Merge f677ef143c5ef254af6e03aaa22fc772db1bd0ee 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

@@ -1421,6 +1421,10 @@ from .ufctv import (
     UFCTVIE,
     UFCArabiaIE,
 )
+from .unige import (
+    UnigeIE,
+    UnigePlaylistIE,
+)
 from .uktvplay import UKTVPlayIE
 from .digiteka import DigitekaIE
 from .dlive import (

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/unige.py

@@ -0,0 +1,99 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    urlencode_postdata,
+)
+
+from youtube_dl.compat import (
+    compat_HTTPError,
+)
+
+
+class UnigeIE(InfoExtractor):
+    _VALID_URL = r'https://mediaserver.unige.ch/play/(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'https://mediaserver.unige.ch/play/196613',
+        'md5': 'xxxx',
+        'info_dict': {
+            'id': '196613',
+            'display_id': '196613',
+            'ext': 'mp4',
+        },
+    }, {
+        'url': 'https://mediaserver.unige.ch/proxy/196613/VN3-2569-2023-2024-09-19.mp4',
+        'only_matching': True,
+    }]
+
+    def _login(self, video_id):
+        # Login credentials are per video group
+
+        username, password = self._get_login_info(netrc_machine=f'unige-mediaserver-{video_id}')
+        if not username or not password:
+            self.raise_login_required('You need a username/pwd to access this video')
+
+        try:
+            secure_wp = f'https://mediaserver.unige.ch/proxy/{video_id}/secure.php?view=play&id={video_id}'
+            self._download_webpage(
+                secure_wp, None, 'Logging in',
+                data=urlencode_postdata({
+                    'httpd_username': username,
+                    'httpd_password': password,
+                }), headers={
+                    'Referer': secure_wp,
+                    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
+                })
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400:
+                raise ExtractorError(
+                    'Unable to login: incorrect username and/or password',
+                    expected=True)
+            raise
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        try:
+            # This dumb download only checks if we need to login, as authentication
+            # is unique (and sometimes optional) for each video
+            secure_wp = f'https://mediaserver.unige.ch/proxy/{video_id}/secure.php?view=play&id={video_id}'
+            self._download_webpage(secure_wp, f'secure_{video_id}')
+        except ExtractorError as e:
+            if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                self._login(video_id)
+            else:
+                # The video doesn't require login
+                pass
+
+        video_title = self._html_search_regex(r'<h1>(.+?)</h1>', webpage.replace('\n', ''), 'unige')
+        course_title = self._html_search_regex(r'<a href="/collection/[-\w+]+">(?P<course>.*)</a></div>', webpage, 'unige')
+        course_id = self._html_search_regex(r'<a href="/collection/(?P<courseid>[-\w+]+)">', webpage, 'unige')
+
+        video_url = self._search_regex(
+            r'<source src="([^"]+)"', webpage, 'video URL')
+
+        return {
+            'id': video_id,
+            'title': video_title,
+            'url': video_url,
+            'channel': course_title,
+            'channel_id': course_id,
+        }
+
+
+class UnigePlaylistIE(InfoExtractor):
+    _VALID_URL = r'https://mediaserver.unige.ch/collection/(?P<id>[-\w+]+)'
+
+    def _real_extract(self, url):
+        collection_id = self._match_id(url)
+
+        rss = self._download_xml(url + '.rss', collection_id)
+
+        entries = [self.url_result(video.text, 'Unige')
+                   for video in rss.findall('./channel/item/link')]
+        title_text = rss.find('./channel/title').text
+
+        return self.playlist_result(entries, collection_id, title_text)

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: