Merge 156c58b07f7c66c98aea45762c3e26ff55987990 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

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

youtube_dl/extractor/atvat.py

@@ -12,15 +12,15 @@ from ..utils import (
 class ATVAtIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?atv\.at/(?:[^/]+/){2}(?P<id>[dv]\d+)'
     _TESTS = [{
-        'url': 'http://atv.at/aktuell/di-210317-2005-uhr/v1698449/',
-        'md5': 'c3b6b975fb3150fc628572939df205f2',
+        'url': 'https://www.atv.at/bauer-sucht-frau-die-zweite-chance/folge-1/d3390693/',
+        'md5': 'c471605591009dfb6e6c54f7e62e2807',
         'info_dict': {
-            'id': '1698447',
+            'id': '3390684',
             'ext': 'mp4',
-            'title': 'DI, 21.03.17 | 20:05 Uhr 1/1',
+            'title': 'Bauer sucht Frau - Die zweite Chance Folge 1',
         }
     }, {
-        'url': 'http://atv.at/aktuell/meinrad-knapp/d8416/',
+        'url': 'https://www.atv.at/bauer-sucht-frau-staffel-17/fuenfte-eventfolge/d3339537/',
         'only_matching': True,
     }]
 
@@ -28,48 +28,52 @@ class ATVAtIE(InfoExtractor):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
         video_data = self._parse_json(unescapeHTML(self._search_regex(
-            [r'flashPlayerOptions\s*=\s*(["\'])(?P<json>(?:(?!\1).)+)\1',
-             r'class="[^"]*jsb_video/FlashPlayer[^"]*"[^>]+data-jsb="(?P<json>[^"]+)"'],
+            r'var\splaylist\s*=\s*(?P<json>\[.*\]);',
             webpage, 'player data', group='json')),
-            display_id)['config']['initial_video']
+            display_id)
 
-        video_id = video_data['id']
-        video_title = video_data['title']
+        first_video = video_data[0]
+        video_id = first_video['id']
+        video_title = first_video.get('tvShowTitle', first_video['title'])
 
-        parts = []
-        for part in video_data.get('parts', []):
-            part_id = part['id']
-            part_title = part['title']
+        def process_source_entry(self, source, part_id):
+            source_url = source.get('url')
+            if not source_url:
+                return None
+            ext = determine_ext(source_url)
+            if ext == 'm3u8':
+                return self._extract_m3u8_formats(
+                    source_url, part_id, 'mp4', 'm3u8_native',
+                    m3u8_id='hls', fatal=False)
+            else:
+                return {
+                    'url': source_url,
+                }
 
-            formats = []
-            for source in part.get('sources', []):
-                source_url = source.get('src')
-                if not source_url:
+        def process_entry(self, entry):
+            part_id = entry['id']
+            formats = [process_source_entry(self, source, part_id) for source in entry["sources"]]
+            formats_flat = []
+            for f in formats:
+                if f is None:
                     continue
-                ext = determine_ext(source_url)
-                if ext == 'm3u8':
-                    formats.extend(self._extract_m3u8_formats(
-                        source_url, part_id, 'mp4', 'm3u8_native',
-                        m3u8_id='hls', fatal=False))
+                elif type(f) is list:
+                    formats_flat.extend(f)
                 else:
-                    formats.append({
-                        'format_id': source.get('delivery'),
-                        'url': source_url,
-                    })
-            self._sort_formats(formats)
+                    formats_flat.append(f)
+            self._sort_formats(formats_flat)
+            return {
+                'id': entry['id'],
+                'title': entry['title'],
+                'duration': int_or_none(entry.get('duration')),
+                'formats': formats_flat
+            }
 
-            parts.append({
-                'id': part_id,
-                'title': part_title,
-                'thumbnail': part.get('preview_image_url'),
-                'duration': int_or_none(part.get('duration')),
-                'is_live': part.get('is_livestream'),
-                'formats': formats,
-            })
+        entries = [process_entry(self, entry) for entry in video_data]
 
         return {
             '_type': 'multi_video',
             'id': video_id,
             'title': video_title,
-            'entries': parts,
+            'entries': entries,
         }

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/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/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: