X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/1270059fa7cf720bc0533c2cdfe6370923ae4e20..20205e79eb4216762c923790b27dbe38f945293a:/youtube_dl/extractor/dplay.py diff --git a/youtube_dl/extractor/dplay.py b/youtube_dl/extractor/dplay.py index b734467..a7b9db5 100644 --- a/youtube_dl/extractor/dplay.py +++ b/youtube_dl/extractor/dplay.py @@ -1,73 +1,68 @@ # coding: utf-8 from __future__ import unicode_literals -import json import re -import time from .common import InfoExtractor -from ..compat import ( - compat_HTTPError, - compat_str, - compat_urlparse, -) +from ..compat import compat_HTTPError from ..utils import ( determine_ext, ExtractorError, float_or_none, int_or_none, - remove_end, - try_get, - unified_strdate, unified_timestamp, - update_url_query, - USER_AGENTS, ) class DPlayIE(InfoExtractor): - _VALID_URL = r'https?://(?Pwww\.(?Pdplay\.(?Pdk|se|no)))/(?:video(?:er|s)/)?(?P[^/]+/[^/?#]+)' + _VALID_URL = r'''(?x)https?:// + (?P + (?:www\.)?(?Pdplay\.(?Pdk|fi|jp|se|no))| + (?Pes|it)\.dplay\.com + )/[^/]+/(?P[^/]+/[^/?#]+)''' _TESTS = [{ # 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': { - 'id': '3172', - 'display_id': 'nugammalt-77-handelser-som-format-sverige/season-1-svensken-lar-sig-njuta-av-livet', + 'id': '13628', + 'display_id': 'nugammalt-77-handelser-som-format-sverige/nugammalt-77-handelser-som-format-sverige-101', 'ext': 'mp4', 'title': 'Svensken lär sig njuta av livet', 'description': 'md5:d3819c9bccffd0fe458ca42451dd50d8', - 'duration': 2650, - 'timestamp': 1365454320, + 'duration': 2649.856, + 'timestamp': 1365453720, 'upload_date': '20130408', - 'creator': 'Kanal 5 (Home)', + 'creator': 'Kanal 5', 'series': 'Nugammalt - 77 händelser som format Sverige', 'season_number': 1, 'episode_number': 1, - 'age_limit': 0, + }, + 'params': { + 'format': 'bestvideo', + 'skip_download': True, }, }, { # 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': { - 'id': '70816', - 'display_id': 'mig-og-min-mor/season-6-episode-12', + 'id': '104465', + 'display_id': 'ted-bundy-mind-of-a-monster/ted-bundy-mind-of-a-monster', 'ext': 'mp4', - 'title': 'Episode 12', - 'description': 'md5:9c86e51a93f8a4401fc9641ef9894c90', - 'duration': 2563, - 'timestamp': 1429696800, - 'upload_date': '20150422', - 'creator': 'Kanal 4 (Home)', - 'series': 'Mig og min mor', - 'season_number': 6, - 'episode_number': 12, - 'age_limit': 0, + 'title': 'Ted Bundy: Mind Of A Monster', + 'description': 'md5:8b780f6f18de4dae631668b8a9637995', + 'duration': 5290.027, + 'timestamp': 1570694400, + 'upload_date': '20191010', + 'creator': 'ID - Investigation Discovery', + 'series': 'Ted Bundy: Mind Of A Monster', + 'season_number': 1, + 'episode_number': 1, + }, + '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 'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7', @@ -88,276 +83,165 @@ class DPlayIE(InfoExtractor): 'format': 'bestvideo', 'skip_download': True, }, + 'skip': 'Available for Premium users', }, { - - 'url': 'https://www.dplay.dk/videoer/singleliv/season-5-episode-3', - 'only_matching': True, - }, { - 'url': 'https://www.dplay.se/videos/sofias-anglar/sofias-anglar-1001', - 'only_matching': True, - }] - - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('id') - domain = mobj.group('domain') - - self._initialize_geo_bypass([mobj.group('country').upper()]) - - 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') - disco_base = 'https://disco-api.%s' % host - self._download_json( - '%s/token' % disco_base, display_id, 'Downloading token', - query={ - 'realm': host.replace('.', ''), - }) - video = self._download_json( - '%s/content/videos/%s' % (disco_base, display_id), display_id, - headers={ - 'Referer': url, - 'x-disco-client': 'WEB:UNKNOWN:dplay-client:0.0.1', - }, query={ - 'include': 'show' - }) - video_id = video['data']['id'] - info = video['data']['attributes'] - title = info['name'] - formats = [] - for format_id, format_dict in self._download_json( - '%s/playback/videoPlaybackInfo/%s' % (disco_base, video_id), - display_id)['data']['attributes']['streaming'].items(): - if not isinstance(format_dict, dict): - continue - format_url = format_dict.get('url') - if not format_url: - continue - ext = determine_ext(format_url) - if format_id == 'dash' or ext == 'mpd': - formats.extend(self._extract_mpd_formats( - format_url, display_id, mpd_id='dash', fatal=False)) - elif format_id == 'hls' or ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - format_url, display_id, 'mp4', - entry_protocol='m3u8_native', m3u8_id='hls', - fatal=False)) - else: - formats.append({ - 'url': format_url, - 'format_id': format_id, - }) - self._sort_formats(formats) - - series = None - try: - included = video.get('included') - if isinstance(included, list): - show = next(e for e in included if e.get('type') == 'show') - series = try_get( - show, lambda x: x['attributes']['name'], compat_str) - except StopIteration: - pass - - return { - 'id': video_id, - 'display_id': display_id, - 'title': title, - 'description': info.get('description'), - 'duration': float_or_none( - info.get('videoDuration'), scale=1000), - 'timestamp': unified_timestamp(info.get('publishStart')), - 'series': series, - 'season_number': int_or_none(info.get('seasonNumber')), - 'episode_number': int_or_none(info.get('episodeNumber')), - 'age_limit': int_or_none(info.get('minimum_age')), - 'formats': formats, - } - - 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[^/?#]+)' - _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', + 'display_id': 'biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij', 'ext': 'mp4', - 'title': 'Biografie imbarazzanti: Luigi Di Maio: la psicosi di Stanislawskij', + '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': 'Luigi Di Maio: la psicosi di Stanislawskij', + 'episode': 'Episode 1', '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 + }, { + '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, + }, { + 'url': 'https://www.dplay.jp/video/gold-rush/24086', + 'only_matching': True, + }] - 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 + def _get_disco_api_info(self, url, display_id, disco_host, realm, country): + geo_countries = [country.upper()] + self._initialize_geo_bypass({ + 'countries': geo_countries, + }) + disco_base = 'https://%s/' % disco_host + token = self._download_json( + disco_base + 'token', display_id, 'Downloading token', + query={ + 'realm': realm, + })['data']['attributes']['token'] + headers = { + 'Referer': url, + 'Authorization': 'Bearer ' + token, + } + video = self._download_json( + disco_base + 'content/videos/' + display_id, display_id, + headers=headers, query={ + 'fields[channel]': 'name', + 'fields[image]': 'height,src,width', + 'fields[show]': 'name', + 'fields[tag]': 'name', + 'fields[video]': 'description,episodeNumber,name,publishStart,seasonNumber,videoDuration', + 'include': 'images,primaryChannel,show,tags' + }) + video_id = video['data']['id'] + info = video['data']['attributes'] + title = info['name'].strip() + formats = [] + try: + streaming = self._download_json( + 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): + continue + format_url = format_dict.get('url') + if not format_url: + continue + ext = determine_ext(format_url) + if format_id == 'dash' or ext == 'mpd': + formats.extend(self._extract_mpd_formats( + format_url, display_id, mpd_id='dash', fatal=False)) + elif format_id == 'hls' or ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + format_url, display_id, 'mp4', + entry_protocol='m3u8_native', m3u8_id='hls', + fatal=False)) else: - video_id = try_get(info, lambda x: x['data']['id']) - - if not info: - info_url = self._search_regex( - r'url\s*[:=]\s*["\']((?:https?:)?//[^/]+/playback/videoPlaybackInfo/\d+)', - webpage, '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, - }) - 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') - - series = self._html_search_regex( - r'(?s)]+class=["\'].*?\bshow_title\b.*?["\'][^>]*>(.+?)', - webpage, 'series', fatal=False) - episode = self._search_regex( - r']+class=["\'].*?\bdesc_ep\b.*?["\'][^>]*>\s*
\s*([^<]+)', - webpage, 'episode', fatal=False) + formats.append({ + 'url': format_url, + 'format_id': format_id, + }) + self._sort_formats(formats) - mobj = re.search( - r'(?s)]+class=["\']dates["\'][^>]*>.+?\bS\.(?P\d+)\s+E\.(?P\d+)\s*-\s*(?P\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 + creator = series = None + tags = [] + thumbnails = [] + included = video.get('included') or [] + if isinstance(included, list): + for e in included: + attributes = e.get('attributes') + if not attributes: + 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 { - 'id': compat_str(video_id or display_id), + 'id': video_id, 'display_id': display_id, 'title': title, - 'description': self._og_search_description(webpage), - 'thumbnail': self._og_search_thumbnail(webpage), + 'description': info.get('description'), + 'duration': float_or_none(info.get('videoDuration'), 1000), + 'timestamp': unified_timestamp(info.get('publishStart')), 'series': series, - 'season_number': season_number, - 'episode': episode, - 'episode_number': episode_number, - 'upload_date': upload_date, + 'season_number': int_or_none(info.get('seasonNumber')), + 'episode_number': int_or_none(info.get('episodeNumber')), + 'creator': creator, + 'tags': tags, + 'thumbnails': thumbnails, 'formats': formats, } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + display_id = mobj.group('id') + domain = mobj.group('domain').lstrip('www.') + country = mobj.group('country') or mobj.group('subdomain_country') + host = 'disco-api.' + domain if domain.startswith('dplay.') else 'eu2-prod.disco-api.com' + return self._get_disco_api_info( + url, display_id, host, 'dplay' + country, country)