[utils] locked_file: Fix for PyPy on Windows

Closes #4060

.github/workflows/core.yml

@@ -10,12 +10,15 @@ jobs:
       matrix:
         os: [ubuntu-latest]
         # CPython 3.9 is in quick-test
-        python-version: ['3.6', '3.7', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8, pypy-3.9]
+        python-version: ['3.6', '3.7', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8]
         run-tests-ext: [sh]
         include:
-        # atleast one of the tests must be in windows
+        # atleast one of each CPython/PyPy tests must be in windows
         - os: windows-latest
-          python-version: 3.8
+          python-version: '3.8'
+          run-tests-ext: bat
+        - os: windows-latest
+          python-version: pypy-3.9
           run-tests-ext: bat
     steps:
     - uses: actions/checkout@v2

.github/workflows/download.yml

@@ -9,11 +9,15 @@ jobs:
       fail-fast: true
       matrix:
         os: [ubuntu-latest]
-        python-version: ['3.6', '3.7', '3.9', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8, pypy-3.9]
+        python-version: ['3.6', '3.7', '3.9', '3.10', 3.11-dev, pypy-3.6, pypy-3.7, pypy-3.8]
         run-tests-ext: [sh]
         include:
+        # atleast one of each CPython/PyPy tests must be in windows
         - os: windows-latest
-          python-version: 3.8
+          python-version: '3.8'
+          run-tests-ext: bat
+        - os: windows-latest
+          python-version: pypy-3.9
           run-tests-ext: bat
     steps:
     - uses: actions/checkout@v2

README.md

@@ -325,6 +325,7 @@ You can also fork the project on github and run your fork's [build workflow](.gi
     -h, --help                      Print this help text and exit
     --version                       Print program version and exit
     -U, --update                    Update this program to latest version
+    --no-update                     Do not update (default)
     -i, --ignore-errors             Ignore download and postprocessing errors.
                                     The download will be considered successful
                                     even if the postprocessing fails

yt_dlp/YoutubeDL.py

@@ -3060,7 +3060,7 @@ class YoutubeDL:
 
                 success = True
                 merger, fd = FFmpegMergerPP(self), None
-                if info_dict.get('url'):
+                if info_dict.get('protocol') or info_dict.get('url'):
                     fd = get_suitable_downloader(info_dict, self.params, to_stdout=temp_filename == '-')
                     if fd is not FFmpegFD and (
                             info_dict.get('section_start') or info_dict.get('section_end')):

yt_dlp/extractor/common.py

@@ -1419,6 +1419,10 @@ class InfoExtractor:
             'ViewAction': 'view',
         }
 
+        def is_type(e, *expected_types):
+            type = variadic(traverse_obj(e, '@type'))
+            return any(x in type for x in expected_types)
+
         def extract_interaction_type(e):
             interaction_type = e.get('interactionType')
             if isinstance(interaction_type, dict):
@@ -1432,9 +1436,7 @@ class InfoExtractor:
             if not isinstance(interaction_statistic, list):
                 return
             for is_e in interaction_statistic:
-                if not isinstance(is_e, dict):
-                    continue
-                if is_e.get('@type') != 'InteractionCounter':
+                if not is_type(is_e, 'InteractionCounter'):
                     continue
                 interaction_type = extract_interaction_type(is_e)
                 if not interaction_type:
@@ -1471,7 +1473,7 @@ class InfoExtractor:
                 info['chapters'] = chapters
 
         def extract_video_object(e):
-            assert e['@type'] == 'VideoObject'
+            assert is_type(e, 'VideoObject')
             author = e.get('author')
             info.update({
                 'url': traverse_obj(e, 'contentUrl', 'embedUrl', expected_type=url_or_none),
@@ -1503,13 +1505,12 @@ class InfoExtractor:
                 if at_top_level and set(e.keys()) == {'@context', '@graph'}:
                     traverse_json_ld(variadic(e['@graph'], allowed_types=(dict,)), at_top_level=False)
                     break
-                item_type = e.get('@type')
-                if expected_type is not None and expected_type != item_type:
+                if expected_type is not None and not is_type(e, expected_type):
                     continue
                 rating = traverse_obj(e, ('aggregateRating', 'ratingValue'), expected_type=float_or_none)
                 if rating is not None:
                     info['average_rating'] = rating
-                if item_type in ('TVEpisode', 'Episode'):
+                if is_type(e, 'TVEpisode', 'Episode'):
                     episode_name = unescapeHTML(e.get('name'))
                     info.update({
                         'episode': episode_name,
@@ -1519,39 +1520,39 @@ class InfoExtractor:
                     if not info.get('title') and episode_name:
                         info['title'] = episode_name
                     part_of_season = e.get('partOfSeason')
-                    if isinstance(part_of_season, dict) and part_of_season.get('@type') in ('TVSeason', 'Season', 'CreativeWorkSeason'):
+                    if is_type(part_of_season, 'TVSeason', 'Season', 'CreativeWorkSeason'):
                         info.update({
                             'season': unescapeHTML(part_of_season.get('name')),
                             'season_number': int_or_none(part_of_season.get('seasonNumber')),
                         })
                     part_of_series = e.get('partOfSeries') or e.get('partOfTVSeries')
-                    if isinstance(part_of_series, dict) and part_of_series.get('@type') in ('TVSeries', 'Series', 'CreativeWorkSeries'):
+                    if is_type(part_of_series, 'TVSeries', 'Series', 'CreativeWorkSeries'):
                         info['series'] = unescapeHTML(part_of_series.get('name'))
-                elif item_type == 'Movie':
+                elif is_type(e, 'Movie'):
                     info.update({
                         'title': unescapeHTML(e.get('name')),
                         'description': unescapeHTML(e.get('description')),
                         'duration': parse_duration(e.get('duration')),
                         'timestamp': unified_timestamp(e.get('dateCreated')),
                     })
-                elif item_type in ('Article', 'NewsArticle'):
+                elif is_type(e, 'Article', 'NewsArticle'):
                     info.update({
                         'timestamp': parse_iso8601(e.get('datePublished')),
                         'title': unescapeHTML(e.get('headline')),
                         'description': unescapeHTML(e.get('articleBody') or e.get('description')),
                     })
-                    if traverse_obj(e, ('video', 0, '@type')) == 'VideoObject':
+                    if is_type(traverse_obj(e, ('video', 0)), 'VideoObject'):
                         extract_video_object(e['video'][0])
-                    elif traverse_obj(e, ('subjectOf', 0, '@type')) == 'VideoObject':
+                    elif is_type(traverse_obj(e, ('subjectOf', 0)), 'VideoObject'):
                         extract_video_object(e['subjectOf'][0])
-                elif item_type == 'VideoObject':
+                elif is_type(e, 'VideoObject'):
                     extract_video_object(e)
                     if expected_type is None:
                         continue
                     else:
                         break
                 video = e.get('video')
-                if isinstance(video, dict) and video.get('@type') == 'VideoObject':
+                if is_type(video, 'VideoObject'):
                     extract_video_object(video)
                 if expected_type is None:
                     continue

yt_dlp/extractor/generic.py

@@ -2563,6 +2563,56 @@ class GenericIE(InfoExtractor):
                 }
             }]
         },
+        {
+            'note': 'Rumble embed',
+            'url': 'https://rumble.com/vdmum1-moose-the-dog-helps-girls-dig-a-snow-fort.html',
+            'md5': '53af34098a7f92c4e51cf0bd1c33f009',
+            'info_dict': {
+                'id': 'vb0ofn',
+                'ext': 'mp4',
+                'timestamp': 1612662578,
+                'uploader': 'LovingMontana',
+                'channel': 'LovingMontana',
+                'upload_date': '20210207',
+                'title': 'Winter-loving dog helps girls dig a snow fort ',
+                'channel_url': 'https://rumble.com/c/c-546523',
+                'thumbnail': 'https://sp.rmbl.ws/s8/1/5/f/x/x/5fxxb.OvCc.1-small-Moose-The-Dog-Helps-Girls-D.jpg',
+                'duration': 103,
+            }
+        },
+        {
+            'note': 'Rumble JS embed',
+            'url': 'https://therightscoop.com/what-does-9-plus-1-plus-1-equal-listen-to-this-audio-of-attempted-kavanaugh-assassins-call-and-youll-get-it',
+            'md5': '4701209ac99095592e73dbba21889690',
+            'info_dict': {
+                'id': 'v15eqxl',
+                'ext': 'mp4',
+                'channel': 'Mr Producer Media',
+                'duration': 92,
+                'title': '911 Audio From The Man Who Wanted To Kill Supreme Court Justice Kavanaugh',
+                'channel_url': 'https://rumble.com/c/RichSementa',
+                'thumbnail': 'https://sp.rmbl.ws/s8/1/P/j/f/A/PjfAe.OvCc-small-911-Audio-From-The-Man-Who-.jpg',
+                'timestamp': 1654892716,
+                'uploader': 'Mr Producer Media',
+                'upload_date': '20220610',
+            }
+        },
+        {
+            'note': 'JSON LD with multiple @type',
+            'url': 'https://www.nu.nl/280161/video/hoe-een-bladvlo-dit-verwoestende-japanse-onkruid-moet-vernietigen.html',
+            'md5': 'c7949f34f57273013fb7ccb1156393db',
+            'info_dict': {
+                'id': 'ipy2AcGL',
+                'ext': 'mp4',
+                'description': 'md5:6a9d644bab0dc2dc06849c2505d8383d',
+                'thumbnail': r're:https://media\.nu\.nl/m/.+\.jpg',
+                'title': 'Hoe een bladvlo dit verwoestende Japanse onkruid moet vernietigen',
+                'timestamp': 1586577474,
+                'upload_date': '20200411',
+                'age_limit': 0,
+                'duration': 111.0,
+            }
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -3891,15 +3941,10 @@ class GenericIE(InfoExtractor):
         json_ld = self._search_json_ld(webpage, video_id, default={})
         if json_ld.get('url') not in (url, None):
             self.report_detected('JSON LD')
-            if determine_ext(json_ld['url']) == 'm3u8':
-                json_ld['formats'], json_ld['subtitles'] = self._extract_m3u8_formats_and_subtitles(
-                    json_ld['url'], video_id, 'mp4')
-                json_ld.pop('url')
-                self._sort_formats(json_ld['formats'])
-            else:
-                json_ld['_type'] = 'url_transparent'
-                json_ld['url'] = smuggle_url(json_ld['url'], {'force_videoid': video_id, 'to_generic': True})
-            return merge_dicts(json_ld, info_dict)
+            return merge_dicts({
+                '_type': 'url_transparent',
+                'url': smuggle_url(json_ld['url'], {'force_videoid': video_id, 'to_generic': True}),
+            }, json_ld, info_dict)
 
         def check_video(vurl):
             if YoutubeIE.suitable(vurl):

yt_dlp/extractor/jwplatform.py

@@ -5,7 +5,7 @@ from ..utils import unsmuggle_url
 
 
 class JWPlatformIE(InfoExtractor):
-    _VALID_URL = r'(?:https?://(?:content\.jwplatform|cdn\.jwplayer)\.com/(?:(?:feed|player|thumb|preview)s|jw6|v2/media)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
+    _VALID_URL = r'(?:https?://(?:content\.jwplatform|cdn\.jwplayer)\.com/(?:(?:feed|player|thumb|preview|manifest)s|jw6|v2/media)/|jwplatform:)(?P<id>[a-zA-Z0-9]{8})'
     _TESTS = [{
         'url': 'http://content.jwplatform.com/players/nPripu9l-ALJ3XQCI.js',
         'md5': 'fa8899fa601eb7c83a64e9d568bdf325',

yt_dlp/extractor/rumble.py

@@ -50,13 +50,14 @@ class RumbleEmbedIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    @staticmethod
-    def _extract_urls(webpage):
-        return [
-            mobj.group('url')
-            for mobj in re.finditer(
-                r'(?:<(?:script|iframe)[^>]+\bsrc=|["\']embedUrl["\']\s*:\s*)["\'](?P<url>%s)' % RumbleEmbedIE._VALID_URL,
-                webpage)]
+    @classmethod
+    def _extract_urls(cls, webpage):
+        embeds = tuple(re.finditer(
+            fr'(?:<(?:script|iframe)[^>]+\bsrc=|["\']embedUrl["\']\s*:\s*)["\'](?P<url>{cls._VALID_URL})', webpage))
+        if embeds:
+            return [mobj.group('url') for mobj in embeds]
+        return [f'https://rumble.com/embed/{mobj.group("id")}' for mobj in re.finditer(
+            r'<script>\s*Rumble\(\s*"play"\s*,\s*{\s*[\'"]video[\'"]\s*:\s*[\'"](?P<id>[0-9a-z]+)[\'"]', webpage)]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)

yt_dlp/extractor/zdf.py

@@ -69,7 +69,7 @@ class ZDFBaseIE(InfoExtractor):
             f.update({
                 'url': format_url,
                 'format_id': join_nonempty('http', meta.get('type'), meta.get('quality')),
-                'tbr': int_or_none(self._search_regex(r'_(\d+)k_', format_url, default=None))
+                'tbr': int_or_none(self._search_regex(r'_(\d+)k_', format_url, 'tbr', default=None))
             })
             new_formats = [f]
         formats.extend(merge_dicts(f, {

yt_dlp/options.py

@@ -312,6 +312,10 @@ def create_parser():
         '-U', '--update',
         action='store_true', dest='update_self',
         help='Update this program to latest version')
+    general.add_option(
+        '--no-update',
+        action='store_false', dest='update_self',
+        help='Do not update (default)')
     general.add_option(
         '-i', '--ignore-errors',
         action='store_true', dest='ignoreerrors',

yt_dlp/utils.py

@@ -2003,7 +2003,8 @@ if sys.platform == 'win32':
         if not LockFileEx(msvcrt.get_osfhandle(f.fileno()),
                           (0x2 if exclusive else 0x0) | (0x0 if block else 0x1),
                           0, whole_low, whole_high, f._lock_file_overlapped_p):
-            raise BlockingIOError('Locking file failed: %r' % ctypes.FormatError())
+            # NB: No argument form of "ctypes.FormatError" does not work on PyPy
+            raise BlockingIOError(f'Locking file failed: {ctypes.FormatError(ctypes.GetLastError())!r}')
 
     def _unlock_file(f):
         assert f._lock_file_overlapped_p