[dplay] improve extraction

- add support for dplay.fi, dplay.jp and es.dplay.com(closes #16969)
- fix it.dplay.com extraction(closes #22826)
- update tests
- extract creator, tags and thumbnails
- handle playback API call errors
This commit is contained in:
Remita Amine 2019-10-26 14:58:29 +01:00
parent 235dbb434b
commit 0b98f3a751
2 changed files with 133 additions and 269 deletions

View file

@ -1,74 +1,68 @@
# coding: utf-8 # coding: utf-8
from __future__ import unicode_literals from __future__ import unicode_literals
import json
import re import re
import time
from .common import InfoExtractor from .common import InfoExtractor
from ..compat import ( from ..compat import compat_HTTPError
compat_HTTPError,
compat_str,
compat_urlparse,
)
from ..utils import ( from ..utils import (
determine_ext, determine_ext,
ExtractorError, ExtractorError,
float_or_none, float_or_none,
int_or_none, int_or_none,
remove_end,
try_get,
unified_strdate,
unified_timestamp, unified_timestamp,
update_url_query,
urljoin,
USER_AGENTS,
) )
class DPlayIE(InfoExtractor): class DPlayIE(InfoExtractor):
_VALID_URL = r'https?://(?P<domain>www\.(?P<host>dplay\.(?P<country>dk|se|no)))/(?:video(?:er|s)/)?(?P<id>[^/]+/[^/?#]+)' _VALID_URL = r'''(?x)https?://
(?P<domain>
(?:www\.)?(?P<host>dplay\.(?P<country>dk|fi|jp|se|no))|
(?P<subdomain_country>es|it)\.dplay\.com
)/[^/]+/(?P<id>[^/]+/[^/?#]+)'''
_TESTS = [{ _TESTS = [{
# non geo restricted, via secure api, unsigned download hls URL # non geo restricted, via secure api, unsigned download hls URL
'url': 'http://www.dplay.se/nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet/', 'url': 'https://www.dplay.se/videos/nugammalt-77-handelser-som-format-sverige/nugammalt-77-handelser-som-format-sverige-101',
'info_dict': { 'info_dict': {
'id': '3172', 'id': '13628',
'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet', 'display_id': 'nugammalt-77-handelser-som-format-sverige/nugammalt-77-handelser-som-format-sverige-101',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Svensken lär sig njuta av livet', 'title': 'Svensken lär sig njuta av livet',
'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8', 'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8',
'duration': 2650, 'duration': 2649.856,
'timestamp': 1365454320, 'timestamp': 1365453720,
'upload_date': '20130408', 'upload_date': '20130408',
'creator': 'Kanal 5 (Home)', 'creator': 'Kanal 5',
'series': 'Nugammalt - 77 händelser som format Sverige', 'series': 'Nugammalt - 77 händelser som format Sverige',
'season_number': 1, 'season_number': 1,
'episode_number': 1, 'episode_number': 1,
'age_limit': 0, },
'params': {
'format': 'bestvideo',
'skip_download': True,
}, },
}, { }, {
# geo restricted, via secure api, unsigned download hls URL # geo restricted, via secure api, unsigned download hls URL
'url': 'http://www.dplay.dk/mig-og-min-mor/season-6-episode-12/', 'url': 'http://www.dplay.dk/videoer/ted-bundy-mind-of-a-monster/ted-bundy-mind-of-a-monster',
'info_dict': { 'info_dict': {
'id': '70816', 'id': '104465',
'display_id': 'mig-og-min-mor/season-6-episode-12', 'display_id': 'ted-bundy-mind-of-a-monster/ted-bundy-mind-of-a-monster',
'ext': 'mp4', 'ext': 'mp4',
'title': 'Episode 12', 'title': 'Ted Bundy: Mind Of A Monster',
'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90', 'description': 'md5:8b780f6f18de4dae631668b8a9637995',
'duration': 2563, 'duration': 5290.027,
'timestamp': 1429696800, 'timestamp': 1570694400,
'upload_date': '20150422', 'upload_date': '20191010',
'creator': 'Kanal 4 (Home)', 'creator': 'ID - Investigation Discovery',
'series': 'Mig og min mor', 'series': 'Ted Bundy: Mind Of A Monster',
'season_number': 6, 'season_number': 1,
'episode_number': 12, 'episode_number': 1,
'age_limit': 0, },
'params': {
'format': 'bestvideo',
'skip_download': True,
}, },
}, {
# geo restricted, via direct unsigned hls URL
'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
'only_matching': True,
}, { }, {
# disco-api # disco-api
'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7', 'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
@ -89,19 +83,59 @@ class DPlayIE(InfoExtractor):
'format': 'bestvideo', 'format': 'bestvideo',
'skip_download': True, 'skip_download': True,
}, },
'skip': 'Available for Premium users',
}, { }, {
'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
'url': 'https://www.dplay.dk/videoer/singleliv/season-5-episode-3', 'md5': '2b808ffb00fc47b884a172ca5d13053c',
'info_dict': {
'id': '6918',
'display_id': 'biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij',
'ext': 'mp4',
'title': 'Luigi Di Maio: la psicosi di Stanislawskij',
'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
'thumbnail': r're:^https?://.*\.jpe?g',
'upload_date': '20160524',
'timestamp': 1464076800,
'series': 'Biografie imbarazzanti',
'season_number': 1,
'episode': 'Episode 1',
'episode_number': 1,
},
}, {
'url': 'https://es.dplay.com/dmax/la-fiebre-del-oro/temporada-8-episodio-1/',
'info_dict': {
'id': '21652',
'display_id': 'la-fiebre-del-oro/temporada-8-episodio-1',
'ext': 'mp4',
'title': 'Episodio 1',
'description': 'md5:b9dcff2071086e003737485210675f69',
'thumbnail': r're:^https?://.*\.png',
'upload_date': '20180709',
'timestamp': 1531173540,
'series': 'La fiebre del oro',
'season_number': 8,
'episode': 'Episode 1',
'episode_number': 1,
},
'params': {
'skip_download': True,
},
}, {
'url': 'https://www.dplay.fi/videot/shifting-gears-with-aaron-kaufman/episode-16',
'only_matching': True, 'only_matching': True,
}, { }, {
'url': 'https://www.dplay.se/videos/sofias-anglar/sofias-anglar-1001', 'url': 'https://www.dplay.jp/video/gold-rush/24086',
'only_matching': True, 'only_matching': True,
}] }]
def _get_disco_api_info(self, url, display_id, disco_host, realm): def _get_disco_api_info(self, url, display_id, disco_host, realm, country):
disco_base = 'https://' + disco_host geo_countries = [country.upper()]
self._initialize_geo_bypass({
'countries': geo_countries,
})
disco_base = 'https://%s/' % disco_host
token = self._download_json( token = self._download_json(
'%s/token' % disco_base, display_id, 'Downloading token', disco_base + 'token', display_id, 'Downloading token',
query={ query={
'realm': realm, 'realm': realm,
})['data']['attributes']['token'] })['data']['attributes']['token']
@ -110,17 +144,30 @@ class DPlayIE(InfoExtractor):
'Authorization': 'Bearer ' + token, 'Authorization': 'Bearer ' + token,
} }
video = self._download_json( video = self._download_json(
'%s/content/videos/%s' % (disco_base, display_id), display_id, disco_base + 'content/videos/' + display_id, display_id,
headers=headers, query={ headers=headers, query={
'include': 'show' 'include': 'images,primaryChannel,show,tags'
}) })
video_id = video['data']['id'] video_id = video['data']['id']
info = video['data']['attributes'] info = video['data']['attributes']
title = info['name'] title = info['name'].strip()
formats = [] formats = []
for format_id, format_dict in self._download_json( try:
'%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id), streaming = self._download_json(
display_id, headers=headers)['data']['attributes']['streaming'].items(): disco_base + 'playback/videoPlaybackInfo/' + video_id,
display_id, headers=headers)['data']['attributes']['streaming']
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
error = info['errors'][0]
error_code = error.get('code')
if error_code == 'access.denied.geoblocked':
self.raise_geo_restricted(countries=geo_countries)
elif error_code == 'access.denied.missingpackage':
self.raise_login_required()
raise ExtractorError(info['errors'][0]['detail'], expected=True)
raise
for format_id, format_dict in streaming.items():
if not isinstance(format_dict, dict): if not isinstance(format_dict, dict):
continue continue
format_url = format_dict.get('url') format_url = format_dict.get('url')
@ -142,235 +189,55 @@ class DPlayIE(InfoExtractor):
}) })
self._sort_formats(formats) self._sort_formats(formats)
series = None creator = series = None
try: tags = []
included = video.get('included') thumbnails = []
if isinstance(included, list): included = video.get('included') or []
show = next(e for e in included if e.get('type') == 'show') if isinstance(included, list):
series = try_get( for e in included:
show, lambda x: x['attributes']['name'], compat_str) attributes = e.get('attributes')
except StopIteration: if not attributes:
pass continue
e_type = e.get('type')
if e_type == 'channel':
creator = attributes.get('name')
elif e_type == 'image':
src = attributes.get('src')
if src:
thumbnails.append({
'url': src,
'width': int_or_none(attributes.get('width')),
'height': int_or_none(attributes.get('height')),
})
if e_type == 'show':
series = attributes.get('name')
elif e_type == 'tag':
name = attributes.get('name')
if name:
tags.append(name)
return { return {
'id': video_id, 'id': video_id,
'display_id': display_id, 'display_id': display_id,
'title': title, 'title': title,
'description': info.get('description'), 'description': info.get('description'),
'duration': float_or_none( 'duration': float_or_none(info.get('videoDuration'), 1000),
info.get('videoDuration'), scale=1000),
'timestamp': unified_timestamp(info.get('publishStart')), 'timestamp': unified_timestamp(info.get('publishStart')),
'series': series, 'series': series,
'season_number': int_or_none(info.get('seasonNumber')), 'season_number': int_or_none(info.get('seasonNumber')),
'episode_number': int_or_none(info.get('episodeNumber')), 'episode_number': int_or_none(info.get('episodeNumber')),
'age_limit': int_or_none(info.get('minimum_age')), 'age_limit': int_or_none(info.get('minimum_age')),
'creator': creator,
'tags': tags,
'thumbnails': thumbnails,
'formats': formats, 'formats': formats,
} }
def _real_extract(self, url): def _real_extract(self, url):
mobj = re.match(self._VALID_URL, url) mobj = re.match(self._VALID_URL, url)
display_id = mobj.group('id') display_id = mobj.group('id')
domain = mobj.group('domain') domain = mobj.group('domain').lstrip('www.')
country = mobj.group('country') or mobj.group('subdomain_country')
self._initialize_geo_bypass({ host = 'disco-api.' + domain if domain.startswith('dplay.') else 'eu2-prod.disco-api.com'
'countries': [mobj.group('country').upper()], return self._get_disco_api_info(
}) url, display_id, host, 'dplay' + country, country)
webpage = self._download_webpage(url, display_id)
video_id = self._search_regex(
r'data-video-id=["\'](\d+)', webpage, 'video id', default=None)
if not video_id:
host = mobj.group('host')
return self._get_disco_api_info(
url, display_id, 'disco-api.' + host, host.replace('.', ''))
info = self._download_json(
'http://%s/api/v2/ajax/videos?video_id=%s' % (domain, video_id),
video_id)['data'][0]
title = info['title']
PROTOCOLS = ('hls', 'hds')
formats = []
def extract_formats(protocol, manifest_url):
if protocol == 'hls':
m3u8_formats = self._extract_m3u8_formats(
manifest_url, video_id, ext='mp4',
entry_protocol='m3u8_native', m3u8_id=protocol, fatal=False)
# Sometimes final URLs inside m3u8 are unsigned, let's fix this
# ourselves. Also fragments' URLs are only served signed for
# Safari user agent.
query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
for m3u8_format in m3u8_formats:
m3u8_format.update({
'url': update_url_query(m3u8_format['url'], query),
'http_headers': {
'User-Agent': USER_AGENTS['Safari'],
},
})
formats.extend(m3u8_formats)
elif protocol == 'hds':
formats.extend(self._extract_f4m_formats(
manifest_url + '&hdcore=3.8.0&plugin=flowplayer-3.8.0.0',
video_id, f4m_id=protocol, fatal=False))
domain_tld = domain.split('.')[-1]
if domain_tld in ('se', 'dk', 'no'):
for protocol in PROTOCOLS:
# Providing dsc-geo allows to bypass geo restriction in some cases
self._set_cookie(
'secure.dplay.%s' % domain_tld, 'dsc-geo',
json.dumps({
'countryCode': domain_tld.upper(),
'expiry': (time.time() + 20 * 60) * 1000,
}))
stream = self._download_json(
'https://secure.dplay.%s/secure/api/v2/user/authorization/stream/%s?stream_type=%s'
% (domain_tld, video_id, protocol), video_id,
'Downloading %s stream JSON' % protocol, fatal=False)
if stream and stream.get(protocol):
extract_formats(protocol, stream[protocol])
# The last resort is to try direct unsigned hls/hds URLs from info dictionary.
# Sometimes this does work even when secure API with dsc-geo has failed (e.g.
# http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/).
if not formats:
for protocol in PROTOCOLS:
if info.get(protocol):
extract_formats(protocol, info[protocol])
self._sort_formats(formats)
subtitles = {}
for lang in ('se', 'sv', 'da', 'nl', 'no'):
for format_id in ('web_vtt', 'vtt', 'srt'):
subtitle_url = info.get('subtitles_%s_%s' % (lang, format_id))
if subtitle_url:
subtitles.setdefault(lang, []).append({'url': subtitle_url})
return {
'id': video_id,
'display_id': display_id,
'title': title,
'description': info.get('video_metadata_longDescription'),
'duration': int_or_none(info.get('video_metadata_length'), scale=1000),
'timestamp': int_or_none(info.get('video_publish_date')),
'creator': info.get('video_metadata_homeChannel'),
'series': info.get('video_metadata_show'),
'season_number': int_or_none(info.get('season')),
'episode_number': int_or_none(info.get('episode')),
'age_limit': int_or_none(info.get('minimum_age')),
'formats': formats,
'subtitles': subtitles,
}
class DPlayItIE(InfoExtractor):
_VALID_URL = r'https?://it\.dplay\.com/[^/]+/[^/]+/(?P<id>[^/?#]+)'
_GEO_COUNTRIES = ['IT']
_TEST = {
'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
'md5': '2b808ffb00fc47b884a172ca5d13053c',
'info_dict': {
'id': '6918',
'display_id': 'luigi-di-maio-la-psicosi-di-stanislawskij',
'ext': 'mp4',
'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij',
'description': 'md5:3c7a4303aef85868f867a26f5cc14813',
'thumbnail': r're:^https?://.*\.jpe?g',
'upload_date': '20160524',
'series': 'Biografie imbarazzanti',
'season_number': 1,
'episode': 'Luigi Di Maio: la psicosi di Stanislawskij',
'episode_number': 1,
},
}
def _real_extract(self, url):
display_id = self._match_id(url)
webpage = self._download_webpage(url, display_id)
title = remove_end(self._og_search_title(webpage), ' | Dplay')
video_id = None
info = self._search_regex(
r'playback_json\s*:\s*JSON\.parse\s*\(\s*("(?:\\.|[^"\\])+?")',
webpage, 'playback JSON', default=None)
if info:
for _ in range(2):
info = self._parse_json(info, display_id, fatal=False)
if not info:
break
else:
video_id = try_get(info, lambda x: x['data']['id'])
if not info:
info_url = self._search_regex(
(r'playback_json_url\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1',
r'url\s*[:=]\s*["\'](?P<url>(?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)'),
webpage, 'info url', group='url')
info_url = urljoin(url, info_url)
video_id = info_url.rpartition('/')[-1]
try:
info = self._download_json(
info_url, display_id, headers={
'Authorization': 'Bearer %s' % self._get_cookies(url).get(
'dplayit_token').value,
'Referer': url,
})
if isinstance(info, compat_str):
info = self._parse_json(info, display_id)
except ExtractorError as e:
if isinstance(e.cause, compat_HTTPError) and e.cause.code in (400, 403):
info = self._parse_json(e.cause.read().decode('utf-8'), display_id)
error = info['errors'][0]
if error.get('code') == 'access.denied.geoblocked':
self.raise_geo_restricted(
msg=error.get('detail'), countries=self._GEO_COUNTRIES)
raise ExtractorError(info['errors'][0]['detail'], expected=True)
raise
hls_url = info['data']['attributes']['streaming']['hls']['url']
formats = self._extract_m3u8_formats(
hls_url, display_id, ext='mp4', entry_protocol='m3u8_native',
m3u8_id='hls')
self._sort_formats(formats)
series = self._html_search_regex(
r'(?s)<h1[^>]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)</h1>',
webpage, 'series', fatal=False)
episode = self._search_regex(
r'<p[^>]+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*<br/>\s*<b>([^<]+)',
webpage, 'episode', fatal=False)
mobj = re.search(
r'(?s)<span[^>]+class=["\']dates["\'][^>]*>.+?\bS\.(?P<season_number>\d+)\s+E\.(?P<episode_number>\d+)\s*-\s*(?P<upload_date>\d{2}/\d{2}/\d{4})',
webpage)
if mobj:
season_number = int(mobj.group('season_number'))
episode_number = int(mobj.group('episode_number'))
upload_date = unified_strdate(mobj.group('upload_date'))
else:
season_number = episode_number = upload_date = None
return {
'id': compat_str(video_id or display_id),
'display_id': display_id,
'title': title,
'description': self._og_search_description(webpage),
'thumbnail': self._og_search_thumbnail(webpage),
'series': series,
'season_number': season_number,
'episode': episode,
'episode_number': episode_number,
'upload_date': upload_date,
'formats': formats,
}

View file

@ -277,10 +277,7 @@ from .douyutv import (
DouyuShowIE, DouyuShowIE,
DouyuTVIE, DouyuTVIE,
) )
from .dplay import ( from .dplay import DPlayIE
DPlayIE,
DPlayItIE,
)
from .dreisat import DreiSatIE from .dreisat import DreiSatIE
from .drbonanza import DRBonanzaIE from .drbonanza import DRBonanzaIE
from .drtuber import DrTuberIE from .drtuber import DrTuberIE