Merge f2767e2ce0 into 5904853ae5

Closes #7442
Authored by: bashonly

yt_dlp/extractor/_extractors.py

@@ -350,6 +350,7 @@
 from .chilloutzone import ChilloutzoneIE
 from .chzzk import (
     CHZZKLiveIE,
+    CHZZKChannelIE,
     CHZZKVideoIE,
 )
 from .cinemax import CinemaxIE

yt_dlp/extractor/cbc.py

@@ -151,7 +151,7 @@ def _real_extract(self, url):
 
 class CBCPlayerIE(InfoExtractor):
     IE_NAME = 'cbc.ca:player'
-    _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>(?:\d\.)?\d+)'
+    _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/(?:video/)?|i/caffeine/syndicate/\?mediaId=))(?P<id>(?:\d\.)?\d+)'
     _TESTS = [{
         'url': 'http://www.cbc.ca/player/play/2683190193',
         'md5': '64d25f841ddf4ddb28a235338af32e2c',
@@ -277,6 +277,28 @@ class CBCPlayerIE(InfoExtractor):
             'location': 'Canada',
             'media_type': 'Full Program',
         },
+    }, {
+        'url': 'https://www.cbc.ca/player/play/video/1.7194274',
+        'md5': '188b96cf6bdcb2540e178a6caa957128',
+        'info_dict': {
+            'id': '2334524995812',
+            'ext': 'mp4',
+            'title': '#TheMoment a rare white spirit moose was spotted in Alberta',
+            'description': 'md5:18ae269a2d0265c5b0bbe4b2e1ac61a3',
+            'timestamp': 1714788791,
+            'duration': 77.678,
+            'subtitles': {'eng': [{'ext': 'vtt', 'protocol': 'm3u8_native'}]},
+            'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/201/543/THE_MOMENT.jpg',
+            'uploader': 'CBCC-NEW',
+            'chapters': 'count:0',
+            'upload_date': '20240504',
+            'categories': 'count:3',
+            'series': 'The National',
+            'tags': 'count:15',
+            'creators': ['encoder'],
+            'location': 'Canada',
+            'media_type': 'Excerpt',
+        },
     }, {
         'url': 'cbcplayer:1.7159484',
         'only_matching': True,

yt_dlp/extractor/chzzk.py

@@ -87,6 +87,91 @@ def _real_extract(self, url):
         }
 
 
+def _make_vod_result(data):
+    assert isinstance(data, dict)
+    if not data.get('videoId'):
+        return
+    return {
+        '_type': 'url_transparent',
+        'id': data.get('videoId'),
+        'title': data.get('videoTitle'),
+        'url': f'https://chzzk.naver.com/video/{data.get("videoNo")}',
+        'thumbnail': data.get('thumbnailImageUrl'),
+        'timestamp': data.get('publishDateAt'),
+        'view_count': data.get('readCount'),
+        'duration': data.get('duration'),
+        'age_limit': 19 if data.get('adult') else 0,
+        'was_live': True if data.get('videoType') == 'REPLAY' else False,
+    }
+
+
+class CHZZKChannelIE(InfoExtractor):
+    IE_NAME = 'chzzk:channel'
+    _VALID_URL = r'https?://chzzk\.naver\.com/(?P<id>[\da-f]{32})'
+    _TESTS = [{
+        'note': 'Both video and replay included',
+        'url': 'https://chzzk.naver.com/68f895c59a1043bc5019b5e08c83a5c5',
+        'info_dict': {
+            'id': '68f895c59a1043bc5019b5e08c83a5c5',
+            'channel_id': '68f895c59a1043bc5019b5e08c83a5c5',
+            'description': '나는 머찐 라디유 나는 머찐 라디유' * 26 + '나는 라디유',
+            'channel': '라디유radiyu',
+            'title': '라디유radiyu',
+            'channel_is_verified': False,
+        },
+        'playlist_mincount': 4,
+    }, {
+        'note': 'Video list paging',
+        'url': 'https://chzzk.naver.com/2d4aa2f79b0a397d032c479ef1b37a67',
+        'info_dict': {
+            'id': '2d4aa2f79b0a397d032c479ef1b37a67',
+            'channel_id': '2d4aa2f79b0a397d032c479ef1b37a67',
+            'description': '',
+            'channel': '진짜후추',
+            'title': '진짜후추',
+            'channel_is_verified': False,
+        },
+        'playlist_mincount': 86,
+    }]
+
+    def _real_extract(self, url):
+        channel_id = self._match_id(url)
+        # even if channel got banned, the video list is still available. but downloading videos leads to error
+        channel_meta = self._download_json(
+            f'https://api.chzzk.naver.com/service/v1/channels/{channel_id}', channel_id,
+            note='Downloading channel info', errnote='Unable to download channel info')
+        if channel_meta.get('code') != 200:
+            raise ExtractorError('The channel is not available: %s(%s)' % (
+                channel_meta.get("message"), channel_meta.get("code")), expected=True)
+
+        channel_meta = channel_meta.get('content')
+        if not channel_meta or not channel_meta.get('channelId'):
+            raise ExtractorError('The channel does not exist', expected=True)
+
+        total_count = 0
+        videos = []
+        current_count = 0
+        while current_count <= total_count:
+            videos_page = self._download_json(
+                f'https://api.chzzk.naver.com/service/v1/channels/{channel_id}/videos', channel_meta['channelName'],
+                query={'sortType': 'LATEST', 'pagingType': 'PAGE', 'size': 18, 'videoType': '', 'page': current_count},
+                note='Downloading videos page(%s/%s)' % (current_count + 1, total_count + 1), errnote='Unable to download videos page')
+            if videos_page.get('code') != 200:
+                raise ExtractorError('Unable to download videos page: %s(%s){videos_page.get("message")}' % (
+                    videos_page.get("message"), videos_page.get("code")), expected=True)
+            videos_page = videos_page['content']
+            videos.extend(map(_make_vod_result, videos_page['data']))
+            total_count = videos_page['totalPages']
+            current_count += 1
+        if len(videos) != total_count:
+            self.to_screen('Warning: The total count of videos is not equal to the actual count: %s != %s' % (len(videos), total_count))
+
+        return self.playlist_result(
+            videos, channel_id, channel_meta['channelName'], channel_meta['channelDescription'],
+            channel=channel_meta['channelName'], channel_id=channel_id, channel_is_verified=channel_meta['verifiedMark'],
+        )
+
+
 class CHZZKVideoIE(InfoExtractor):
     IE_NAME = 'chzzk:video'
     _VALID_URL = r'https?://chzzk\.naver\.com/video/(?P<id>\d+)'

yt_dlp/extractor/crunchyroll.py

@@ -53,15 +53,19 @@ def _set_auth_info(self, response):
         CrunchyrollBaseIE._AUTH_EXPIRY = time_seconds(seconds=traverse_obj(response, ('expires_in', {float_or_none}), default=300) - 10)
 
     def _request_token(self, headers, data, note='Requesting token', errnote='Failed to request token'):
-        try:  # TODO: Add impersonation support here
+        try:
             return self._download_json(
                 f'{self._BASE_URL}/auth/v1/token', None, note=note, errnote=errnote,
-                headers=headers, data=urlencode_postdata(data))
+                headers=headers, data=urlencode_postdata(data), impersonate=True)
         except ExtractorError as error:
             if not isinstance(error.cause, HTTPError) or error.cause.status != 403:
                 raise
+            if target := error.cause.response.extensions.get('impersonate'):
+                raise ExtractorError(f'Got HTTP Error 403 when using impersonate target "{target}"')
             raise ExtractorError(
-                'Request blocked by Cloudflare; navigate to Crunchyroll in your browser, '
+                'Request blocked by Cloudflare. '
+                'Install the required impersonation dependency if possible, '
+                'or else navigate to Crunchyroll in your browser, '
                 'then pass the fresh cookies (with --cookies-from-browser or --cookies) '
                 'and your browser\'s User-Agent (with --user-agent)', expected=True)
 

yt_dlp/extractor/youporn.py

@@ -72,15 +72,15 @@ class YouPornIE(InfoExtractor):
             'id': '16290308',
             'age_limit': 18,
             'categories': [],
-            'description': 'md5:00ea70f642f431c379763c17c2f396bc',
+            'description': str,  # TODO: detect/remove SEO spam description in ytdl backport
             'display_id': 'tinderspecial-trailer1',
             'duration': 298.0,
             'ext': 'mp4',
             'upload_date': '20201123',
             'uploader': 'Ersties',
             'tags': [],
-            'thumbnail': 'https://fi1.ypncdn.com/202011/23/16290308/original/8/tinderspecial-trailer1-8(m=eaAaaEPbaaaa).jpg',
-            'timestamp': 1606089600,
+            'thumbnail': r're:https://.+\.jpg',
+            'timestamp': 1606147564,
             'title': 'Tinder In Real Life',
             'view_count': int,
         }
@@ -88,11 +88,17 @@ class YouPornIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id, display_id = self._match_valid_url(url).group('id', 'display_id')
-        definitions = self._download_json(
-            f'https://www.youporn.com/api/video/media_definitions/{video_id}/', display_id or video_id)
+        self._set_cookie('.youporn.com', 'age_verified', '1')
+        webpage = self._download_webpage(f'https://www.youporn.com/watch/{video_id}', video_id)
+        definitions = self._search_json(r'\bplayervars\s*:', webpage, 'player vars', video_id)['mediaDefinitions']
 
-        def get_format_data(data, f):
-            return traverse_obj(data, lambda _, v: v['format'] == f and url_or_none(v['videoUrl']))
+        def get_format_data(data, stream_type):
+            info_url = traverse_obj(data, (lambda _, v: v['format'] == stream_type, 'videoUrl', {url_or_none}, any))
+            if not info_url:
+                return []
+            return traverse_obj(
+                self._download_json(info_url, video_id, f'Downloading {stream_type} info JSON', fatal=False),
+                lambda _, v: v['format'] == stream_type and url_or_none(v['videoUrl']))
 
         formats = []
         # Try to extract only the actual master m3u8 first, avoiding the duplicate single resolution "master" m3u8s
@@ -123,10 +129,6 @@ def get_format_data(data, f):
             f['height'] = height
             formats.append(f)
 
-        webpage = self._download_webpage(
-            'http://www.youporn.com/watch/%s' % video_id, display_id,
-            headers={'Cookie': 'age_verified=1'})
-
         title = self._html_search_regex(
             r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>',
             webpage, 'title', default=None) or self._og_search_title(