aboutsummaryrefslogtreecommitdiff
path: root/youtube_dl/extractor/generic.py
diff options
context:
space:
mode:
Diffstat (limited to 'youtube_dl/extractor/generic.py')
-rw-r--r--youtube_dl/extractor/generic.py233
1 files changed, 150 insertions, 83 deletions
diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py
index 9a7b0d25d..f6b984300 100644
--- a/youtube_dl/extractor/generic.py
+++ b/youtube_dl/extractor/generic.py
@@ -9,6 +9,8 @@ from .common import InfoExtractor
from .youtube import YoutubeIE
from ..compat import (
compat_urllib_parse,
+ compat_urllib_parse_unquote,
+ compat_urllib_request,
compat_urlparse,
compat_xml_parse_error,
)
@@ -32,6 +34,7 @@ from .brightcove import BrightcoveIE
from .nbc import NBCSportsVPlayerIE
from .ooyala import OoyalaIE
from .rutv import RUTVIE
+from .tvc import TVCIE
from .sportbox import SportBoxEmbedIE
from .smotri import SmotriIE
from .condenast import CondeNastIE
@@ -39,6 +42,7 @@ from .udn import UDNEmbedIE
from .senateisvp import SenateISVPIE
from .bliptv import BlipTVIE
from .svt import SVTIE
+from .pornhub import PornHubIE
class GenericIE(InfoExtractor):
@@ -46,6 +50,97 @@ class GenericIE(InfoExtractor):
_VALID_URL = r'.*'
IE_NAME = 'generic'
_TESTS = [
+ # Direct link to a video
+ {
+ 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
+ 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
+ 'info_dict': {
+ 'id': 'trailer',
+ 'ext': 'mp4',
+ 'title': 'trailer',
+ 'upload_date': '20100513',
+ }
+ },
+ # Direct link to media delivered compressed (until Accept-Encoding is *)
+ {
+ 'url': 'http://calimero.tk/muzik/FictionJunction-Parallel_Hearts.flac',
+ 'md5': '128c42e68b13950268b648275386fc74',
+ 'info_dict': {
+ 'id': 'FictionJunction-Parallel_Hearts',
+ 'ext': 'flac',
+ 'title': 'FictionJunction-Parallel_Hearts',
+ 'upload_date': '20140522',
+ },
+ 'expected_warnings': [
+ 'URL could be a direct video link, returning it as such.'
+ ]
+ },
+ # Direct download with broken HEAD
+ {
+ 'url': 'http://ai-radio.org:8000/radio.opus',
+ 'info_dict': {
+ 'id': 'radio',
+ 'ext': 'opus',
+ 'title': 'radio',
+ },
+ 'params': {
+ 'skip_download': True, # infinite live stream
+ },
+ 'expected_warnings': [
+ r'501.*Not Implemented'
+ ],
+ },
+ # Direct link with incorrect MIME type
+ {
+ 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
+ 'md5': '4ccbebe5f36706d85221f204d7eb5913',
+ 'info_dict': {
+ 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
+ 'id': '5_Lennart_Poettering_-_Systemd',
+ 'ext': 'webm',
+ 'title': '5_Lennart_Poettering_-_Systemd',
+ 'upload_date': '20141120',
+ },
+ 'expected_warnings': [
+ 'URL could be a direct video link, returning it as such.'
+ ]
+ },
+ # 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:.*groundbreaking video review series.*'
+ },
+ 'playlist_mincount': 11,
+ },
+ # RSS feed with enclosure
+ {
+ 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
+ 'info_dict': {
+ 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
+ 'ext': 'm4v',
+ 'upload_date': '20150228',
+ 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
+ }
+ },
+ # google redirect
+ {
+ 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
+ 'info_dict': {
+ 'id': 'cmQHVoWB5FY',
+ 'ext': 'mp4',
+ 'upload_date': '20130224',
+ 'uploader_id': 'TheVerge',
+ 'description': 're:^Chris Ziegler takes a look at the\.*',
+ 'uploader': 'The Verge',
+ 'title': 'First Firefox OS phones side-by-side',
+ },
+ 'params': {
+ 'skip_download': False,
+ }
+ },
{
'url': 'http://www.hodiho.fr/2013/02/regis-plante-sa-jeep.html',
'md5': '85b90ccc9d73b4acd9138d3af4c27f89',
@@ -125,17 +220,6 @@ class GenericIE(InfoExtractor):
'skip_download': True, # m3u8 download
},
},
- # Direct link to a video
- {
- 'url': 'http://media.w3.org/2010/05/sintel/trailer.mp4',
- 'md5': '67d406c2bcb6af27fa886f31aa934bbe',
- 'info_dict': {
- 'id': 'trailer',
- 'ext': 'mp4',
- 'title': 'trailer',
- 'upload_date': '20100513',
- }
- },
# ooyala video
{
'url': 'http://www.rollingstone.com/music/videos/norwegian-dj-cashmere-cat-goes-spartan-on-with-me-premiere-20131219',
@@ -160,22 +244,6 @@ class GenericIE(InfoExtractor):
},
'add_ie': ['Ooyala'],
},
- # google redirect
- {
- 'url': 'http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&cad=rja&ved=0CCUQtwIwAA&url=http%3A%2F%2Fwww.youtube.com%2Fwatch%3Fv%3DcmQHVoWB5FY&ei=F-sNU-LLCaXk4QT52ICQBQ&usg=AFQjCNEw4hL29zgOohLXvpJ-Bdh2bils1Q&bvm=bv.61965928,d.bGE',
- 'info_dict': {
- 'id': 'cmQHVoWB5FY',
- 'ext': 'mp4',
- 'upload_date': '20130224',
- 'uploader_id': 'TheVerge',
- 'description': 're:^Chris Ziegler takes a look at the\.*',
- 'uploader': 'The Verge',
- 'title': 'First Firefox OS phones side-by-side',
- },
- 'params': {
- 'skip_download': False,
- }
- },
# embed.ly video
{
'url': 'http://www.tested.com/science/weird/460206-tested-grinding-coffee-2000-frames-second/',
@@ -225,6 +293,15 @@ class GenericIE(InfoExtractor):
'skip_download': True,
},
},
+ # TVC embed
+ {
+ 'url': 'http://sch1298sz.mskobr.ru/dou_edu/karamel_ki/filial_galleries/video/iframe_src_http_tvc_ru_video_iframe_id_55304_isplay_false_acc_video_id_channel_brand_id_11_show_episodes_episode_id_32307_frameb/',
+ 'info_dict': {
+ 'id': '55304',
+ 'ext': 'mp4',
+ 'title': 'Дошкольное воспитание',
+ },
+ },
# SportBox embed
{
'url': 'http://www.vestifinance.ru/articles/25753',
@@ -407,16 +484,6 @@ class GenericIE(InfoExtractor):
'title': 'Busty Blonde Siri Tit Fuck While Wank at HandjobHub.com',
}
},
- # 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:.*groundbreaking video review series.*'
- },
- 'playlist_mincount': 11,
- },
# Multiple brightcove videos
# https://github.com/rg3/youtube-dl/issues/2283
{
@@ -470,21 +537,6 @@ class GenericIE(InfoExtractor):
'uploader': 'thoughtworks.wistia.com',
},
},
- # Direct download with broken HEAD
- {
- 'url': 'http://ai-radio.org:8000/radio.opus',
- 'info_dict': {
- 'id': 'radio',
- 'ext': 'opus',
- 'title': 'radio',
- },
- 'params': {
- 'skip_download': True, # infinite live stream
- },
- 'expected_warnings': [
- r'501.*Not Implemented'
- ],
- },
# Soundcloud embed
{
'url': 'http://nakedsecurity.sophos.com/2014/10/29/sscc-171-are-you-sure-that-1234-is-a-bad-password-podcast/',
@@ -516,21 +568,6 @@ class GenericIE(InfoExtractor):
},
'playlist_mincount': 2,
},
- # Direct link with incorrect MIME type
- {
- 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
- 'md5': '4ccbebe5f36706d85221f204d7eb5913',
- 'info_dict': {
- 'url': 'http://ftp.nluug.nl/video/nluug/2014-11-20_nj14/zaal-2/5_Lennart_Poettering_-_Systemd.webm',
- 'id': '5_Lennart_Poettering_-_Systemd',
- 'ext': 'webm',
- 'title': '5_Lennart_Poettering_-_Systemd',
- 'upload_date': '20141120',
- },
- 'expected_warnings': [
- 'URL could be a direct video link, returning it as such.'
