Merge a784be7395cbb0426b8111ca126e73fe4cf653cc into 4e714f9df1ed2cccd51df60d45ff5504abe827b7

* 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

@@ -1701,3 +1701,4 @@ from .zingmp3 import (
 )
 from .zoom import ZoomIE
 from .zype import ZypeIE
+from .invidious import InvidiousIE, InvidiousPlaylistIE

youtube_dl/extractor/invidious.py

@@ -0,0 +1,217 @@
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..compat import compat_urllib_parse
+
+INSTANCES = [
+    'y.com.sb',
+    'yt.artemislena.eu'
+]
+
+INSTANCES_HOST_REGEX = '(?:' + '|'.join([instance.replace('.', r'\.') for instance in INSTANCES]) + ')'
+
+
+class InvidiousIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?' + INSTANCES_HOST_REGEX + r'/watch\?v=(?P<id>.+)'
+    _TESTS = [
+        {
+            'url': 'https://y.com.sb/watch?v=xKTygGa6hg0',
+            'info_dict': {
+                'id': 'xKTygGa6hg0',
+                'ext': 'mp4',
+                'title': 'Coding in C++ - Creating a Player Controller - CRYENGINE Summer Academy S1E5 - [Tutorial]',
+                'uploader': 'CRYENGINE',
+                'uploader_id': 'UCtaXcIVFp8HEpthm7qwtKCQ',
+                'description': 'md5:7aa75816d40ffccdbf3e15a90b05fca3',
+            }
+        },
+        {
+            'url': 'https://yt.artemislena.eu/watch?v=BaW_jenozKc',
+            'md5': '5515885fed58607bfae88f7d2090bc93',
+            'info_dict': {
+                'id': 'BaW_jenozKc',
+                'ext': 'mp4',
+                'title': 'youtube-dl test video "\'/\\ä↭𝕐',
+                'uploader': 'Philipp Hagemeister',
+                'uploader_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
+                'channel_id': 'UCLqxVugv74EIW3VWh2NOa3Q',
+                'description': 'test chars:  "\'/\\ä↭𝕐\ntest URL: https://github.com/rg3/youtube-dl/issues/1892\n\nThis is a test video for youtube-dl.\n\nFor more information, contact phihag@phihag.de .',
+                'tags': ['youtube-dl'],
+                'duration': 10,
+                'view_count': int,
+                'like_count': int,
+                'dislike_count': int,
+            }
+        },
+    ]
+
+    def __init__(self, downloader=None):
+        super().__init__(downloader)
+
+    # type is either 'video' or 'audio'
+    # ext is the file extension
+    @staticmethod
+    def _get_additional_format_data(format_type, bitrate, resolution, fps):
+        out = {}
+
+        try:
+            type_and_ext, codecs = format_type.split(';')
+        except Exception:
+            pass
+
+        try:
+            type_, ext = type_and_ext.split('/')
+            # codec = codecs.split('"')[1]
+            out['ext'] = ext
+            # if type_ == 'audio':
+            #     out['acodec'] = codec
+            # elif type_ == 'video':
+            #     out['vcodec'] = codec
+        except Exception:
+            pass
+
+        try:
+            bitrate = float(bitrate) / 1000
+            # if type_ == 'audio':
+            #     out['abr'] = bitrate
+            # elif type_ == 'video':
+            #     out['vbr'] = bitrate
+            # out['tbr'] = bitrate
+        except Exception:
+            pass
+
+        try:
+            if type_ == 'audio':
+                out['resolution'] = type_and_ext + ' @ ' + str(bitrate) + 'k - audio only'
+            elif type_ == 'video':
+                out['resolution'] = resolution + ' - ' + type_and_ext + ' @ ' + str(fps) + 'fps - video only'
+        except Exception:
+            pass
+
+        return out
+
+    def _patch_url(self, url):
+        return compat_urllib_parse.urlparse(url)._replace(netloc=self.url_netloc).geturl()
+
+    def _get_formats(self, api_response):
+        all_formats = []
+
+        # Video/audio only
+        for format_ in api_response.get('adaptiveFormats') or []:
+            all_formats.append({
+                'url': self._patch_url(format_['url']),
+                'format_id': format_.get('itag'),
+                # 'fps': format_.get('fps'),
+                # 'container': format_.get('container')
+            } | InvidiousIE._get_additional_format_data(format_.get('type'), format_.get('bitrate'), format_.get('resolution'), format_.get('fps')))
+
+        # Both video and audio
+        for format_ in api_response.get('formatStreams') or []:
+            all_formats.append({
+                'url': self._patch_url(format_['url']),
+                'format_id': format_.get('itag'),
+                # 'fps': format_.get('fps'),
+                # 'container': format_.get('container')
+            } | InvidiousIE._get_additional_format_data(format_.get('type'), format_.get('bitrate'), format_.get('resolution'), format_.get('fps')))
+
+        return all_formats
+
+    def _get_thumbnails(self, api_response):
+        thumbnails = []
+        video_thumbnails = api_response.get('videoThumbnails') or []
+
+        for inversed_quality, thumbnail in enumerate(video_thumbnails):
+            thumbnails.append({
+                'id': thumbnail.get('quality'),
+                'url': thumbnail.get('url'),
+                'quality': len(video_thumbnails) - inversed_quality,
+                'width': thumbnail.get('width'),
+                'height': thumbnail.get('height')
+            })
+
+        return thumbnails
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = None
+
+        # host_url will contain `http[s]://example.com` where `example.com` is the used invidious instance.
+        url_parsed = compat_urllib_parse.urlparse(url)
+        self.url_netloc = url_parsed.netloc
+        host_url = url_parsed.scheme + '://' + url_parsed.netloc
+
+        api_response = self._download_json(host_url + '/api/v1/videos/' + video_id, video_id)
+
+        def download_webpage_and(fn, fatal=True):
+            global webpage
+            if webpage is None:
+                webpage = self._download_webpage(url, video_id, fatal=fatal)
+            return fn()
+
+        out = {
+            'id': video_id,
+            'title': api_response.get('title') or download_webpage_and(lambda: self._og_search_title(webpage)),
+            'description': api_response.get('description') or download_webpage_and(lambda: self._og_search_description(webpage)),
+
+            'release_timestamp': api_response.get('published'),
+
+            'uploader': api_response.get('author'),
+            'uploader_id': api_response.get('authorId'),
+            'channel': api_response.get('author'),
+            'channel_id': api_response.get('authorId'),
+            'channel_url': host_url + api_response.get('authorUrl'),
+
+            'duration': api_response.get('lengthSeconds'),
+
+            'view_count': api_response.get('viewCount'),
+            'like_count': api_response.get('likeCount'),
+            'dislike_count': api_response.get('dislikeCount'),
+
+            # 'isFamilyFriendly': 18 if api_response.get('isFamilyFriendly') == False else None
+
+            'tags': api_response.get('keywords'),
+            'is_live': api_response.get('liveNow'),
+
+            'formats': self._get_formats(api_response),
+            'thumbnails': self._get_thumbnails(api_response)
+        }
+
+        if api_response.get('isFamilyFriendly') is False:
+            out['age_limit'] = 18
+
+        return out
+
+
+class InvidiousPlaylistIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://)?(?:www\.)?' + INSTANCES_HOST_REGEX + r'/playlist\?list=(?P<id>.+)'
+    _TEST = {
+        'url': 'https://yt.artemislena.eu/playlist?list=PLowKtXNTBypGqImE405J2565dvjafglHU',
+        'md5': 'de4a9175071169961fe7cf2b6740da12',
+        'info_dict': {
+            'id': 'HyznrdDSSGM',
+            'ext': 'mp4',
+            'title': '8-bit computer update',
+            'uploader': 'Ben Eater',
+            'uploader_id': 'UCS0N5baNlQWJCUrhCEo8WlA',
+            'description': 'An update on my plans to build another 8-bit computer from scratch and make videos of the whole process! Buy a kit and build your own! https://eater.net/8bit/kits\n\nSupport me on Patreon: https://www.patreon.com/beneater',
+        }
+    }
+
+    def _get_entries(self, api_response):
+        return [InvidiousIE(self._downloader)._real_extract(self.host_url + '/watch?v=' + video['videoId'])
+                for video in api_response['videos']]
+
+    def _real_extract(self, url):
+        playlist_id = self._match_id(url)
+
+        # host_url will contain `http[s]://example.com` where `example.com` is the used invidious instance.
+        url_parsed = compat_urllib_parse.urlparse(url)
+        self.host_url = url_parsed.scheme + '://' + url_parsed.netloc
+
+        api_response = self._download_json(self.host_url + '/api/v1/playlists/' + playlist_id, playlist_id)
+        return InfoExtractor.playlist_result(self._get_entries(api_response), playlist_id, api_response.get('title'), api_response.get('description')) | {
+            'release_timestamp': api_response.get('updated'),
+
+            'uploader': api_response.get('author'),
+            'uploader_id': api_response.get('authorId'),
+        }

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',