Compare commits

...

2 Commits

Author SHA1 Message Date
Remita Amine
c669554ef5 [medialaan] add support DPG Media MyChannels based websites
closes #14871
closes #15597
closes #16106
closes #16489
2021-01-27 14:58:47 +01:00
Remita Amine
11b68df7a4 [abcnews] fix extraction(closes #12394)(closes #27920) 2021-01-27 12:28:22 +01:00
5 changed files with 227 additions and 288 deletions

View File

@ -1,14 +1,15 @@
# coding: utf-8
from __future__ import unicode_literals
import calendar
import re
import time
from .amp import AMPIE
from .common import InfoExtractor
from .youtube import YoutubeIE
from ..compat import compat_urlparse
from ..utils import (
parse_duration,
parse_iso8601,
try_get,
)
class AbcNewsVideoIE(AMPIE):
@ -18,8 +19,8 @@ class AbcNewsVideoIE(AMPIE):
(?:
abcnews\.go\.com/
(?:
[^/]+/video/(?P<display_id>[0-9a-z-]+)-|
video/embed\?.*?\bid=
(?:[^/]+/)*video/(?P<display_id>[0-9a-z-]+)-|
video/(?:embed|itemfeed)\?.*?\bid=
)|
fivethirtyeight\.abcnews\.go\.com/video/embed/\d+/
)
@ -49,6 +50,12 @@ class AbcNewsVideoIE(AMPIE):
}, {
'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478',
'only_matching': True,
}, {
'url': 'http://abcnews.go.com/video/itemfeed?id=46979033',
'only_matching': True,
}, {
'url': 'https://abcnews.go.com/GMA/News/video/history-christmas-story-67894761',
'only_matching': True,
}]
def _real_extract(self, url):
@ -69,28 +76,23 @@ class AbcNewsIE(InfoExtractor):
_VALID_URL = r'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)'
_TESTS = [{
'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY',
# Youtube Embeds
'url': 'https://abcnews.go.com/Entertainment/peter-billingsley-child-actor-christmas-story-hollywood-power/story?id=51286501',
'info_dict': {
'id': '10505354',
'ext': 'flv',
'display_id': 'dramatic-video-rare-death-job-america',
'title': 'Occupational Hazards',
'description': 'Nightline investigates the dangers that lurk at various jobs.',
'thumbnail': r're:^https?://.*\.jpg$',
'upload_date': '20100428',
'timestamp': 1272412800,
'id': '51286501',
'title': "Peter Billingsley: From child actor in 'A Christmas Story' to Hollywood power player",
'description': 'Billingsley went from a child actor to Hollywood power player.',
},
'add_ie': ['AbcNewsVideo'],
'playlist_count': 5,
}, {
'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818',
'info_dict': {
'id': '38897857',
'ext': 'mp4',
'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016',
'title': 'Justin Timberlake Drops Hints For Secret Single',
'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.',
'upload_date': '20160515',
'timestamp': 1463329500,
'upload_date': '20160505',
'timestamp': 1462442280,
},
'params': {
# m3u8 download
@ -102,49 +104,55 @@ class AbcNewsIE(InfoExtractor):
}, {
'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
'only_matching': True,
}, {
# inline.type == 'video'
'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343',
'only_matching': True,
}]
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
display_id = mobj.group('display_id')
video_id = mobj.group('id')
story_id = self._match_id(url)
webpage = self._download_webpage(url, story_id)
story = self._parse_json(self._search_regex(
r"window\['__abcnews__'\]\s*=\s*({.+?});",
webpage, 'data'), story_id)['page']['content']['story']['everscroll'][0]
article_contents = story.get('articleContents') or {}
webpage = self._download_webpage(url, video_id)
video_url = self._search_regex(
r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL')
full_video_url = compat_urlparse.urljoin(url, video_url)
def entries():
featured_video = story.get('featuredVideo') or {}
feed = try_get(featured_video, lambda x: x['video']['feed'])
if feed:
yield {
'_type': 'url',
'id': featured_video.get('id'),
'title': featured_video.get('name'),
'url': feed,
'thumbnail': featured_video.get('images'),
'description': featured_video.get('description'),
'timestamp': parse_iso8601(featured_video.get('uploadDate')),
'duration': parse_duration(featured_video.get('duration')),
'ie_key': AbcNewsVideoIE.ie_key(),
}
youtube_url = YoutubeIE._extract_url(webpage)
for inline in (article_contents.get('inlines') or []):
inline_type = inline.get('type')
if inline_type == 'iframe':
iframe_url = try_get(inline, lambda x: x['attrs']['src'])
if iframe_url:
yield self.url_result(iframe_url)
elif inline_type == 'video':
video_id = inline.get('id')
if video_id:
yield {
'_type': 'url',
'id': video_id,
'url': 'http://abcnews.go.com/video/embed?id=' + video_id,
'thumbnail': inline.get('imgSrc') or inline.get('imgDefault'),
'description': inline.get('description'),
'duration': parse_duration(inline.get('duration')),
'ie_key': AbcNewsVideoIE.ie_key(),
}
timestamp = None
date_str = self._html_search_regex(
r'<span[^>]+class="timestamp">([^<]+)</span>',
webpage, 'timestamp', fatal=False)
if date_str:
tz_offset = 0
if date_str.endswith(' ET'): # Eastern Time
tz_offset = -5
date_str = date_str[:-3]
date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p']
for date_format in date_formats:
try:
timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format))
except ValueError:
continue
if timestamp is not None:
timestamp -= tz_offset * 3600
entry = {
'_type': 'url_transparent',
'ie_key': AbcNewsVideoIE.ie_key(),
'url': full_video_url,
'id': video_id,
'display_id': display_id,
'timestamp': timestamp,
}
if youtube_url:
entries = [entry, self.url_result(youtube_url, ie=YoutubeIE.ie_key())]
return self.playlist_result(entries)
return entry
return self.playlist_result(
entries(), story_id, article_contents.get('headline'),
article_contents.get('subHead'))

View File

@ -1459,6 +1459,7 @@ from .vrv import (
VRVSeriesIE,
)
from .vshare import VShareIE
from .vtm import VTMIE
from .medialaan import MedialaanIE
from .vube import VubeIE
from .vuclip import VuClipIE

View File

@ -128,6 +128,7 @@ from .zype import ZypeIE
from .odnoklassniki import OdnoklassnikiIE
from .kinja import KinjaEmbedIE
from .arcpublishing import ArcPublishingIE
from .medialaan import MedialaanIE
class GenericIE(InfoExtractor):
@ -2223,6 +2224,20 @@ class GenericIE(InfoExtractor):
'duration': 1581,
},
},
{
# MyChannels SDK embed
# https://www.24kitchen.nl/populair/deskundige-dit-waarom-sommigen-gevoelig-zijn-voor-voedselallergieen
'url': 'https://www.demorgen.be/nieuws/burgemeester-rotterdam-richt-zich-in-videoboodschap-tot-relschoppers-voelt-het-goed~b0bcfd741/',
'md5': '90c0699c37006ef18e198c032d81739c',
'info_dict': {
'id': '194165',
'ext': 'mp4',
'title': 'Burgemeester Aboutaleb spreekt relschoppers toe',
'timestamp': 1611740340,
'upload_date': '20210127',
'duration': 159,
},
},
]
def report_following_redirect(self, new_url):
@ -2462,6 +2477,9 @@ class GenericIE(InfoExtractor):
webpage = self._webpage_read_content(
full_response, url, video_id, prefix=first_bytes)
if '<title>DPG Media Privacy Gate</title>' in webpage:
webpage = self._download_webpage(url, video_id)
self.report_extraction(video_id)
# Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest?
@ -2593,6 +2611,11 @@ class GenericIE(InfoExtractor):
if arc_urls:
return self.playlist_from_matches(arc_urls, video_id, video_title, ie=ArcPublishingIE.ie_key())
mychannels_urls = MedialaanIE._extract_urls(webpage)
if mychannels_urls:
return self.playlist_from_matches(
mychannels_urls, video_id, video_title, ie=MedialaanIE.ie_key())
# Look for embedded rtl.nl player
matches = re.findall(
r'<iframe[^>]+?src="((?:https?:)?//(?:(?:www|static)\.)?rtl\.nl/(?:system/videoplayer/[^"]+(?:video_)?)?embed[^"]+)"',

View File

@ -2,268 +2,113 @@ from __future__ import unicode_literals
import re
from .gigya import GigyaBaseIE
from ..compat import compat_str
from .common import InfoExtractor
from ..utils import (
extract_attributes,
int_or_none,
parse_duration,
try_get,
unified_timestamp,
mimetype2ext,
parse_iso8601,
)
class MedialaanIE(GigyaBaseIE):
class MedialaanIE(InfoExtractor):
_VALID_URL = r'''(?x)
https?://
(?:www\.|nieuws\.)?
(?:
(?P<site_id>vtm|q2|vtmkzoom)\.be/
(?:
video(?:/[^/]+/id/|/?\?.*?\baid=)|
(?:[^/]+/)*
)
(?:embed\.)?mychannels.video/embed/|
embed\.mychannels\.video/(?:s(?:dk|cript)/)?production/|
(?:www\.)?(?:
(?:
7sur7|
demorgen|
hln|
joe|
qmusic
)\.be|
(?:
[abe]d|
bndestem|
destentor|
gelderlander|
pzc|
tubantia|
volkskrant
)\.nl
)/video/(?:[^/]+/)*[^/?&#]+~p
)
(?P<id>[^/?#&]+)
(?P<id>\d+)
'''
_NETRC_MACHINE = 'medialaan'
_APIKEY = '3_HZ0FtkMW_gOyKlqQzW5_0FHRC7Nd5XpXJZcDdXY4pk5eES2ZWmejRW5egwVm4ug-'
_SITE_TO_APP_ID = {
'vtm': 'vtm_watch',
'q2': 'q2',
'vtmkzoom': 'vtmkzoom',
}
_TESTS = [{
# vod
'url': 'http://vtm.be/video/volledige-afleveringen/id/vtm_20170219_VM0678361_vtmwatch',
'url': 'https://www.bndestem.nl/video/de-terugkeer-van-ally-de-aap-en-wie-vertrekt-er-nog-bij-nac~p193993',
'info_dict': {
'id': 'vtm_20170219_VM0678361_vtmwatch',
'id': '193993',
'ext': 'mp4',
'title': 'Allemaal Chris afl. 6',
'description': 'md5:4be86427521e7b07e0adb0c9c554ddb2',
'timestamp': 1487533280,
'upload_date': '20170219',
'duration': 2562,
'series': 'Allemaal Chris',
'season': 'Allemaal Chris',
'season_number': 1,
'season_id': '256936078124527',
'episode': 'Allemaal Chris afl. 6',
'episode_number': 6,
'episode_id': '256936078591527',
'title': 'De terugkeer van Ally de Aap en wie vertrekt er nog bij NAC?',
'timestamp': 1611663540,
'upload_date': '20210126',
'duration': 238,
},
'params': {
'skip_download': True,
},
'skip': 'Requires account credentials',
}, {
# clip
'url': 'http://vtm.be/video?aid=168332',
'info_dict': {
'id': '168332',
'ext': 'mp4',
'title': '"Veronique liegt!"',
'description': 'md5:1385e2b743923afe54ba4adc38476155',
'timestamp': 1489002029,
'upload_date': '20170308',
'duration': 96,
},
}, {
# vod
'url': 'http://vtm.be/video/volledige-afleveringen/id/257107153551000',
'url': 'https://www.gelderlander.nl/video/kanalen/degelderlander~c320/series/snel-nieuws~s984/noodbevel-in-doetinchem-politie-stuurt-mensen-centrum-uit~p194093',
'only_matching': True,
}, {
# vod
'url': 'http://vtm.be/video?aid=163157',
'url': 'https://embed.mychannels.video/sdk/production/193993?options=TFTFF_default',
'only_matching': True,
}, {
# vod
'url': 'http://www.q2.be/video/volledige-afleveringen/id/2be_20170301_VM0684442_q2',
'url': 'https://embed.mychannels.video/script/production/193993',
'only_matching': True,
}, {
# clip
'url': 'http://vtmkzoom.be/k3-dansstudio/een-nieuw-seizoen-van-k3-dansstudio',
'url': 'https://embed.mychannels.video/production/193993',
'only_matching': True,
}, {
# http/s redirect
'url': 'https://vtmkzoom.be/video?aid=45724',
'info_dict': {
'id': '257136373657000',
'ext': 'mp4',
'title': 'K3 Dansstudio Ushuaia afl.6',
},
'params': {
'skip_download': True,
},
'skip': 'Requires account credentials',
'url': 'https://mychannels.video/embed/193993',
'only_matching': True,
}, {
# nieuws.vtm.be
'url': 'https://nieuws.vtm.be/stadion/stadion/genk-nog-moeilijk-programma',
'url': 'https://embed.mychannels.video/embed/193993',
'only_matching': True,
}]
def _real_initialize(self):
self._logged_in = False
def _login(self):
username, password = self._get_login_info()
if username is None:
self.raise_login_required()
auth_data = {
'APIKey': self._APIKEY,
'sdk': 'js_6.1',
'format': 'json',
'loginID': username,
'password': password,
}
auth_info = self._gigya_login(auth_data)
self._uid = auth_info['UID']
self._uid_signature = auth_info['UIDSignature']
self._signature_timestamp = auth_info['signatureTimestamp']
self._logged_in = True
@staticmethod
def _extract_urls(webpage):
entries = []
for element in re.findall(r'(<div[^>]+data-mychannels-type="video"[^>]*>)', webpage):
mychannels_id = extract_attributes(element).get('data-mychannels-id')
if mychannels_id:
entries.append('https://mychannels.video/embed/' + mychannels_id)
return entries
def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url)
video_id, site_id = mobj.group('id', 'site_id')
production_id = self._match_id(url)
production = self._download_json(
'https://embed.mychannels.video/sdk/production/' + production_id,
production_id, query={'options': 'UUUU_default'})['productions'][0]
title = production['title']
webpage = self._download_webpage(url, video_id)
config = self._parse_json(
self._search_regex(
r'videoJSConfig\s*=\s*JSON\.parse\(\'({.+?})\'\);',
webpage, 'config', default='{}'), video_id,
transform_source=lambda s: s.replace(
'\\\\', '\\').replace(r'\"', '"').replace(r"\'", "'"))
vod_id = config.get('vodId') or self._search_regex(
(r'\\"vodId\\"\s*:\s*\\"(.+?)\\"',
r'"vodId"\s*:\s*"(.+?)"',
r'<[^>]+id=["\']vod-(\d+)'),
webpage, 'video_id', default=None)
# clip, no authentication required
if not vod_id:
player = self._parse_json(
self._search_regex(
r'vmmaplayer\(({.+?})\);', webpage, 'vmma player',
default=''),
video_id, transform_source=lambda s: '[%s]' % s, fatal=False)
if player:
video = player[-1]
if video['videoUrl'] in ('http', 'https'):
return self.url_result(video['url'], MedialaanIE.ie_key())
info = {
'id': video_id,
'url': video['videoUrl'],
'title': video['title'],
'thumbnail': video.get('imageUrl'),
'timestamp': int_or_none(video.get('createdDate')),
'duration': int_or_none(video.get('duration')),
}
formats = []
for source in (production.get('sources') or []):
src = source.get('src')
if not src:
continue
ext = mimetype2ext(source.get('type'))
if ext == 'm3u8':
formats.extend(self._extract_m3u8_formats(
src, production_id, 'mp4', 'm3u8_native',
m3u8_id='hls', fatal=False))
else:
info = self._parse_html5_media_entries(
url, webpage, video_id, m3u8_id='hls')[0]
info.update({
'id': video_id,
'title': self._html_search_meta('description', webpage),
'duration': parse_duration(self._html_search_meta('duration', webpage)),
formats.append({
'ext': ext,
'url': src,
})
# vod, authentication required
else:
if not self._logged_in:
self._login()
self._sort_formats(formats)
settings = self._parse_json(
self._search_regex(
r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);',
webpage, 'drupal settings', default='{}'),
video_id)
def get(container, item):
return try_get(
settings, lambda x: x[container][item],
compat_str) or self._search_regex(
r'"%s"\s*:\s*"([^"]+)' % item, webpage, item,
default=None)
app_id = get('vod', 'app_id') or self._SITE_TO_APP_ID.get(site_id, 'vtm_watch')
sso = get('vod', 'gigyaDatabase') or 'vtm-sso'
data = self._download_json(
'http://vod.medialaan.io/api/1.0/item/%s/video' % vod_id,
video_id, query={
'app_id': app_id,
'user_network': sso,
'UID': self._uid,
'UIDSignature': self._uid_signature,
'signatureTimestamp': self._signature_timestamp,
})
formats = self._extract_m3u8_formats(
data['response']['uri'], video_id, entry_protocol='m3u8_native',
ext='mp4', m3u8_id='hls')
self._sort_formats(formats)
info = {
'id': vod_id,
'formats': formats,
}
api_key = get('vod', 'apiKey')
channel = get('medialaanGigya', 'channel')
if api_key:
videos = self._download_json(
'http://vod.medialaan.io/vod/v2/videos', video_id, fatal=False,
query={
'channels': channel,
'ids': vod_id,
'limit': 1,
'apikey': api_key,
})
if videos:
video = try_get(
videos, lambda x: x['response']['videos'][0], dict)
if video:
def get(container, item, expected_type=None):
return try_get(
video, lambda x: x[container][item], expected_type)
def get_string(container, item):
return get(container, item, compat_str)
info.update({
'series': get_string('program', 'title'),
'season': get_string('season', 'title'),
'season_number': int_or_none(get('season', 'number')),
'season_id': get_string('season', 'id'),
'episode': get_string('episode', 'title'),
'episode_number': int_or_none(get('episode', 'number')),
'episode_id': get_string('episode', 'id'),
'duration': int_or_none(
video.get('duration')) or int_or_none(
video.get('durationMillis'), scale=1000),
'title': get_string('episode', 'title'),
'description': get_string('episode', 'text'),
'timestamp': unified_timestamp(get_string(
'publication', 'begin')),
})
if not info.get('title'):
info['title'] = try_get(
config, lambda x: x['videoConfig']['title'],
compat_str) or self._html_search_regex(
r'\\"title\\"\s*:\s*\\"(.+?)\\"', webpage, 'title',
default=None) or self._og_search_title(webpage)
if not info.get('description'):
info['description'] = self._html_search_regex(
r'<div[^>]+class="field-item\s+even">\s*<p>(.+?)</p>',
webpage, 'description', default=None)
return info
return {
'id': production_id,
'title': title,
'formats': formats,
'thumbnail': production.get('posterUrl'),
'timestamp': parse_iso8601(production.get('publicationDate'), ' '),
'duration': int_or_none(production.get('duration')) or None,
}

View File

@ -0,0 +1,62 @@
# coding: utf-8
from __future__ import unicode_literals
from .common import InfoExtractor
from ..utils import (
int_or_none,
parse_iso8601,
try_get,
)
class VTMIE(InfoExtractor):
_VALID_URL = r'https?://(?:www\.)?vtm\.be/([^/?&#]+)~v(?P<id>[0-9a-f]{8}(?:-[0-9a-f]{4}){3}-[0-9a-f]{12})'
_TEST = {
'url': 'https://vtm.be/gast-vernielt-genkse-hotelkamer~ve7534523-279f-4b4d-a5c9-a33ffdbe23e1',
'md5': '37dca85fbc3a33f2de28ceb834b071f8',
'info_dict': {
'id': '192445',
'ext': 'mp4',
'title': 'Gast vernielt Genkse hotelkamer',
'timestamp': 1611060180,
'upload_date': '20210119',
'duration': 74,
# TODO: fix url _type result processing
# 'series': 'Op Interventie',
}
}
def _real_extract(self, url):
uuid = self._match_id(url)
video = self._download_json(
'https://omc4vm23offuhaxx6hekxtzspi.appsync-api.eu-west-1.amazonaws.com/graphql',
uuid, query={
'query': '''{
getComponent(type: Video, uuid: "%s") {
... on Video {
description
duration
myChannelsVideo
program {
title
}
publishedAt
title
}
}
}''' % uuid,
}, headers={
'x-api-key': 'da2-lz2cab4tfnah3mve6wiye4n77e',
})['data']['getComponent']
return {
'_type': 'url',
'id': uuid,
'title': video.get('title'),
'url': 'http://mychannels.video/embed/%d' % video['myChannelsVideo'],
'description': video.get('description'),
'timestamp': parse_iso8601(video.get('publishedAt')),
'duration': int_or_none(video.get('duration')),
'series': try_get(video, lambda x: x['program']['title']),
'ie_key': 'Medialaan',
}