Merge 4232427a8d4615085c3229b88c09d6adb90d2922 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

@@ -707,6 +707,7 @@ from .mixcloud import (
     MixcloudUserIE,
     MixcloudPlaylistIE,
 )
+from .megacartoons import MegaCartoonsIE
 from .mlb import (
     MLBIE,
     MLBVideoIE,

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/megacartoons.py

@@ -0,0 +1,110 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from ..utils import (
+    bug_reports_message,
+    JSON_LD_RE,
+    merge_dicts,
+    NO_DEFAULT,
+    RegexNotFoundError,
+    try_get,
+    url_or_none,
+)
+from ..compat import compat_str
+
+from .common import InfoExtractor
+
+
+class MegaCartoonsIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?megacartoons\.net/(?P<id>[a-zA-Z\d-]+)/'
+    _TESTS = [{
+        'url': 'https://www.megacartoons.net/help-wanted/',
+        'md5': '4ba9be574f9a17abe0c074e2f955fded',
+        'info_dict': {
+            'id': 'help-wanted',
+            'ext': 'mp4',
+            'title': 'Help Wanted - SpongeBob SquarePants',
+            'upload_date': '20200223',
+            'timestamp': 1582416000,
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'description': 'md5:2c909daa6c6cb16b2d4d791dd1a31632'
+        }
+    }, {
+        'url': 'https://www.megacartoons.net/1000-years-of-courage/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.megacartoons.net/911-2/',
+        'only_matching': True,
+    }]
+
+    # adapted from common.py pending yt-dlp back-port
+    def _search_json_ld(self, html, video_id, expected_type=None, **kwargs):
+        json_ld_list = list(re.finditer(JSON_LD_RE, html))
+        default = kwargs.get('default', NO_DEFAULT)
+        fatal = kwargs.get('fatal', True) if default is NO_DEFAULT else False
+        json_ld = []
+        for mobj in json_ld_list:
+            json_ld_item = self._parse_json(
+                mobj.group('json_ld'), video_id, fatal=fatal)
+            if not json_ld_item:
+                continue
+            if isinstance(json_ld_item, dict):
+                json_ld.append(json_ld_item)
+            elif isinstance(json_ld_item, (list, tuple)):
+                json_ld.extend(json_ld_item)
+        if json_ld:
+            # handle initial '@graph' with one level of children
+            if len(json_ld) > 0 and '@graph' in json_ld[0] and '@context' in json_ld[0]:
+                # should always be hit here
+                context = json_ld[0]['@context']
+                json_ld_g = json_ld[0]['@graph'] or []
+                for item in json_ld_g:
+                    item.setdefault('@context', context)
+                json_ld = json_ld_g + json_ld[1:]
+            json_ld = self._json_ld(json_ld, video_id, fatal=fatal, expected_type=expected_type)
+        if json_ld:
+            return json_ld
+        if default is not NO_DEFAULT:
+            return default
+        elif fatal:
+            raise RegexNotFoundError('Unable to extract JSON-LD')
+        else:
+            self.report_warning('unable to extract JSON-LD %s' % bug_reports_message())
+            return {}
+
+    def _real_extract(self, url):
+        # ID is equal to the episode name
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        info = self._search_json_ld(webpage, video_id, fatal=False) or {}
+
+        info.update({
+            'id': video_id,
+            # Try to find a good title or fallback to the ID
+            'title': info.get('title') or self._og_search_title(webpage) or video_id.replace('-', ' ').capitalize(),
+        })
+
+        if 'url' not in info or 'thumbnail' not in info:
+            # Video data is stored in a json -> extract it from the raw html
+            url_json = self._parse_json(self._html_search_regex(r'''<div\b[^>]+\bdata-item\s*=\s*(["'])(?P<videourls>\{.*})\1''', webpage, 'videourls', group='videourls', default='{}'), video_id, fatal=False) or {}
+
+            video_url = url_or_none(try_get(url_json, lambda x: x['sources'][0]['src'], compat_str) or self._og_search_video_url(webpage))   # Get the video url
+            video_thumbnail = url_or_none(url_json.get('splash') or self._og_search_thumbnail(webpage))            # Get the thumbnail
+            info = merge_dicts(info, {
+                'url': video_url,
+                'thumbnail': video_thumbnail,
+            })
+
+        # Find the <article> class in the html
+        article = self._search_regex(
+            r'(?s)<article\b[^>]*?\bclass\s*=\s*[^>]*?\bpost\b[^>]*>(.+?)</article\b', webpage, 'post', default='')
+
+        # Extract the actual description from it
+        info['description'] = (
+            self._html_search_regex(r'(?s)<p>\s*([^<]+)\s*</p>', article, 'videodescription', fatal=False)
+            or self._og_search_description(webpage))
+
+        return info

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: