Merge 69a40d3eb06316ac944aea2a0d28d2ce15d447cd 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

@@ -1154,7 +1154,10 @@ from .slideshare import SlideshareIE
 from .slideslive import SlidesLiveIE
 from .slutload import SlutloadIE
 from .snotr import SnotrIE
-from .sohu import SohuIE
+from .sohu import (
+    SohuIE,
+    SohuPlaylistIE,
+)
 from .sonyliv import SonyLIVIE
 from .soundcloud import (
     SoundcloudEmbedIE,

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/extractor/sohu.py

@@ -1,6 +1,7 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import math
 import re
 
 from .common import InfoExtractor
@@ -11,12 +12,13 @@ from ..compat import (
 from ..utils import (
     ExtractorError,
     int_or_none,
-    try_get,
+    try_get, NO_DEFAULT, HTMLAttributeParser,
 )
 
 
 class SohuIE(InfoExtractor):
-    _VALID_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
+    _VALID_URL = r'https?://(?:my\.)?tv\.sohu\.com/.+?/.+(?:\.html|\.shtml).*?'
+    _VALID_URL_OG_URL = r'https?://(?P<mytv>my\.)?tv\.sohu\.com/.+?/(?(mytv)|n)(?P<id>\d+)\.shtml.*?'
 
     # Sohu videos give different MD5 sums on Travis CI and my machine
     _TESTS = [{
@@ -82,6 +84,58 @@ class SohuIE(InfoExtractor):
         'params': {
             'skip_download': True
         }
+    }, {
+        'url': 'https://tv.sohu.com/v/dXMvMjMyNzk5ODg5Lzc4NjkzNDY0LnNodG1s.html',
+        'info_dict': {
+            'id': '78693464',
+            'ext': 'mp4',
+            'title': '【爱范品】第31期：MWC见不到的奇葩手机',
+        }
+    }, {
+        'note': 'Video in issue #18542: https://github.com/ytdl-org/youtube-dl/issues/18542',
+        'url': 'https://tv.sohu.com/v/MjAxNzA3MTMvbjYwMDA1MzM4MS5zaHRtbA==.html',
+        'info_dict': {
+            'id': '600053381',
+            'ext': 'mp4',
+            'title': '试听：侯旭《孤鸟》',
+        },
+    }, {
+        'note': 'Video in issue #28944: https://github.com/ytdl-org/youtube-dl/issues/28944',
+        'url': 'https://tv.sohu.com/v/dXMvNTAyMjA5MTMvNjg1NjIyNTYuc2h0bWw=.html?src=pl',
+        'info_dict': {
+            'id': '68562256',
+            'ext': 'mp4',
+            'title': "Cryin'  [HD 1080p]  Chris.Botti(feat. Steven Tyler",
+        },
+    }, {
+        'note': 'Multipart video with new address format',
+        'url': 'https://tv.sohu.com/v/dXMvMjQyNTYyMTYzLzc4OTEwMzM5LnNodG1s.html?src=pl',
+        'info_dict': {
+            'id': '78910339',
+            'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '78910339_part1',
+                'ext': 'mp4',
+                'duration': 294,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }, {
+            'info_dict': {
+                'id': '78910339_part2',
+                'ext': 'mp4',
+                'duration': 300,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }, {
+            'info_dict': {
+                'id': '78910339_part3',
+                'ext': 'mp4',
+                'duration': 150,
+                'title': '【神探苍实战秘籍】第13期 战争之影 赫卡里姆',
+            }
+        }]
     }]
 
     def _real_extract(self, url):
@@ -97,7 +151,12 @@ class SohuIE(InfoExtractor):
                 'Downloading JSON data for %s' % vid_id,
                 headers=self.geo_verification_headers())
 
-        mobj = re.match(self._VALID_URL, url)
+        mobj = re.match(self._VALID_URL_OG_URL, url)
+        if mobj is None:
+            webpage = self._download_webpage(url, '')
+            url = self._og_search_url(webpage)
+            mobj = re.match(self._VALID_URL_OG_URL, url)
+
         video_id = mobj.group('id')
         mytv = mobj.group('mytv') is not None
 
@@ -200,3 +259,65 @@ class SohuIE(InfoExtractor):
             }
 
         return info
+
+
+class SohuPlaylistIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:my\.)?tv\.sohu\.com/pl/(?P<pl_id>\d+)$'
+    _URL_IN_PLAYLIST = re.compile(r'<strong>.*?</strong>')
+    parser = HTMLAttributeParser()
+    _TESTS = [{
+        'url': 'https://my.tv.sohu.com/pl/9637148',
+        'info_dict': {
+            'title': '【语文大师】初中必背、常考经典古诗词',
+            'id': '9637148',
+        },
+        'playlist_count': 70,
+    }, {
+        'url': 'https://my.tv.sohu.com/pl/9700995',
+        'info_dict': {
+            'title': '牛人游戏',
+            'id': '9700995',
+        },
+        'playlist_count': 198,
+    },
+    ]
+
+    def _real_extract(self, url):
+        mobj = re.match(self._VALID_URL, url)
+        playlist_id = mobj.group('pl_id')
+
+        webpage = self._download_webpage(url, playlist_id)
+        title = self._get_playlist_title(webpage)
+        all_pages = self._get_all_pages_in_playlist(webpage, url)
+        video_list = self._get_video_list(all_pages, playlist_id)
+        playlist = self.playlist_result(self._entries(video_list), playlist_id, title)
+
+        return playlist
+
+    def _get_playlist_title(self, webpage):
+        title = self._search_regex(r'<title>(.*?)</title>', webpage, 'title')
+        return re.sub(r'(?:^栏目：| -搜狐视频$)', '', title)
+
+    def _entries(self, video_list):
+        entries = []
+        for mobj in re.finditer(self._URL_IN_PLAYLIST, video_list):
+            self.parser.feed(mobj.group(0))
+            url = self.parser.attrs['href']
+            title = self.parser.attrs['title']
+            entry = self.url_result(url, SohuIE.ie_key(), '', title)
+            entries.append(entry)
+        return entries
+
+    def _get_all_pages_in_playlist(self, first_page, url):
+        pgcount = int(self._search_regex(r'var pgcount = \'(\d+)\'', first_page, 'pgcount'))
+        pgsize = int(self._search_regex(r'var pgsize = \'(\d+)\'', first_page, 'pgsize'))
+        return [url + '/index%d.shtml' % (i + 1) for i in range(0, math.ceil(pgcount / pgsize))]
+
+    def _get_video_list(self, all_pages, playlist_id):
+        video_list = ''
+        for i, url in enumerate(all_pages):
+            webpage = self._download_webpage(url, "playlist " + playlist_id + " page: %d" % (1 + i))
+            video_list += self._search_regex(
+                r'<ul class="uList cfix">(.*?)</ul>',
+                webpage, 'video list', NO_DEFAULT, True, re.DOTALL)
+        return video_list

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: