Merge 68ad5cb00d71c50b2cabaadad43f83b0ba8792f3 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

* thx seproDev, yt-dlp/yt-dlp/pull/12694/commits/ae69e3c
* also add documenting comment in `InfoExtractor`

youtube_dl/extractor/bokecc.py

@@ -32,7 +32,7 @@ class BokeCCBaseIE(InfoExtractor):
 
 
 class BokeCCIE(BokeCCBaseIE):
-    _IE_DESC = 'CC视频'
+    IE_DESC = 'CC视频'
     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
 
     _TESTS = [{

youtube_dl/extractor/cloudy.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec'
+    IE_DESC = 'cloudy.ec'
     _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',

youtube_dl/extractor/common.py

@@ -422,6 +422,8 @@ class InfoExtractor(object):
     _GEO_COUNTRIES = None
     _GEO_IP_BLOCKS = None
     _WORKING = True
+    # supply this in public subclasses: used in supported sites list, etc
+    # IE_DESC = 'short description of IE'
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""

youtube_dl/extractor/extractors.py

@@ -954,6 +954,7 @@ from .playfm import PlayFMIE
 from .playplustv import PlayPlusTVIE
 from .plays import PlaysTVIE
 from .playstuff import PlayStuffIE
+from .playsuisse import PlaySuisseIE
 from .playtvak import PlaytvakIE
 from .playvid import PlayvidIE
 from .playwire import PlaywireIE

youtube_dl/extractor/itv.py

@@ -35,15 +35,6 @@ from ..utils import (
 
 class ITVBaseIE(InfoExtractor):
 
-    def _search_nextjs_data(self, webpage, video_id, **kw):
-        transform_source = kw.pop('transform_source', None)
-        fatal = kw.pop('fatal', True)
-        return self._parse_json(
-            self._search_regex(
-                r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
-                webpage, 'next.js data', group='js', fatal=fatal, **kw),
-            video_id, transform_source=transform_source, fatal=fatal)
-
     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
         if errnote is False:
             return False
@@ -109,7 +100,9 @@ class ITVBaseIE(InfoExtractor):
 
 class ITVIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
-    _IE_DESC = 'ITVX'
+    IE_DESC = 'ITVX'
+    _WORKING = False
+
     _TESTS = [{
         'note': 'Hub URLs redirect to ITVX',
         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
@@ -270,7 +263,7 @@ class ITVIE(ITVBaseIE):
                 'ext': determine_ext(href, 'vtt'),
             })
 
-        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
+        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default={})
         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
         info = self._og_extract(webpage, require_title=not title)
@@ -323,7 +316,7 @@ class ITVIE(ITVBaseIE):
 
 class ITVBTCCIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
-    _IE_DESC = 'ITV articles: News, British Touring Car Championship'
+    IE_DESC = 'ITV articles: News, British Touring Car Championship'
     _TESTS = [{
         'note': 'British Touring Car Championship',
         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',

youtube_dl/extractor/playsuisse.py

@@ -0,0 +1,299 @@
+# coding: utf-8
+from __future__ import unicode_literals
+import json
+
+from .common import InfoExtractor
+from ..utils import std_headers
+
+
+class PlaySuisseIE(InfoExtractor):
+    _MEDIA_URL = 'https://4bbepzm4ef.execute-api.eu-central-1.amazonaws.com/prod/graphql'
+    _VALID_URL = r'https?://(?:www\.)?playsuisse\.ch/watch/(?P<id1>[0-9]+)'
+    _TESTS = [
+        {
+            'url': 'https://www.playsuisse.ch/watch/763211/0',
+            'md5': '0d716b7a16c3e6ab784ef817ee9a20c1',
+            'info_dict': {
+                'id': '763211',
+                'ext': 'mp4',
+                'title': 'Wilder S01E01 - Knochen',
+                'description': 'md5:8ea7a8076ba000cd9e8bc132fd0afdd8'
+            }
+        },
+        {
+            'url': 'https://www.playsuisse.ch/watch/808675/0',
+            'md5': '7aa043e69fea5044db2da8bb58bca239',
+            'info_dict': {
+                'id': '808675',
+                'ext': 'mp4',
+                'title': 'Der Läufer',
+                'description': 'md5:'
+            }
+        },
+        {
+            'url': 'https://www.playsuisse.ch/watch/817913/0',
+            'md5': '50721c46ca0b3a9836eb61ecb0ed7097',
+            'info_dict': {
+                'id': '42',
+                'ext': 'mp4',
+                'title': 'Nr. 47 S01E01 - Die Einweihungsparty',
+                'description': 'md5:'
+            }
+        }
+    ]
+
+    def _get_media_data(self, media_id):
+        locale = std_headers.get('locale', 'de').strip()
+        # TODO find out why the locale has no effect in request
+        response = self._download_json(
+            self._MEDIA_URL,
+            media_id,
+            data=json.dumps({
+                'operationName': 'AssetWatch',
+                'query': self._GRAPHQL_QUERY,
+                'variables': {
+                    "assetId": media_id
+                }
+            }).encode('utf-8'),
+            headers={'Content-Type': 'application/json', 'locale': locale})
+
+        return response['data']['asset']
+
+    def _real_extract(self, url):
+        media_id, = self._VALID_URL_RE.match(url).groups()
+        media_data = self._get_media_data(media_id)
+
+        def _get_title(data):
+            if data.get('seriesName'):
+                return '{} S{:02}E{:02} - {}'.format(
+                    data['seriesName'],
+                    int(data['seasonNumber']),
+                    int(data['episodeNumber']),
+                    data['name'])
+            else:
+                return data['name']
+
+        title = _get_title(media_data)
+        description = media_data['description']
+        thumbnails = [
+            {
+                'id': thumb['id'],
+                'url': thumb['url']
+            }
+            for key, thumb in media_data.items()
+            if key.startswith('thumbnail') and thumb is not None
+        ]
+
+        formats = []
+
+        for media in media_data['medias']:
+            # NOTE subtitle info is included in the m3u8 file, but it's not supported by ytdl
+            # https://github.com/ytdl-org/youtube-dl/issues/6106
+            if media['type'] == 'HLS':
+                formats.extend(self._extract_m3u8_formats(
+                    media['url'],
+                    media_id,
+                    'mp4',
+                    'm3u8_native',
+                    m3u8_id="HLS",
+                    fatal=False))
+
+            # NOTE seems to be 404 for all tested media
+            elif media['type'] == 'DASH' and False:
+                formats.extend(self._extract_mpd_formats(
+                    media['url'],
+                    media_id,
+                    mpd_id='dash',
+                    fatal=False
+                ))
+
+        return {
+            'id': media_id,
+            'title': title,
+            'description': description,
+            'thumbnails': thumbnails,
+            'formats': formats,
+        }
+
+    _GRAPHQL_QUERY = '''\
+query AssetWatch($assetId: ID!) {
+  asset(assetId: $assetId) {
+    ...Asset
+    __typename
+  }
+}
+fragment Asset on Asset {
+  ...AssetDetails
+  episodes {
+    ...AssetDetails
+    __typename
+  }
+  __typename
+}
+fragment AssetDetails on Asset {
+  audioLanguages
+  awards
+  bu
+  contentCategories
+  contentCodes
+  contentTypes
+  contractType
+  countries
+  creators
+  creditsTimeInSecs
+  description
+  descriptionLong
+  directors
+  downloadable
+  duration
+  editorialContentCategoriesDatalab {
+    id
+    title
+    __typename
+  }
+  editorialContentMetaCategoriesDatalab {
+    id
+    title
+    __typename
+  }
+  endDate
+  episodeNumber
+  episodesInSequence
+  externalId
+  firstEpisodeDuration
+  id
+  image16x9 {
+    ...ImageDetails
+    __typename
+  }
+  image2x3 {
+    ...ImageDetails
+    __typename
+  }
+  image16x9WithTitle {
+    ...ImageDetails
+    __typename
+  }
+  image2x3WithTitle {
+    ...ImageDetails
+    __typename
+  }
+  mainCast
+  name
+  nextEpisode {
+    id
+    episodeNumber
+    seasonNumber
+    numberOfEpisodesInSeason
+    image16x9 {
+      ...ImageDetails
+      __typename
+    }
+    __typename
+  }
+  numberOfSeasons
+  otherKeyPeople
+  parentalRating
+  popularity
+  premium
+  presenters
+  primaryLanguage
+  productionCompanies
+  productionCountries
+  provider
+  ratings
+  regions
+  restrictions
+  seasons {
+    seasonNumber
+    assetIds
+    __typename
+  }
+  seasonNumber
+  seriesId
+  seriesName
+  nextEpisode {
+    ...NextEpisodeDetails
+    __typename
+  }
+  parentId
+  startDate
+  subtitleLanguages
+  tagline
+  targetAudience
+  themes
+  thumbnail16x9 {
+    ...ImageDetails
+    __typename
+  }
+  thumbnail2x3 {
+    ...ImageDetails
+    __typename
+  }
+  thumbnail16x9WithTitle {
+    ...ImageDetails
+    __typename
+  }
+  thumbnail2x3WithTitle {
+    ...ImageDetails
+    __typename
+  }
+  type
+  writers
+  year
+  medias {
+    ...MediaDetails
+    __typename
+  }
+  trailerMedias {
+    ...MediaDetails
+    __typename
+  }
+  sponsors {
+    ...SponsorDetails
+    __typename
+  }
+  sponsorEndDate
+  __typename
+}
+fragment ImageDetails on Image {
+  id
+  url
+  alt
+  __typename
+}
+fragment MediaDetails on Media {
+  id
+  type
+  url
+  duration
+  __typename
+}
+fragment SponsorDetails on Sponsor {
+  id
+  name
+  description
+  type
+  externalId
+  image16x9 {
+    ...ImageDetails
+    __typename
+  }
+  thumbnail16x9 {
+    ...ImageDetails
+    __typename
+  }
+  __typename
+}
+fragment NextEpisodeDetails on NextEpisode {
+  id
+  episodeNumber
+  seasonNumber
+  numberOfEpisodesInSeason
+  image16x9 {
+    ...ImageDetails
+    __typename
+  }
+  __typename
+}
+'''

youtube_dl/extractor/senateisvp.py

@@ -47,7 +47,7 @@ class SenateISVPIE(InfoExtractor):
         ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
         ['arch', '', 'http://ussenate-f.akamaihd.net/']
     ]
-    _IE_NAME = 'senate.gov'
+    IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
         'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png',

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError: