Upgrade yt_dlp and download script

2025-05-02 16:11:08 -05:00
parent 3a2e8eeb08
commit d68d9ce4f9
1194 changed files with 60099 additions and 44436 deletions
--- a/plugins/youtube_download/yt_dlp/extractor/dailymotion.py
+++ b/plugins/youtube_download/yt_dlp/extractor/dailymotion.py
@@ -1,6 +1,7 @@
 import functools
 import json
 import re
+import urllib.parse

 from .common import InfoExtractor
 from ..networking.exceptions import HTTPError
@@ -9,11 +10,14 @@ from ..utils import (
    OnDemandPagedList,
    age_restricted,
    clean_html,
+    extract_attributes,
    int_or_none,
    traverse_obj,
    try_get,
    unescapeHTML,
    unsmuggle_url,
+    update_url,
+    url_or_none,
    urlencode_postdata,
 )

@@ -44,36 +48,41 @@ class DailymotionBaseInfoExtractor(InfoExtractor):
        self._FAMILY_FILTER = ff == 'on' if ff else age_restricted(18, self.get_param('age_limit'))
        self._set_dailymotion_cookie('ff', 'on' if self._FAMILY_FILTER else 'off')

+    def _get_token(self, xid):
+        cookies = self._get_dailymotion_cookies()
+        token = self._get_cookie_value(cookies, 'access_token') or self._get_cookie_value(cookies, 'client_token')
+        if token:
+            return token
+
+        data = {
+            'client_id': 'f1a362d288c1b98099c7',
+            'client_secret': 'eea605b96e01c796ff369935357eca920c5da4c5',
+        }
+        username, password = self._get_login_info()
+        if username:
+            data.update({
+                'grant_type': 'password',
+                'password': password,
+                'username': username,
+            })
+        else:
+            data['grant_type'] = 'client_credentials'
+        try:
+            token = self._download_json(
+                'https://graphql.api.dailymotion.com/oauth/token',
+                None, 'Downloading Access Token',
+                data=urlencode_postdata(data))['access_token']
+        except ExtractorError as e:
+            if isinstance(e.cause, HTTPError) and e.cause.status == 400:
+                raise ExtractorError(self._parse_json(
+                    e.cause.response.read().decode(), xid)['error_description'], expected=True)
+            raise
+        self._set_dailymotion_cookie('access_token' if username else 'client_token', token)
+        return token
+
    def _call_api(self, object_type, xid, object_fields, note, filter_extra=None):
        if not self._HEADERS.get('Authorization'):
-            cookies = self._get_dailymotion_cookies()
-            token = self._get_cookie_value(cookies, 'access_token') or self._get_cookie_value(cookies, 'client_token')
-            if not token:
-                data = {
-                    'client_id': 'f1a362d288c1b98099c7',
-                    'client_secret': 'eea605b96e01c796ff369935357eca920c5da4c5',
-                }
-                username, password = self._get_login_info()
-                if username:
-                    data.update({
-                        'grant_type': 'password',
-                        'password': password,
-                        'username': username,
-                    })
-                else:
-                    data['grant_type'] = 'client_credentials'
-                try:
-                    token = self._download_json(
-                        'https://graphql.api.dailymotion.com/oauth/token',
-                        None, 'Downloading Access Token',
-                        data=urlencode_postdata(data))['access_token']
-                except ExtractorError as e:
-                    if isinstance(e.cause, HTTPError) and e.cause.status == 400:
-                        raise ExtractorError(self._parse_json(
-                            e.cause.response.read().decode(), xid)['error_description'], expected=True)
-                    raise
-                self._set_dailymotion_cookie('access_token' if username else 'client_token', token)
-            self._HEADERS['Authorization'] = 'Bearer ' + token
+            self._HEADERS['Authorization'] = f'Bearer {self._get_token(xid)}'

        resp = self._download_json(
            'https://graphql.api.dailymotion.com/', xid, note, data=json.dumps({
@@ -81,7 +90,7 @@ class DailymotionBaseInfoExtractor(InfoExtractor):
  %s(xid: "%s"%s) {
    %s
  }
-}''' % (object_type, xid, ', ' + filter_extra if filter_extra else '', object_fields),
+}''' % (object_type, xid, ', ' + filter_extra if filter_extra else '', object_fields),  # noqa: UP031
            }).encode(), headers=self._HEADERS)
        obj = resp['data'][object_type]
        if not obj:
@@ -91,15 +100,23 @@ class DailymotionBaseInfoExtractor(InfoExtractor):

 class DailymotionIE(DailymotionBaseInfoExtractor):
    _VALID_URL = r'''(?ix)
-                    https?://
+                    (?:https?:)?//
+                    (?:
+                        dai\.ly/|
                        (?:
-                            (?:(?:www|touch|geo)\.)?dailymotion\.[a-z]{2,3}/(?:(?:(?:(?:embed|swf|\#)/)|player\.html\?)?video|swf)|
-                            (?:www\.)?lequipe\.fr/video
+                            (?:(?:www|touch|geo)\.)?dailymotion\.[a-z]{2,3}|
+                            (?:www\.)?lequipe\.fr
+                        )/
+                        (?:
+                            swf/(?!video)|
+                            (?:(?:crawler|embed|swf)/)?video/|
+                            player(?:/[\da-z]+)?\.html\?(?:video|(?P<is_playlist>playlist))=
                        )
-                        [/=](?P<id>[^/?_&]+)(?:.+?\bplaylist=(?P<playlist_id>x[0-9a-z]+))?
-                    '''
+                    )
+                    (?P<id>[^/?_&#]+)(?:[\w-]*\?playlist=(?P<playlist_id>x[0-9a-z]+))?
+    '''
    IE_NAME = 'dailymotion'
-    _EMBED_REGEX = [r'<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.com/(?:embed|swf)/video/.+?)\1']
+    _EMBED_REGEX = [rf'(?ix)<(?:(?:embed|iframe)[^>]+?src=|input[^>]+id=[\'"]dmcloudUrlEmissionSelect[\'"][^>]+value=)["\'](?P<url>{_VALID_URL[5:]})']
    _TESTS = [{
        'url': 'http://www.dailymotion.com/video/x5kesuj_office-christmas-party-review-jason-bateman-olivia-munn-t-j-miller_news',
        'md5': '074b95bdee76b9e3654137aee9c79dfe',
@@ -107,13 +124,17 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
            'id': 'x5kesuj',
            'ext': 'mp4',
            'title': 'Office Christmas Party Review –  Jason Bateman, Olivia Munn, T.J. Miller',
-            'description': 'Office Christmas Party Review -  Jason Bateman, Olivia Munn, T.J. Miller',
+            'description': 'Office Christmas Party Review - Jason Bateman, Olivia Munn, T.J. Miller',
            'duration': 187,
            'timestamp': 1493651285,
            'upload_date': '20170501',
            'uploader': 'Deadline',
            'uploader_id': 'x1xm8ri',
            'age_limit': 0,
+            'view_count': int,
+            'like_count': int,
+            'tags': ['hollywood', 'celeb', 'celebrity', 'movies', 'red carpet'],
+            'thumbnail': r're:https://(?:s[12]\.)dmcdn\.net/v/K456B1cmt4ZcZ9KiM/x1080',
        },
    }, {
        'url': 'https://geo.dailymotion.com/player.html?video=x89eyek&mute=true',
@@ -132,8 +153,8 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
            'view_count': int,
            'like_count': int,
            'tags': ['en_quete_d_esprit'],
-            'thumbnail': 'https://s2.dmcdn.net/v/Tncwi1YGKdvFbDuDY/x1080',
-        }
+            'thumbnail': r're:https://(?:s[12]\.)dmcdn\.net/v/Tncwi1clTH6StrxMP/x1080',
+        },
    }, {
        'url': 'https://www.dailymotion.com/video/x2iuewm_steam-machine-models-pricing-listed-on-steam-store-ign-news_videogames',
        'md5': '2137c41a8e78554bb09225b8eb322406',
@@ -201,6 +222,111 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
    }, {
        'url': 'https://www.dailymotion.com/video/x3z49k?playlist=xv4bw',
        'only_matching': True,
+    }, {
+        'url': 'https://geo.dailymotion.com/player/x86gw.html?video=k46oCapRs4iikoz9DWy',
+        'only_matching': True,
+    }, {
+        'url': 'https://geo.dailymotion.com/player/xakln.html?video=x8mjju4&customConfig%5BcustomParams%5D=%2Ffr-fr%2Ftennis%2Fwimbledon-mens-singles%2Farticles-video',
+        'only_matching': True,
+    }, {  # playlist-only
+        'url': 'https://geo.dailymotion.com/player/xf7zn.html?playlist=x7wdsj',
+        'only_matching': True,
+    }, {
+        'url': 'https://geo.dailymotion.com/player/xmyye.html?video=x93blhi',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.dailymotion.com/crawler/video/x8u4owg',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.dailymotion.com/embed/video/x8u4owg',
+        'only_matching': True,
+    }, {
+        'url': 'https://dai.ly/x94cnnk',
+        'only_matching': True,
+    }]
+    _WEBPAGE_TESTS = [{
+        # https://geo.dailymotion.com/player/xmyye.html?video=x93blhi
+        'url': 'https://www.financialounge.com/video/2024/08/01/borse-europee-in-rosso-dopo-la-fed-a-milano-volano-mediobanca-e-tim-edizione-del-1-agosto/',
