Merge 2103cec4a5b3931c9a5d31dd9aed673773bba26e 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

@@ -737,6 +737,7 @@ from .mtv import (
 )
 from .muenchentv import MuenchenTVIE
 from .mwave import MwaveIE, MwaveMeetGreetIE
+from .mxplayer import MxplayerIE
 from .mychannels import MyChannelsIE
 from .myspace import MySpaceIE, MySpaceAlbumIE
 from .myspass import MySpassIE

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/mxplayer.py

@@ -0,0 +1,102 @@
+from __future__ import unicode_literals
+
+import re
+from urllib.parse import urljoin
+from .common import InfoExtractor
+from ..utils import (
+    url_or_none, js_to_json, ExtractorError)
+
+
+# VALID_STREAMS = ('dash', 'hls', )
+VALID_STREAMS = ('dash', )
+
+
+class MxplayerIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?mxplayer\.in/movie/(?P<slug>[a-z0-9]+(?:-[a-z0-9]+)*)'
+    # _VALID_URL = r'https?://(?:www\.)?mxplayer\.in/movie/(?P<title>.*)[-](?P<id>.+)$'
+    _TEST = {
+        'url': 'https://www.mxplayer.in/movie/watch-knock-knock-hindi-dubbed-movie-online-b9fa28df3bfb8758874735bbd7d2655a?watch=true',
+        'info_dict': {
+            'id': 'b9fa28df3bfb8758874735bbd7d2655a',
+            'ext': 'mp4',
+            'title': 'Knock Knock Movie | Watch 2015 Knock Knock Full Movie Online- MX Player',
+        },
+        'params': {
+            'skip_download': True,
+            'format': 'bestvideo+bestaudio'
+        }
+    }
+
+    def _get_best_stream_url(self, stream):
+        best_stream = list(filter(None, [v for k, v in stream.items()]))
+        return best_stream.pop(0) if len(best_stream) else None
+
+    def _get_stream_urls(self, video_dict):
+        stream_dict = video_dict.get('stream', {'provider': {}})
+        stream_provider = stream_dict.get('provider')
+
+        if not stream_dict[stream_provider]:
+            message = 'No stream provider found'
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+
+        streams = []
+        for stream_name, v in stream_dict[stream_provider].items():
+            if stream_name in VALID_STREAMS:
+                stream_url = self._get_best_stream_url(v)
+                if stream_url is None:
+                    continue
+                streams.append((stream_name, stream_url))
+        return streams
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        video_slug = mobj.group('slug')
+
+        video_id = video_slug.split('-')[-1]
+
+        webpage = self._download_webpage(url, video_id)
+
+        window_state_json = self._html_search_regex(
+            r'(?s)<script>window\.state\s*[:=]\s(\{.+\})\n(\w+).*(</script>).*',
+            webpage, 'WindowState')
+
+        source = self._parse_json(js_to_json(window_state_json), video_id)
+        if not source:
+            message = 'source not found'
+            raise ExtractorError('%s said: %s' % (self.IE_NAME, message), expected=True)
+
+        config_dict = source['config']
+        video_dict = source['entities'][video_id]
+        stream_urls = self._get_stream_urls(video_dict)
+
+        title = self._og_search_title(webpage, fatal=True, default=video_dict['title'])
+
+        formats = []
+        headers = {'Referer': url}
+        for stream_name, stream_url in stream_urls:
+            if stream_name == 'dash':
+                format_url = url_or_none(urljoin(config_dict['videoCdnBaseUrl'], stream_url))
+                if not format_url:
+                    continue
+                formats.extend(self._extract_mpd_formats(
+                    format_url, video_id, mpd_id='dash', headers=headers))
+
+        self._sort_formats(formats)
+        info = {
+            'id': video_id,
+            'ext': 'mpd',
+            'title': title,
+            'description': video_dict.get('description'),
+            'formats': formats
+        }
+
+        if video_dict.get('imageInfo'):
+            info['thumbnails'] = list(map(lambda i: dict(i, **{
+                'url': urljoin(config_dict['imageBaseUrl'], i['url'])
+            }), video_dict['imageInfo']))
+
+        if video_dict.get('webUrl'):
+            last_part = video_dict['webUrl'].split("/")[-1]
+            info['display_id'] = last_part.replace(video_id, "").rstrip("-")
+
+        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: