Update to ytdl-2021.02.10

Except: [archiveorg] Fix and improve extraction (5fc53690cbe6abb11941a3f4846b566a7472753e)
This commit is contained in:
pukkandan 2021-02-11 02:52:55 +05:30
parent 539d158c50
commit cc2db87805
8 changed files with 345 additions and 237 deletions

View file

@ -1,6 +1,5 @@
# Supported sites
- **1tv**: Первый канал
- **1up.com**
- **20min**
- **220.ro**
- **23video**
@ -394,6 +393,8 @@ # Supported sites
- **HungamaSong**
- **Hypem**
- **ign.com**
- **IGNArticle**
- **IGNVideo**
- **IHeartRadio**
- **iheartradio:podcast**
- **imdb**: Internet Movie Database trailers
@ -701,7 +702,6 @@ # Supported sites
- **parliamentlive.tv**: UK parliament videos
- **Patreon**
- **pbs**: Public Broadcasting Service (PBS) and member stations: PBS: Public Broadcasting Service, APT - Alabama Public Television (WBIQ), GPB/Georgia Public Broadcasting (WGTV), Mississippi Public Broadcasting (WMPN), Nashville Public Television (WNPT), WFSU-TV (WFSU), WSRE (WSRE), WTCI (WTCI), WPBA/Channel 30 (WPBA), Alaska Public Media (KAKM), Arizona PBS (KAET), KNME-TV/Channel 5 (KNME), Vegas PBS (KLVX), AETN/ARKANSAS ETV NETWORK (KETS), KET (WKLE), WKNO/Channel 10 (WKNO), LPB/LOUISIANA PUBLIC BROADCASTING (WLPB), OETA (KETA), Ozarks Public Television (KOZK), WSIU Public Broadcasting (WSIU), KEET TV (KEET), KIXE/Channel 9 (KIXE), KPBS San Diego (KPBS), KQED (KQED), KVIE Public Television (KVIE), PBS SoCal/KOCE (KOCE), ValleyPBS (KVPT), CONNECTICUT PUBLIC TELEVISION (WEDH), KNPB Channel 5 (KNPB), SOPTV (KSYS), Rocky Mountain PBS (KRMA), KENW-TV3 (KENW), KUED Channel 7 (KUED), Wyoming PBS (KCWC), Colorado Public Television / KBDI 12 (KBDI), KBYU-TV (KBYU), Thirteen/WNET New York (WNET), WGBH/Channel 2 (WGBH), WGBY (WGBY), NJTV Public Media NJ (WNJT), WLIW21 (WLIW), mpt/Maryland Public Television (WMPB), WETA Television and Radio (WETA), WHYY (WHYY), PBS 39 (WLVT), WVPT - Your Source for PBS and More! (WVPT), Howard University Television (WHUT), WEDU PBS (WEDU), WGCU Public Media (WGCU), WPBT2 (WPBT), WUCF TV (WUCF), WUFT/Channel 5 (WUFT), WXEL/Channel 42 (WXEL), WLRN/Channel 17 (WLRN), WUSF Public Broadcasting (WUSF), ETV (WRLK), UNC-TV (WUNC), PBS Hawaii - Oceanic Cable Channel 10 (KHET), Idaho Public Television (KAID), KSPS (KSPS), OPB (KOPB), KWSU/Channel 10 & KTNW/Channel 31 (KWSU), WILL-TV (WILL), Network Knowledge - WSEC/Springfield (WSEC), WTTW11 (WTTW), Iowa Public Television/IPTV (KDIN), Nine Network (KETC), PBS39 Fort Wayne (WFWA), WFYI Indianapolis (WFYI), Milwaukee Public Television (WMVS), WNIN (WNIN), WNIT Public Television (WNIT), WPT (WPNE), WVUT/Channel 22 (WVUT), WEIU/Channel 51 (WEIU), WQPT-TV (WQPT), WYCC PBS Chicago (WYCC), WIPB-TV (WIPB), WTIU (WTIU), CET (WCET), ThinkTVNetwork (WPTD), WBGU-TV (WBGU), WGVU TV (WGVU), NET1 (KUON), Pioneer Public Television (KWCM), SDPB Television (KUSD), TPT (KTCA), KSMQ (KSMQ), KPTS/Channel 8 (KPTS), KTWU/Channel 11 (KTWU), East Tennessee PBS (WSJK), WCTE-TV (WCTE), WLJT, Channel 11 (WLJT), WOSU TV (WOSU), WOUB/WOUC (WOUB), WVPB (WVPB), WKYU-PBS (WKYU), KERA 13 (KERA), MPBN (WCBB), Mountain Lake PBS (WCFE), NHPTV (WENH), Vermont PBS (WETK), witf (WITF), WQED Multimedia (WQED), WMHT Educational Telecommunications (WMHT), Q-TV (WDCQ), WTVS Detroit Public TV (WTVS), CMU Public Television (WCMU), WKAR-TV (WKAR), WNMU-TV Public TV 13 (WNMU), WDSE - WRPT (WDSE), WGTE TV (WGTE), Lakeland Public Television (KAWE), KMOS-TV - Channels 6.1, 6.2 and 6.3 (KMOS), MontanaPBS (KUSM), KRWG/Channel 22 (KRWG), KACV (KACV), KCOS/Channel 13 (KCOS), WCNY/Channel 24 (WCNY), WNED (WNED), WPBS (WPBS), WSKG Public TV (WSKG), WXXI (WXXI), WPSU (WPSU), WVIA Public Media Studios (WVIA), WTVI (WTVI), Western Reserve PBS (WNEO), WVIZ/PBS ideastream (WVIZ), KCTS 9 (KCTS), Basin PBS (KPBT), KUHT / Channel 8 (KUHT), KLRN (KLRN), KLRU (KLRU), WTJX Channel 12 (WTJX), WCVE PBS (WCVE), KBTC Public Television (KBTC)
- **pcmag**
- **PearVideo**
- **PeerTube**
- **People**

View file

@ -19,55 +19,46 @@
_TESTS = [
(
'https://s.ytimg.com/yts/jsbin/html5player-vflHOr_nV.js',
'js',
86,
'>=<;:/.-[+*)(\'&%$#"!ZYX0VUTSRQPONMLKJIHGFEDCBA\\yxwvutsrqponmlkjihgfedcba987654321',
),
(
'https://s.ytimg.com/yts/jsbin/html5player-vfldJ8xgI.js',
'js',
85,
'3456789a0cdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRS[UVWXYZ!"#$%&\'()*+,-./:;<=>?@',
),
(
'https://s.ytimg.com/yts/jsbin/html5player-vfle-mVwz.js',
'js',
90,
']\\[@?>=<;:/.-,+*)(\'&%$#"hZYXWVUTSRQPONMLKJIHGFEDCBAzyxwvutsrqponmlkjiagfedcb39876',
),
(
'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl0Cbn9e.js',
'js',
84,
'O1I3456789abcde0ghijklmnopqrstuvwxyzABCDEFGHfJKLMN2PQRSTUVW@YZ!"#$%&\'()*+,-./:;<=',
),
(
'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js',
'js',
'2ACFC7A61CA478CD21425E5A57EBD73DDC78E22A.2094302436B2D377D14A3BBA23022D023B8BC25AA',
'A52CB8B320D22032ABB3A41D773D2B6342034902.A22E87CDD37DBE75A5E52412DC874AC16A7CFCA2',
),
(
'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflBb0OQx.js',
'js',
84,
'123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQ0STUVWXYZ!"#$%&\'()*+,@./:;<=>'
),
(
'https://s.ytimg.com/yts/jsbin/html5player-en_US-vfl9FYC6l.js',
'js',
83,
'123456789abcdefghijklmnopqr0tuvwxyzABCDETGHIJKLMNOPQRS>UVWXYZ!"#$%&\'()*+,-./:;<=F'
),
(
'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflCGk6yw/html5player.js',
'js',
'4646B5181C6C3020DF1D9C7FCFEA.AD80ABF70C39BD369CCCAE780AFBB98FA6B6CB42766249D9488C288',
'82C8849D94266724DC6B6AF89BBFA087EACCD963.B93C07FBA084ACAEFCF7C9D1FD0203C6C1815B6B'
),
(
'https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js',
'js',
'312AA52209E3623129A412D56A40F11CB0AF14AE.3EE09501CB14E3BCDC3B2AE808BF3F1D14E7FBF12',
'112AA5220913623229A412D56A40F11CB0AF14AE.3EE0950FCB14EEBCDC3B2AE808BF331D14E7FBF3',
)
@ -78,6 +69,10 @@ class TestPlayerInfo(unittest.TestCase):
def test_youtube_extract_player_info(self):
PLAYER_URLS = (
('https://www.youtube.com/s/player/64dddad9/player_ias.vflset/en_US/base.js', '64dddad9'),
('https://www.youtube.com/s/player/64dddad9/player_ias.vflset/fr_FR/base.js', '64dddad9'),
('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'),
('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-de_DE.vflset/base.js', '64dddad9'),
('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-tablet-en_US.vflset/base.js', '64dddad9'),
# obsolete
('https://www.youtube.com/yts/jsbin/player_ias-vfle4-e03/en_US/base.js', 'vfle4-e03'),
('https://www.youtube.com/yts/jsbin/player_ias-vfl49f_g4/en_US/base.js', 'vfl49f_g4'),
@ -100,13 +95,13 @@ def setUp(self):
os.mkdir(self.TESTDATA_DIR)
def make_tfunc(url, stype, sig_input, expected_sig):
def make_tfunc(url, sig_input, expected_sig):
m = re.match(r'.*-([a-zA-Z0-9_-]+)(?:/watch_as3|/html5player)?\.[a-z]+$', url)
assert m, '%r should follow URL format' % url
test_id = m.group(1)
def test_func(self):
basename = 'player-%s.%s' % (test_id, stype)
basename = 'player-%s.js' % test_id
fn = os.path.join(self.TESTDATA_DIR, basename)
if not os.path.exists(fn):
@ -114,22 +109,16 @@ def test_func(self):
ydl = FakeYDL()
ie = YoutubeIE(ydl)
if stype == 'js':
with io.open(fn, encoding='utf-8') as testf:
jscode = testf.read()
func = ie._parse_sig_js(jscode)
else:
assert stype == 'swf'
with open(fn, 'rb') as testf:
swfcode = testf.read()
func = ie._parse_sig_swf(swfcode)
src_sig = (
compat_str(string.printable[:sig_input])
if isinstance(sig_input, int) else sig_input)
got_sig = func(src_sig)
self.assertEqual(got_sig, expected_sig)
test_func.__name__ = str('test_signature_' + stype + '_' + test_id)
test_func.__name__ = str('test_signature_js_' + test_id)
setattr(TestSignature, test_func.__name__, test_func)

View file

@ -95,6 +95,9 @@ def _real_extract(self, url):
if 'Ten film jest dostępny dla użytkowników premium' in webpage:
raise ExtractorError('This video is only available for premium users.', expected=True)
if re.search(r'niedostępn[ey] w(?:&nbsp;|\s+)Twoim kraju\s*<', webpage):
self.raise_geo_restricted()
need_confirm_age = False
if self._html_search_regex(r'(<form[^>]+action="[^"]*/a/validatebirth[^"]*")',
webpage, 'birthday validate form', default=None):

View file

@ -502,8 +502,8 @@
from .hypem import HypemIE
from .ign import (
IGNIE,
OneUPIE,
PCMagIE,
IGNVideoIE,
IGNArticleIE,
)
from .iheart import (
IHeartRadioIE,

View file

@ -3,28 +3,39 @@
import re
from .common import InfoExtractor
from ..compat import (
compat_parse_qs,
compat_urllib_parse_urlparse,
)
from ..utils import (
HEADRequest,
determine_ext,
int_or_none,
parse_iso8601,
strip_or_none,
try_get,
)
class IGNIE(InfoExtractor):
class IGNBaseIE(InfoExtractor):
def _call_api(self, slug):
return self._download_json(
'http://apis.ign.com/{0}/v3/{0}s/slug/{1}'.format(self._PAGE_TYPE, slug), slug)
class IGNIE(IGNBaseIE):
"""
Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com.
Some videos of it.ign.com are also supported
"""
_VALID_URL = r'https?://.+?\.ign\.com/(?:[^/]+/)?(?P<type>videos|show_videos|articles|feature|(?:[^/]+/\d+/video))(/.+)?/(?P<name_or_id>.+)'
_VALID_URL = r'https?://(?:.+?\.ign|www\.pcmag)\.com/videos/(?:\d{4}/\d{2}/\d{2}/)?(?P<id>[^/?&#]+)'
IE_NAME = 'ign.com'
_PAGE_TYPE = 'video'
_API_URL_TEMPLATE = 'http://apis.ign.com/video/v3/videos/%s'
_EMBED_RE = r'<iframe[^>]+?["\']((?:https?:)?//.+?\.ign\.com.+?/embed.+?)["\']'
_TESTS = [
{
_TESTS = [{
'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review',
'md5': 'febda82c4bafecd2d44b6e1a18a595f8',
'md5': 'd2e1586d9987d40fad7867bf96a018ea',
'info_dict': {
'id': '8f862beef863986b2785559b9e1aa599',
'ext': 'mp4',
@ -32,13 +43,147 @@ class IGNIE(InfoExtractor):
'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c',
'timestamp': 1370440800,
'upload_date': '20130605',
'uploader_id': 'cberidon@ign.com',
'tags': 'count:9',
}
},
{
}, {
'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data',
'md5': 'f1581a6fe8c5121be5b807684aeac3f6',
'info_dict': {
'id': 'ee10d774b508c9b8ec07e763b9125b91',
'ext': 'mp4',
'title': 'What\'s New Now: Is GoGo Snooping on Your Data?',
'description': 'md5:817a20299de610bd56f13175386da6fa',
'timestamp': 1420571160,
'upload_date': '20150106',
'tags': 'count:4',
}
}, {
'url': 'https://www.ign.com/videos/is-a-resident-evil-4-remake-on-the-way-ign-daily-fix',
'only_matching': True,
}]
def _real_extract(self, url):
display_id = self._match_id(url)
video = self._call_api(display_id)
video_id = video['videoId']
metadata = video['metadata']
title = metadata.get('longTitle') or metadata.get('title') or metadata['name']
formats = []
refs = video.get('refs') or {}
m3u8_url = refs.get('m3uUrl')
if m3u8_url:
formats.extend(self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False))
f4m_url = refs.get('f4mUrl')
if f4m_url:
formats.extend(self._extract_f4m_formats(
f4m_url, video_id, f4m_id='hds', fatal=False))
for asset in (video.get('assets') or []):
asset_url = asset.get('url')
if not asset_url:
continue
formats.append({
'url': asset_url,
'tbr': int_or_none(asset.get('bitrate'), 1000),
'fps': int_or_none(asset.get('frame_rate')),
'height': int_or_none(asset.get('height')),
'width': int_or_none(asset.get('width')),
})
mezzanine_url = try_get(video, lambda x: x['system']['mezzanineUrl'])
if mezzanine_url:
formats.append({
'ext': determine_ext(mezzanine_url, 'mp4'),
'format_id': 'mezzanine',
'preference': 1,
'url': mezzanine_url,
})
self._sort_formats(formats)
thumbnails = []
for thumbnail in (video.get('thumbnails') or []):
thumbnail_url = thumbnail.get('url')
if not thumbnail_url:
continue
thumbnails.append({
'url': thumbnail_url,
})
tags = []
for tag in (video.get('tags') or []):
display_name = tag.get('displayName')
if not display_name:
continue
tags.append(display_name)
return {
'id': video_id,
'title': title,
'description': strip_or_none(metadata.get('description')),
'timestamp': parse_iso8601(metadata.get('publishDate')),
'duration': int_or_none(metadata.get('duration')),
'display_id': display_id,
'thumbnails': thumbnails,
'formats': formats,
'tags': tags,
}
class IGNVideoIE(InfoExtractor):
_VALID_URL = r'https?://.+?\.ign\.com/(?:[a-z]{2}/)?[^/]+/(?P<id>\d+)/(?:video|trailer)/'
_TESTS = [{
'url': 'http://me.ign.com/en/videos/112203/video/how-hitman-aims-to-be-different-than-every-other-s',
'md5': 'dd9aca7ed2657c4e118d8b261e5e9de1',
'info_dict': {
'id': 'e9be7ea899a9bbfc0674accc22a36cc8',
'ext': 'mp4',
'title': 'How Hitman Aims to Be Different Than Every Other Stealth Game - NYCC 2015',
'description': 'Taking out assassination targets in Hitman has never been more stylish.',
'timestamp': 1444665600,
'upload_date': '20151012',
}
}, {
'url': 'http://me.ign.com/ar/angry-birds-2/106533/video/lrd-ldyy-lwl-lfylm-angry-birds',
'only_matching': True,
}, {
# Youtube embed
'url': 'https://me.ign.com/ar/ratchet-clank-rift-apart/144327/trailer/embed',
'only_matching': True,
}, {
# Twitter embed
'url': 'http://adria.ign.com/sherlock-season-4/9687/trailer/embed',
'only_matching': True,
}, {
# Vimeo embed
'url': 'https://kr.ign.com/bic-2018/3307/trailer/embed',
'only_matching': True,
}]
def _real_extract(self, url):
video_id = self._match_id(url)
req = HEADRequest(url.rsplit('/', 1)[0] + '/embed')
url = self._request_webpage(req, video_id).geturl()
ign_url = compat_parse_qs(
compat_urllib_parse_urlparse(url).query).get('url', [None])[0]
if ign_url:
return self.url_result(ign_url, IGNIE.ie_key())
return self.url_result(url)
class IGNArticleIE(IGNBaseIE):
_VALID_URL = r'https?://.+?\.ign\.com/(?:articles(?:/\d{4}/\d{2}/\d{2})?|(?:[a-z]{2}/)?feature/\d+)/(?P<id>[^/?&#]+)'
_PAGE_TYPE = 'article'
_TESTS = [{
'url': 'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind',
'info_dict': {
'id': '100-little-things-in-gta-5-that-will-blow-your-mind',
'id': '524497489e4e8ff5848ece34',
'title': '100 Little Things in GTA 5 That Will Blow Your Mind',
},
'playlist': [
{
@ -49,7 +194,6 @@ class IGNIE(InfoExtractor):
'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.',
'timestamp': 1379339880,
'upload_date': '20130916',
'uploader_id': 'danieljkrupa@gmail.com',
},
},
{
@ -60,173 +204,54 @@ class IGNIE(InfoExtractor):
'description': 'The twisted beauty of GTA 5 in stunning slow motion.',
'timestamp': 1386878820,
'upload_date': '20131212',
'uploader_id': 'togilvie@ign.com',
},
},
],
'params': {
'playlist_items': '2-3',
'skip_download': True,
},
},
{
}, {
'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch',
'md5': '618fedb9c901fd086f6f093564ef8558',
'info_dict': {
'id': '078fdd005f6d3c02f63d795faa1b984f',
'ext': 'mp4',
'id': '53ee806780a81ec46e0790f8',
'title': 'Rewind Theater - Wild Trailer Gamescom 2014',
'description': 'Brian and Jared explore Michel Ancel\'s captivating new preview.',
'timestamp': 1408047180,
'upload_date': '20140814',
'uploader_id': 'jamesduggan1990@gmail.com',
},
},
{
'url': 'http://me.ign.com/en/videos/112203/video/how-hitman-aims-to-be-different-than-every-other-s',
'only_matching': True,
},
{
'url': 'http://me.ign.com/ar/angry-birds-2/106533/video/lrd-ldyy-lwl-lfylm-angry-birds',
'only_matching': True,
},
{
'playlist_count': 2,
}, {
# videoId pattern
'url': 'http://www.ign.com/articles/2017/06/08/new-ducktales-short-donalds-birthday-doesnt-go-as-planned',
'only_matching': True,
},
]
def _find_video_id(self, webpage):
res_id = [
r'"video_id"\s*:\s*"(.*?)"',
r'class="hero-poster[^"]*?"[^>]*id="(.+?)"',
r'data-video-id="(.+?)"',
r'<object id="vid_(.+?)"',
r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"',
r'videoId&quot;\s*:\s*&quot;(.+?)&quot;',
r'videoId["\']\s*:\s*["\']([^"\']+?)["\']',
]
return self._search_regex(res_id, webpage, 'video id', default=None)
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
name_or_id = mobj.group('name_or_id')
page_type = mobj.group('type')
webpage = self._download_webpage(url, name_or_id)
if page_type != 'video':
multiple_urls = re.findall(
r'<param name="flashvars"[^>]*value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]',
webpage)
if multiple_urls:
entries = [self.url_result(u, ie='IGN') for u in multiple_urls]
return {
'_type': 'playlist',
'id': name_or_id,
'entries': entries,
}
video_id = self._find_video_id(webpage)
if not video_id:
return self.url_result(self._search_regex(
self._EMBED_RE, webpage, 'embed url'))
return self._get_video_info(video_id)
def _get_video_info(self, video_id):
api_data = self._download_json(
self._API_URL_TEMPLATE % video_id, video_id)
formats = []
m3u8_url = api_data['refs'].get('m3uUrl')
if m3u8_url:
formats.extend(self._extract_m3u8_formats(
m3u8_url, video_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False))
f4m_url = api_data['refs'].get('f4mUrl')
if f4m_url:
formats.extend(self._extract_f4m_formats(
f4m_url, video_id, f4m_id='hds', fatal=False))
for asset in api_data['assets']:
formats.append({
'url': asset['url'],
'tbr': asset.get('actual_bitrate_kbps'),
'fps': asset.get('frame_rate'),
'height': int_or_none(asset.get('height')),
'width': int_or_none(asset.get('width')),
})
self._sort_formats(formats)
thumbnails = [{
'url': thumbnail['url']
} for thumbnail in api_data.get('thumbnails', [])]
metadata = api_data['metadata']
return {
'id': api_data.get('videoId') or video_id,
'title': metadata.get('longTitle') or metadata.get('name') or metadata.get['title'],
'description': metadata.get('description'),
'timestamp': parse_iso8601(metadata.get('publishDate')),
'duration': int_or_none(metadata.get('duration')),
'display_id': metadata.get('slug') or video_id,
'uploader_id': metadata.get('creator'),
'thumbnails': thumbnails,
'formats': formats,
}
class OneUPIE(IGNIE):
_VALID_URL = r'https?://gamevideos\.1up\.com/(?P<type>video)/id/(?P<name_or_id>.+)\.html'
IE_NAME = '1up.com'
_TESTS = [{
'url': 'http://gamevideos.1up.com/video/id/34976.html',
'md5': 'c9cc69e07acb675c31a16719f909e347',
'info_dict': {
'id': '34976',
'ext': 'mp4',
'title': 'Sniper Elite V2 - Trailer',
'description': 'md5:bf0516c5ee32a3217aa703e9b1bc7826',
'timestamp': 1313099220,
'upload_date': '20110811',
'uploader_id': 'IGN',
}
}]
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
result = super(OneUPIE, self)._real_extract(url)
result['id'] = mobj.group('name_or_id')
return result
class PCMagIE(IGNIE):
_VALID_URL = r'https?://(?:www\.)?pcmag\.com/(?P<type>videos|article2)(/.+)?/(?P<name_or_id>.+)'
IE_NAME = 'pcmag'
_EMBED_RE = r'iframe\.setAttribute\("src",\s*__util.objToUrlString\("http://widgets\.ign\.com/video/embed/content\.html?[^"]*url=([^"]+)["&]'
_TESTS = [{
'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data',
'md5': '212d6154fd0361a2781075f1febbe9ad',
'info_dict': {
'id': 'ee10d774b508c9b8ec07e763b9125b91',
'ext': 'mp4',
'title': '010615_What\'s New Now: Is GoGo Snooping on Your Data?',
'description': 'md5:a7071ae64d2f68cc821c729d4ded6bb3',
'timestamp': 1420571160,
'upload_date': '20150106',
'uploader_id': 'cozzipix@gmail.com',
}
}, {
'url': 'http://www.pcmag.com/article2/0,2817,2470156,00.asp',
'md5': '94130c1ca07ba0adb6088350681f16c1',
'info_dict': {
'id': '042e560ba94823d43afcb12ddf7142ca',
'ext': 'mp4',
'title': 'HTC\'s Weird New Re Camera - What\'s New Now',
'description': 'md5:53433c45df96d2ea5d0fda18be2ca908',
'timestamp': 1412953920,
'upload_date': '20141010',
'uploader_id': 'chris_snyder@pcmag.com',
}
# Youtube embed
'url': 'https://www.ign.com/articles/2021-mvp-named-in-puppy-bowl-xvii',
'only_matching': True,
}, {
# IMDB embed
'url': 'https://www.ign.com/articles/2014/08/07/sons-of-anarchy-final-season-trailer',
'only_matching': True,
}, {
# Facebook embed
'url': 'https://www.ign.com/articles/2017/09/20/marvels-the-punisher-watch-the-new-trailer-for-the-netflix-series',
'only_matching': True,
}, {
# Brightcove embed
'url': 'https://www.ign.com/articles/2016/01/16/supergirl-goes-flying-with-martian-manhunter-in-new-clip',
'only_matching': True,
}]
def _real_extract(self, url):
display_id = self._match_id(url)
article = self._call_api(display_id)
def entries():
media_url = try_get(article, lambda x: x['mediaRelations'][0]['media']['metadata']['url'])
if media_url:
yield self.url_result(media_url, IGNIE.ie_key())
for content in (article.get('content') or []):
for video_url in re.findall(r'(?:\[(?:ignvideo\s+url|youtube\s+clip_id)|<iframe[^>]+src)="([^"]+)"', content):
yield self.url_result(video_url)
return self.playlist_result(
entries(), article.get('articleId'),
strip_or_none(try_get(article, lambda x: x['metadata']['headline'])))

View file

@ -42,8 +42,8 @@ def _real_extract(self, url):
url = url.replace('skola.se/Produkter', 'play.se/program')
webpage = self._download_webpage(url, video_id)
urplayer_data = self._parse_json(self._html_search_regex(
r'data-react-class="components/Player/Player"[^>]+data-react-props="({.+?})"',
webpage, 'urplayer data'), video_id)['currentProduct']
r'data-react-class="routes/Product/components/ProgramContainer/ProgramContainer"[^>]+data-react-props="({.+?})"',
webpage, 'urplayer data'), video_id)['accessibleEpisodes'][0]
episode = urplayer_data['title']
host = self._download_json('http://streaming-loadbalancer.ur.se/loadbalancer.json', video_id)['redirect']

View file

@ -11,11 +11,14 @@
dict_get,
extract_attributes,
ExtractorError,
float_or_none,
int_or_none,
parse_duration,
str_or_none,
try_get,
unified_strdate,
url_or_none,
urljoin,
)
@ -146,36 +149,89 @@ def get_height(s):
video = initials['videoModel']
title = video['title']
formats = []
for format_id, formats_dict in video['sources'].items():
format_urls = set()
format_sizes = {}
sources = try_get(video, lambda x: x['sources'], dict) or {}
for format_id, formats_dict in sources.items():
if not isinstance(formats_dict, dict):
continue
download_sources = try_get(sources, lambda x: x['download'], dict) or {}
for quality, format_dict in download_sources.items():
if not isinstance(format_dict, dict):
continue
format_sizes[quality] = float_or_none(format_dict.get('size'))
for quality, format_item in formats_dict.items():
if format_id == 'download':
# Download link takes some time to be generated,
# skipping for now
continue
if not isinstance(format_item, dict):
continue
format_url = format_item.get('link')
filesize = int_or_none(
format_item.get('size'), invscale=1000000)
else:
format_url = format_item
filesize = None
format_url = url_or_none(format_url)
if not format_url:
if not format_url or format_url in format_urls:
continue
format_urls.add(format_url)
formats.append({
'format_id': '%s-%s' % (format_id, quality),
'url': format_url,
'ext': determine_ext(format_url, 'mp4'),
'height': get_height(quality),
'filesize': filesize,
'filesize': format_sizes.get(quality),
'http_headers': {
'Referer': urlh.geturl(),
},
})
self._sort_formats(formats)
xplayer_sources = try_get(
initials, lambda x: x['xplayerSettings']['sources'], dict)
if xplayer_sources:
hls_sources = xplayer_sources.get('hls')
if isinstance(hls_sources, dict):
for hls_format_key in ('url', 'fallback'):
hls_url = hls_sources.get(hls_format_key)
if not hls_url:
continue
hls_url = urljoin(url, hls_url)
if not hls_url or hls_url in format_urls:
continue
format_urls.add(hls_url)
formats.extend(self._extract_m3u8_formats(
hls_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
standard_sources = xplayer_sources.get('standard')
if isinstance(standard_sources, dict):
for format_id, formats_list in standard_sources.items():
if not isinstance(formats_list, list):
continue
for standard_format in formats_list:
if not isinstance(standard_format, dict):
continue
for standard_format_key in ('url', 'fallback'):
standard_url = standard_format.get(standard_format_key)
if not standard_url:
continue
standard_url = urljoin(url, standard_url)
if not standard_url or standard_url in format_urls:
continue
format_urls.add(standard_url)
ext = determine_ext(standard_url, 'mp4')
if ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
standard_url, video_id, 'mp4', entry_protocol='m3u8_native',
m3u8_id='hls', fatal=False))
continue
quality = (str_or_none(standard_format.get('quality'))
or str_or_none(standard_format.get('label'))
or '')
formats.append({
'format_id': '%s-%s' % (format_id, quality),
'url': standard_url,
'ext': ext,
'height': get_height(quality),
'filesize': format_sizes.get(quality),
'http_headers': {
'Referer': standard_url,
},
})
self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id'))
categories_list = video.get('categories')
if isinstance(categories_list, list):

View file

@ -32,7 +32,7 @@
mimetype2ext,
parse_codecs,
parse_duration,
# qualities,
# qualities, # TODO: Enable this after fixing formatSort
remove_start,
smuggle_url,
str_or_none,
@ -414,7 +414,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
(?(1).+)? # if we found the ID, everything can follow
$""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE}
_PLAYER_INFO_RE = (
r'/(?P<id>[a-zA-Z0-9_-]{8,})/player_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?/base\.js$',
r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
)
_formats = {
@ -621,6 +622,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'uploader': 'AfrojackVEVO',
'uploader_id': 'AfrojackVEVO',
'upload_date': '20131011',
'abr': 129.495,
},
'params': {
'youtube_include_dash_manifest': True,
@ -1134,10 +1136,26 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'only_matching': True,
},
{
# Age-gated video only available with authentication (unavailable
# via embed page workaround)
'url': 'XgnwCQzjau8',
'only_matching': True,
# https://github.com/ytdl-org/youtube-dl/pull/28094
'url': 'OtqTfy26tG0',
'info_dict': {
'id': 'OtqTfy26tG0',
'ext': 'mp4',
'title': 'Burn Out',
'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131',
'upload_date': '20141120',
'uploader': 'The Cinematic Orchestra - Topic',
'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw',
'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw',
'artist': 'The Cinematic Orchestra',
'track': 'Burn Out',
'album': 'Every Day',
'release_data': None,
'release_year': None,
},
'params': {
'skip_download': True,
},
},
]
@ -1230,6 +1248,9 @@ def _parse_sig_js(self, jscode):
funcname = self._search_regex(
(r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\bm=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(h\.s\)\)',
r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2})\(decodeURIComponent\(c\)\)',
r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)',
r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
# Obsolete patterns
@ -1493,7 +1514,9 @@ def feed_entry(name):
formats = []
itags = []
itag_qualities = {}
player_url = None
# TODO: Enable this after fixing formatSort
# q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'])
streaming_data = player_response.get('streamingData') or {}
streaming_formats = streaming_data.get('formats') or []
@ -1502,6 +1525,16 @@ def feed_entry(name):
if fmt.get('targetDurationSec') or fmt.get('drmFamilies'):
continue
itag = str_or_none(fmt.get('itag'))
quality = fmt.get('quality')
if itag and quality:
itag_qualities[itag] = quality
# FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment
# (adding `&sq=0` to the URL) and parsing emsg box to determine the
# number of fragment that would subsequently requested with (`&sq=N`)
if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF':
continue
fmt_url = fmt.get('url')
if not fmt_url:
sc = compat_parse_qs(fmt.get('signatureCipher'))
@ -1521,10 +1554,10 @@ def feed_entry(name):
sp = try_get(sc, lambda x: x['sp'][0]) or 'signature'
fmt_url += '&' + sp + '=' + signature
itag = str_or_none(fmt.get('itag'))
if itag:
itags.append(itag)
quality = fmt.get('quality')
tbr = float_or_none(
fmt.get('averageBitrate') or fmt.get('bitrate'), 1000)
dct = {
'asr': int_or_none(fmt.get('audioSampleRate')),
'filesize': int_or_none(fmt.get('contentLength')),
@ -1532,9 +1565,8 @@ def feed_entry(name):
'format_note': fmt.get('qualityLabel') or quality,
'fps': int_or_none(fmt.get('fps')),
'height': int_or_none(fmt.get('height')),
# 'quality': q(quality), # This does not correctly reflect the overall quality of the format
'tbr': float_or_none(fmt.get(
'averageBitrate') or fmt.get('bitrate'), 1000),
# 'quality': q(quality), # TODO: Enable this after fixing formatSort
'tbr': tbr,
'url': fmt_url,
'width': fmt.get('width'),
}
@ -1545,7 +1577,13 @@ def feed_entry(name):
if mobj:
dct['ext'] = mimetype2ext(mobj.group(1))
dct.update(parse_codecs(mobj.group(2)))
if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none':
no_audio = dct.get('acodec') == 'none'
no_video = dct.get('vcodec') == 'none'
if no_audio:
dct['vbr'] = tbr
if no_video:
dct['abr'] = tbr
if no_audio or no_video:
dct['downloader_options'] = {
# Youtube throttles chunks >~10M
'http_chunk_size': 10485760,
@ -1565,22 +1603,19 @@ def feed_entry(name):
if self._downloader.params.get('youtube_include_dash_manifest'):
dash_manifest_url = streaming_data.get('dashManifestUrl')
if dash_manifest_url:
dash_formats = []
for f in self._extract_mpd_formats(
dash_manifest_url, video_id, fatal=False):
itag = f['format_id']
if itag in itags:
continue
# if itag in itag_qualities: # TODO: Enable this after fixing formatSort
# f['quality'] = q(itag_qualities[itag])
filesize = int_or_none(self._search_regex(
r'/clen/(\d+)', f.get('fragment_base_url')
or f['url'], 'file size', default=None))
if filesize:
f['filesize'] = filesize
dash_formats.append(f)
# Until further investigation prefer DASH formats as non-DASH
# may not be available (see [1])
# 1. https://github.com/ytdl-org/youtube-dl/issues/28070
if dash_formats:
dash_formats_keys = [f['format_id'] for f in dash_formats]
formats = [f for f in formats if f['format_id'] not in dash_formats_keys]
formats.extend(dash_formats)
formats.append(f)
if not formats:
if streaming_data.get('licenseInfos'):
@ -1747,7 +1782,7 @@ def process_language(container, base_url, lang_code, query):
'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')),
'track': mobj.group('track').strip(),
'release_date': release_date,
'release_year': int(release_year),
'release_year': int_or_none(release_year),
})
initial_data = None
@ -2597,9 +2632,9 @@ def _extract_continuation(cls, renderer):
next_continuation = cls._extract_next_continuation_data(renderer)
if next_continuation:
return next_continuation
contents = renderer.get('contents') or renderer.get('items')
if not isinstance(contents, list):
return
contents = []
for key in ('contents', 'items'):
contents.extend(try_get(renderer, lambda x: x[key], list) or [])
for content in contents:
if not isinstance(content, dict):
continue