+        'info_dict': {
+            'id': 'x93blhi',
+            'ext': 'mp4',
+            'title': 'OnAir - 01/08/24',
+            'description': '',
+            'duration': 217,
+            'timestamp': 1722505658,
+            'upload_date': '20240801',
+            'uploader': 'Financialounge',
+            'uploader_id': 'x2vtgmm',
+            'age_limit': 0,
+            'tags': [],
+            'view_count': int,
+            'like_count': int,
+            'thumbnail': r're:https://\w+.dmcdn.net/v/WnEY61cmvMxt2Fi6d/x1080',
+        },
+    }, {
+        # https://geo.dailymotion.com/player/xf7zn.html?playlist=x7wdsj
+        'url': 'https://www.cycleworld.com/blogs/ask-kevin/ducati-continues-to-evolve-with-v4/',
+        'info_dict': {
+            'id': 'x7wdsj',
+        },
+        'playlist_mincount': 50,
+    }, {
+        # https://www.dailymotion.com/crawler/video/x8u4owg
+        'url': 'https://www.leparisien.fr/environnement/video-le-veloto-la-voiture-a-pedales-qui-aimerait-se-faire-une-place-sur-les-routes-09-03-2024-KCYMCPM4WFHJXMSKBUI66UNFPU.php',
+        'info_dict': {
+            'id': 'x8u4owg',
+            'ext': 'mp4',
+            'like_count': int,
+            'uploader': 'Le Parisien',
+            'thumbnail': 'https://www.leparisien.fr/resizer/ho_GwveeYftNkLwg_cEta--5Bv4=/1200x675/cloudfront-eu-central-1.images.arcpublishing.com/leparisien/BFXJNEBN75EUNHGYJLORUC3TX4.jpg',
+            'upload_date': '20240309',
+            'view_count': int,
+            'timestamp': 1709997866,
+            'age_limit': 0,
+            'uploader_id': 'x32f7b',
+            'title': 'VIDÉO. Le «\xa0véloto\xa0», la voiture à pédales qui aimerait se faire une place sur les routes',
+            'duration': 428.0,
+            'description': 'À bord du « véloto », l’alternative à la voiture pour la campagne',
+            'tags': ['biclou', 'vélo', 'véloto', 'campagne', 'voiture', 'environnement', 'véhicules intermédiaires'],
+        },
+    }, {
+        # https://geo.dailymotion.com/player/xry80.html?video=x8vu47w
+        'url': 'https://www.metatube.com/en/videos/546765/This-frogs-decorates-Christmas-tree/',
+        'info_dict': {
+            'id': 'x8vu47w',
+            'ext': 'mp4',
+            'like_count': int,
+            'uploader': 'Metatube',
+            'thumbnail': r're:https://\w+.dmcdn.net/v/W1G_S1coGSFTfkTeR/x1080',
+            'upload_date': '20240326',
+            'view_count': int,
+            'timestamp': 1711496732,
+            'age_limit': 0,
+            'uploader_id': 'x2xpy74',
+            'title': 'Está lindas ranitas ponen su arbolito',
+            'duration': 28,
+            'description': 'Que lindura',
+            'tags': [],
+        },
+    }, {
+        # //geo.dailymotion.com/player/xysxq.html?video=k2Y4Mjp7krAF9iCuINM
+        'url': 'https://lcp.fr/programmes/avant-la-catastrophe-la-naissance-de-la-dictature-nazie-1933-1936-346819',
+        'info_dict': {
+            'id': 'k2Y4Mjp7krAF9iCuINM',
+            'ext': 'mp4',
+            'title': 'Avant la catastrophe la naissance de la dictature nazie 1933 -1936',
+            'description': 'md5:7b620d5e26edbe45f27bbddc1c0257c1',
+            'uploader': 'LCP Assemblée nationale',
+            'uploader_id': 'xbz33d',
+            'view_count': int,
+            'like_count': int,
+            'age_limit': 0,
+            'duration': 3220,
+            'thumbnail': 'https://s1.dmcdn.net/v/Xvumk1djJBUZfjj2a/x1080',
+            'tags': [],
+            'timestamp': 1739919947,
+            'upload_date': '20250218',
+        },
    }]
    _GEO_BYPASS = False
    _COMMON_MEDIA_FIELDS = '''description
@@ -215,17 +341,36 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
        yield from super()._extract_embed_urls(url, webpage)
        for mobj in re.finditer(
                r'(?s)DM\.player\([^,]+,\s*{.*?video[\'"]?\s*:\s*["\']?(?P<id>[0-9a-zA-Z]+).+?}\s*\);', webpage):
-            yield from 'https://www.dailymotion.com/embed/video/' + mobj.group('id')
+            yield 'https://www.dailymotion.com/embed/video/' + mobj.group('id')
+        for mobj in re.finditer(
+                r'(?s)<script [^>]*\bsrc=(["\'])(?:https?:)?//[\w-]+\.dailymotion\.com/player/(?:(?!\1).)+\1[^>]*>', webpage):
+            attrs = extract_attributes(mobj.group(0))
+            player_url = url_or_none(attrs.get('src'))
+            if not player_url:
+                continue
+            player_url = player_url.replace('.js', '.html')
+            if player_url.startswith('//'):
+                player_url = f'https:{player_url}'
+            if video_id := attrs.get('data-video'):
+                query_string = f'video={video_id}'
+            elif playlist_id := attrs.get('data-playlist'):
+                query_string = f'playlist={playlist_id}'
+            else:
+                continue
+            yield update_url(player_url, query=query_string)

