Merge cbaa856993a59cd5038ce218385ced543e711dba 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/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/yourporn.py

@@ -24,44 +24,91 @@ class YourPornIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+    }, {
+        'url': 'https://sxyprn.com/post/5acfc82b12d00.html',
+        'md5': '1df93ede16d87685aa069f56ac69b0e7',
+        'info_dict': {
+            'id': '5acfc82b12d00',
+            'ext': 'mp4',
+            'title': 'Girls Do Porn E157 The Mormon Girl this should be in HD... #GirlsDoPorn #GDP #BigTits #casting',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'duration': 2466,
+            'age_limit': 18,
+        },
     }, {
         'url': 'https://sxyprn.com/post/57ffcb2e1179b.html',
         'only_matching': True,
+    }, {
+        'url': 'https://sxyprn.com/post/5af15f6799de9.html',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-
-        webpage = self._download_webpage(url, video_id)
-
-        parts = self._parse_json(
-            self._search_regex(
-                r'data-vnfo=(["\'])(?P<data>{.+?})\1', webpage, 'data info',
-                group='data'),
-            video_id)[video_id].split('/')
-
-        num = 0
-        for c in parts[6] + parts[7]:
-            if c.isnumeric():
-                num += int(c)
-        parts[5] = compat_str(int(parts[5]) - num)
-        parts[1] += '8'
-        video_url = urljoin(url, '/'.join(parts))
-
-        title = (self._search_regex(
-            r'<[^>]+\bclass=["\']PostEditTA[^>]+>([^<]+)', webpage, 'title',
-            default=None) or self._og_search_description(webpage)).strip()
-        thumbnail = self._og_search_thumbnail(webpage)
-        duration = parse_duration(self._search_regex(
+        get_duration = lambda webpage: parse_duration(self._search_regex(
             r'duration\s*:\s*<[^>]+>([\d:]+)', webpage, 'duration',
             default=None))
 
-        return {
-            'id': video_id,
-            'url': video_url,
-            'title': title,
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'age_limit': 18,
-            'ext': 'mp4',
+        # Only for posts containing a single video is the post_id equal to
+        # video_id. If there a multiple videos there also exists a post with the
+        # video_id and we use this page to fetch the video title.
+        post_id = self._match_id(url)
+
+        webpage = self._download_webpage(url, post_id)
+
+        videos = self._parse_json(
+            self._search_regex(
+                r'data-vnfo=(["\'])(?P<data>{.+?})\1', webpage, 'data info',
+                group='data'),
+            post_id)
+
+        for video_id in videos:
+            parts = videos[video_id].split('/')
+            num = 0
+            for c in parts[6] + parts[7]:
+                if c.isnumeric():
+                    num += int(c)
+            parts[5] = compat_str(int(parts[5]) - num)
+            parts[1] += '8'
+            videos[video_id] = urljoin(url, '/'.join(parts))
+
+        # If there ist only one video in the post (as is most likely) the
+        # video_id is equal to post_id and we save us from re-fetching the page
+        # to obtain the meta data.
+        # This may fail but if needed will be obtained in the next step.
+        titles = {
+            post_id: self._og_search_description(webpage, default=None)
         }
+        thumbnails = {
+            post_id: self._og_search_thumbnail(webpage, default=None)
+        }
+        durations = {
+            post_id: get_duration(webpage)
+        }
+
+        # obtain missing metadata for all videos in the post
+        for video_id in videos:
+            if not titles.get(video_id) or not thumbnails.get(video_id) or video_id not in durations:
+                webpage = self._download_webpage('https://sxyprn.com/post/%s.html' % video_id, video_id)
+            if not titles.get(video_id):
+                titles[video_id] = self._og_search_description(webpage)
+            if not thumbnails.get(video_id):
+                thumbnails[video_id] = self._og_search_thumbnail(webpage)
+            if video_id not in durations:
+                durations[video_id] = get_duration(webpage)
+
+        entries = []
+        for video_id in videos:
+            entries.append({
+                'id': video_id,
+                'url': videos[video_id],
+                'title': titles[video_id],
+                'thumbnail': thumbnails[video_id],
+                'duration': durations[video_id],
+                'age_limit': 18,
+                'ext': 'mp4',
+            })
+
+        if len(entries) == 1:
+            return entries[0]
+        else:
+            return self.playlist_result(entries, post_id, titles[post_id])

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: