',
+ },
+ },
]
- def report_download_webpage(self, video_id):
- """Report webpage download."""
- if not self._downloader.params.get('test', False):
- self._downloader.report_warning('Falling back on generic information extractor.')
- super(GenericIE, self).report_download_webpage(video_id)
-
def report_following_redirect(self, new_url):
"""Report information extraction."""
self._downloader.to_screen('[redirect] Following redirect to %s' % new_url)
url, smuggled_data = unsmuggle_url(url)
force_videoid = None
+ is_intentional = smuggled_data and smuggled_data.get('to_generic')
if smuggled_data and 'force_videoid' in smuggled_data:
force_videoid = smuggled_data['force_videoid']
video_id = force_videoid
'upload_date': upload_date,
}
+ if not self._downloader.params.get('test', False) and not is_intentional:
+ self._downloader.report_warning('Falling back on generic information extractor.')
+
try:
webpage = self._download_webpage(url, video_id)
except ValueError:
# Helper method
def _playlist_from_matches(matches, getter, ie=None):
- urlrs = orderedSet(self.url_result(getter(m), ie) for m in matches)
+ urlrs = orderedSet(
+ self.url_result(self._proto_relative_url(getter(m)), ie)
+ for m in matches)
return self.playlist_result(
urlrs, playlist_id=video_id, playlist_title=video_title)
embedSWF\(?:\s*
)
(["\'])
- (?P<url>(?:https?:)?//(?:www\.)?youtube\.com/
- (?:embed|v)/.+?)
+ (?P<url>(?:https?:)?//(?:www\.)?youtube(?:-nocookie)?\.com/
+ (?:embed|v|p)/.+?)
\1''', webpage)
if matches:
return _playlist_from_matches(
- matches, lambda m: unescapeHTML(m[1]), ie='Youtube')
+ matches, lambda m: unescapeHTML(m[1]))
# Look for embedded Dailymotion player
matches = re.findall(
'title': video_title,
'id': video_id,
}
+ match = re.search(r'(?:id=["\']wistia_|data-wistiaid=["\']|Wistia\.embed\(["\'])(?P<id>[^"\']+)', webpage)
+ if match:
+ return {
+ '_type': 'url_transparent',
+ 'url': 'http://fast.wistia.net/embed/iframe/{0:}'.format(match.group('id')),
+ 'ie_key': 'Wistia',
+ 'uploader': video_uploader,
+ 'title': video_title,
+ 'id': match.group('id')
+ }
# Look for embedded blip.tv player
mobj = re.search(r'<meta\s[^>]*https?://api\.blip\.tv/\w+/redirect/\w+/(\d+)', webpage)
if mobj is not None:
return self.url_result(mobj.group('url'), 'SBS')
+ mobj = re.search(
+ r'<iframe[^>]+?src=(["\'])(?P<url>https?://m\.mlb\.com/shared/video/embed/embed\.html\?.+?)\1',
+ webpage)
+ if mobj is not None:
+ return self.url_result(mobj.group('url'), 'MLB')
+
# Start with something easy: JW Player in SWFObject
found = re.findall(r'flashvars: [\'"](?:.*&)?file=(http[^\'"&]*)', webpage)
if not found:
\s*{[^}]+? ["']?clip["']?\s*:\s*\{\s*
["']?url["']?\s*:\s*["']([^"']+)["']
''', webpage)
- assert found
if not found:
# Try to find twitter cards info
found = re.findall(r'<meta (?:property|name)="twitter:player:stream" (?:content|value)="(.+?)"', webpage)
if m_video_type is not None:
def check_video(vurl):
vpath = compat_urlparse.urlparse(vurl).path
- return '.' in vpath and not vpath.endswith('.swf')
+ vext = determine_ext(vpath)
+ return '.' in vpath and vext not in ('swf', 'png', 'jpg')
found = list(filter(
check_video,
re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)))
if not found:
# HTML5 video
- found = re.findall(r'(?s)<video[^<]*(?:>.*?<source.*?)? src="([^"]+)"', webpage)
+ found = re.findall(r'(?s)<video[^<]*(?:>.*?<source[^>]+)? src="([^"]+)"', webpage)
if not found:
found = re.search(
r'(?i)<meta\s+(?=(?:[a-z-]+="[^"]+"\s+)*http-equiv="refresh")'
- r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'([^\']+)\'"',
+ r'(?:[a-z-]+="[^"]+"\s+)*?content="[0-9]{,2};url=\'?([^\'"]+)',
webpage)
if found:
new_url = found.group(1)