- ]
- },
# Cinchcast embed
{
'url': 'http://undergroundwellness.com/podcasts/306-5-steps-to-permanent-gut-healing/',
@@ -689,16 +726,6 @@ class GenericIE(InfoExtractor):
'age_limit': 0,
},
},
- # RSS feed with enclosure
- {
- 'url': 'http://podcastfeeds.nbcnews.com/audio/podcast/MSNBC-MADDOW-NETCAST-M4V.xml',
- 'info_dict': {
- 'id': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
- 'ext': 'm4v',
- 'upload_date': '20150228',
- 'title': 'pdv_maddow_netcast_m4v-02-27-2015-201624',
- }
- },
# Crooks and Liars embed
{
'url': 'http://crooksandliars.com/2015/04/fox-friends-says-protecting-atheists',
@@ -773,6 +800,18 @@ class GenericIE(InfoExtractor):
# rtmpe downloads
'skip_download': True,
}
+ },
+ # Brightcove URL in single quotes
+ {
+ 'url': 'http://www.sportsnet.ca/baseball/mlb/sn-presents-russell-martin-world-citizen/',
+ 'md5': '4ae374f1f8b91c889c4b9203c8c752af',
+ 'info_dict': {
+ 'id': '4255764656001',
+ 'ext': 'mp4',
+ 'title': 'SN Presents: Russell Martin, World Citizen',
+ 'description': 'To understand why he was the Toronto Blue Jays’ top off-season priority is to appreciate his background and upbringing in Montreal, where he first developed his baseball skills. Written and narrated by Stephen Brunt.',
+ 'uploader': 'Rogers Sportsnet',
+ },
}
]
@@ -894,7 +933,7 @@ class GenericIE(InfoExtractor):
force_videoid = smuggled_data['force_videoid']
video_id = force_videoid
else:
- video_id = os.path.splitext(url.rstrip('/').split('/')[-1])[0]
+ video_id = compat_urllib_parse_unquote(os.path.splitext(url.rstrip('/').split('/')[-1])[0])
self.to_screen('%s: Requesting header' % video_id)
@@ -916,7 +955,9 @@ class GenericIE(InfoExtractor):
full_response = None
if head_response is False:
- full_response = self._request_webpage(url, video_id)
+ request = compat_urllib_request.Request(url)
+ request.add_header('Accept-Encoding', '*')
+ full_response = self._request_webpage(request, video_id)
head_response = full_response
# Check for direct link to a video
@@ -927,7 +968,7 @@ class GenericIE(InfoExtractor):
head_response.headers.get('Last-Modified'))
return {
'id': video_id,
- 'title': os.path.splitext(url_basename(url))[0],
+ 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
'direct': True,
'formats': [{
'format_id': m.group('format_id'),
@@ -941,7 +982,17 @@ class GenericIE(InfoExtractor):
self._downloader.report_warning('Falling back on generic information extractor.')
if not full_response:
- full_response = self._request_webpage(url, video_id)
+ request = compat_urllib_request.Request(url)
+ # Some webservers may serve compressed content of rather big size (e.g. gzipped flac)
+ # making it impossible to download only chunk of the file (yet we need only 512kB to
+ # test whether it's HTML or not). According to youtube-dl default Accept-Encoding
+ # that will always result in downloading the whole file that is not desirable.
+ # Therefore for extraction pass we have to override Accept-Encoding to any in order
+ # to accept raw bytes and being able to download only a chunk.
+ # It may probably better to solve this by checking Content-Type for application/octet-stream
+ # after HEAD request finishes, but not sure if we can rely on this.
+ request.add_header('Accept-Encoding', '*')
+ full_response = self._request_webpage(request, video_id)
# Maybe it's a direct link to a video?
# Be careful not to download the whole thing!
@@ -953,7 +1004,7 @@ class GenericIE(InfoExtractor):
head_response.headers.get('Last-Modified'))
return {
'id': video_id,
- 'title': os.path.splitext(url_basename(url))[0],
+ 'title': compat_urllib_parse_unquote(os.path.splitext(url_basename(url))[0]),
'direct': True,
'url': url,
'upload_date': upload_date,
@@ -1033,7 +1084,7 @@ class GenericIE(InfoExtractor):
# Look for embedded rtl.nl player
matches = re.findall(
- r'<iframe\s+(?:[a-zA-Z-]+="[^"]+"\s+)*?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+video_embed[^"]+)"',
+ r'<iframe[^>]+?src="((?:https?:)?//(?:www\.)?rtl\.nl/system/videoplayer/[^"]+(?:video_)?embed[^"]+)"',
webpage)
if matches:
return _playlist_from_matches(matches, ie='RtlNl')
@@ -1261,11 +1312,27 @@ class GenericIE(InfoExtractor):
if rutv_url:
return self.url_result(rutv_url, 'RUTV')
+ # Look for embedded TVC player
+ tvc_url = TVCIE._extract_url(webpage)
+ if tvc_url:
+ return self.url_result(tvc_url, 'TVC')
+
# Look for embedded SportBox player
sportbox_urls = SportBoxEmbedIE._extract_urls(webpage)
if sportbox_urls:
return _playlist_from_matches(sportbox_urls, ie='SportBoxEmbed')
+ # Look for embedded PornHub player
+ pornhub_url = PornHubIE._extract_url(webpage)
+ if pornhub_url:
+ return self.url_result(pornhub_url, 'PornHub')
+
+ # Look for embedded Tvigle player
+ mobj = re.search(
+ r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//cloud\.tvigle\.ru/video/.+?)\1', webpage)
+ if mobj is not None:
+ return self.url_result(mobj.group('url'), 'Tvigle')
+
# Look for embedded TED player
mobj = re.search(
r'<iframe[^>]+?src=(["\'])(?P<url>https?://embed(?:-ssl)?\.ted\.com/.+?)\1', webpage)
generated by cgit v1.2.3 (git 2.26.2) at 2024-10-05 17:46:39 +0000