Merge cddc1cb9b8f0883890d39decc0343fd42e4dc2e1 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/douyin.py

@@ -0,0 +1,78 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import json
+
+from .common import (
+    InfoExtractor,
+    compat_urllib_parse_unquote,
+)
+
+
+class DouyinIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?douyin\.com/video/(?P<id>[0-9]+)'
+    _TESTS = [{
+        'url': 'https://www.douyin.com/video/6961737553342991651',
+        'md5': 'f0114e6688442972d80fab1083543197',
+        'info_dict': {
+            'id': '6961737553342991651',
+            'ext': 'mp4',
+            'title': '#杨超越  小小水手带你去远航❤️',
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6982497745948921092',
+        'md5': 'bdc8a6b4ce22c887e0064d2813befa27',
+        'info_dict': {
+            'id': '6982497745948921092',
+            'ext': 'mp4',
+            'title': '这个夏日和小羊@杨超越 一起遇见白色幻想',
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6953975910773099811',
+        'md5': 'dde3302460f19db59c47060ff013b902',
+        'info_dict': {
+            'id': '6953975910773099811',
+            'ext': 'mp4',
+            'title': '#一起看海  出现在你的夏日里',
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6950251282489675042',
+        'md5': 'f61844399e85f0ff18cfab91d20fe326',
+        'info_dict': {
+            'id': '6950251282489675042',
+            'ext': 'mp4',
+            'title': '哈哈哈，成功了哈哈哈哈哈哈',
+        }
+    }, {
+        'url': 'https://www.douyin.com/video/6963263655114722595',
+        'md5': '30a852b6efb232df202b80894f51422d',
+        'info_dict': {
+            'id': '6963263655114722595',
+            'ext': 'mp4',
+            'title': '#哪个爱豆的105度最甜 换个角度看看我哈哈',
+        }
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+        # The video player and video source are rendered client-side, the data
+        # contains in a <script id="RENDER_DATA" type="application/json"> tag
+        # quoted, unquote the whole page content then search using regex with
+        # regular string.
+        webpage = compat_urllib_parse_unquote(webpage)
+
+        # As of today, this seems good enough to pinpoint the title
+        title = self._html_search_regex(r'"desc":"([^"]*)"', webpage, 'title')
+
+        # video URLs are in this pattern {"src":"THE_URL"}, in json format,
+        # as a list of CDN urls, all serving the same file
+        urls = json.loads(
+            self._html_search_regex(r'"playAddr":(\[.*?\])', webpage, 'urls'))
+
+        return {
+            'id': video_id,
+            'title': title,
+            'url': 'https:' + urls[0]['src'],
+            'ext': 'mp4',
+        }

youtube_dl/extractor/extractors.py

@@ -334,6 +334,7 @@ from .discoverynetworks import DiscoveryNetworksDeIE
 from .discoveryvr import DiscoveryVRIE
 from .disney import DisneyIE
 from .dispeak import DigitallySpeakingIE
+from .douyin import DouyinIE
 from .dropbox import DropboxIE
 from .dw import (
     DWIE,

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: