]> Raphaël G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/dplay.py
New upstream version 2020.06.16.1
[youtubedl] / youtube_dl / extractor / dplay.py
index 5790553f38ca29107bad44317fedb271dce0883a..a7b9db5689c4df697f1606a343995ef3d29ee03a 100644 (file)
 # coding: utf-8
 from __future__ import unicode_literals
 
-import json
 import re
-import time
 
 from .common import InfoExtractor
-from ..compat import compat_urlparse
+from ..compat import compat_HTTPError
 from ..utils import (
+    determine_ext,
+    ExtractorError,
+    float_or_none,
     int_or_none,
-    update_url_query,
+    unified_timestamp,
 )
 
 
 class DPlayIE(InfoExtractor):
-    _VALID_URL = r'https?://(?P<domain>it\.dplay\.com|www\.dplay\.(?:dk|se|no))/[^/]+/(?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 = [{
-        # geo restricted, via direct unsigned hls URL
-        'url': 'http://it.dplay.com/take-me-out/stagione-1-episodio-25/',
-        'info_dict': {
-            'id': '1255600',
-            'display_id': 'stagione-1-episodio-25',
-            'ext': 'mp4',
-            'title': 'Episodio 25',
-            'description': 'md5:cae5f40ad988811b197d2d27a53227eb',
-            'duration': 2761,
-            'timestamp': 1454701800,
-            'upload_date': '20160205',
-            'creator': 'RTIT',
-            'series': 'Take me out',
-            'season_number': 1,
-            'episode_number': 25,
-            'age_limit': 0,
-        },
-        'expected_warnings': ['Unable to download f4m manifest'],
-    }, {
         # 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': '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': '104465',
+            'display_id': 'ted-bundy-mind-of-a-monster/ted-bundy-mind-of-a-monster',
+            'ext': 'mp4',
+            '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,
+        },
+    }, {
+        # disco-api
+        'url': 'https://www.dplay.no/videoer/i-kongens-klr/sesong-1-episode-7',
+        'info_dict': {
+            'id': '40206',
+            'display_id': 'i-kongens-klr/sesong-1-episode-7',
+            'ext': 'mp4',
+            'title': 'Episode 7',
+            'description': 'md5:e3e1411b2b9aebeea36a6ec5d50c60cf',
+            'duration': 2611.16,
+            'timestamp': 1516726800,
+            'upload_date': '20180123',
+            'series': 'I kongens klær',
+            'season_number': 1,
+            'episode_number': 7,
+        },
+        'params': {
+            'format': 'bestvideo',
+            'skip_download': True,
+        },
+        'skip': 'Available for Premium users',
+    }, {
+        'url': 'http://it.dplay.com/nove/biografie-imbarazzanti/luigi-di-maio-la-psicosi-di-stanislawskij/',
+        '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': '70816',
-            'display_id': 'season-6-episode-12',
+            'id': '21652',
+            'display_id': 'la-fiebre-del-oro/temporada-8-episodio-1',
             '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': '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,
     }, {
-        # geo restricted, via direct unsigned hls URL
-        'url': 'http://www.dplay.no/pga-tour/season-1-hoydepunkter-18-21-februar/',
+        'url': 'https://www.dplay.jp/video/gold-rush/24086',
         'only_matching': True,
     }]
 
-    def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
-        display_id = mobj.group('id')
-        domain = mobj.group('domain')
-
-        webpage = self._download_webpage(url, display_id)
-
-        video_id = self._search_regex(
-            r'data-video-id=["\'](\d+)', webpage, 'video id')
-
-        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')
+    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 = []
-
-        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
-                query = compat_urlparse.parse_qs(compat_urlparse.urlparse(manifest_url).query)
-                for m3u8_format in m3u8_formats:
-                    m3u8_format['url'] = update_url_query(m3u8_format['url'], query)
-                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])
-
+        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:
+                formats.append({
+                    'url': format_url,
+                    'format_id': format_id,
+                })
         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})
+        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': 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')),
+            'description': info.get('description'),
+            'duration': float_or_none(info.get('videoDuration'), 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')),
+            'creator': creator,
+            'tags': tags,
+            'thumbnails': thumbnails,
             'formats': formats,
-            'subtitles': subtitles,
         }
+
+    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)