    def _real_extract(self, url):
        url, smuggled_data = unsmuggle_url(url)
-        video_id, playlist_id = self._match_valid_url(url).groups()
+        video_id, is_playlist, playlist_id = self._match_valid_url(url).group('id', 'is_playlist', 'playlist_id')

-        if playlist_id:
-            if self._yes_playlist(playlist_id, video_id):
-                return self.url_result(
-                    'http://www.dailymotion.com/playlist/' + playlist_id,
-                    'DailymotionPlaylist', playlist_id)
+        if is_playlist:  # We matched the playlist query param as video_id
+            playlist_id = video_id
+            video_id = None
+
+        if self._yes_playlist(playlist_id, video_id):
+            return self.url_result(
+                f'http://www.dailymotion.com/playlist/{playlist_id}',
+                'DailymotionPlaylist', playlist_id)

        password = self.get_param('videopassword')
        media = self._call_api(
@@ -244,8 +389,8 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
      %s
      audienceCount
      isOnAir
-    }''' % (self._COMMON_MEDIA_FIELDS, self._COMMON_MEDIA_FIELDS), 'Downloading media JSON metadata',
-            'password: "%s"' % self.get_param('videopassword') if password else None)
+    }''' % (self._COMMON_MEDIA_FIELDS, self._COMMON_MEDIA_FIELDS), 'Downloading media JSON metadata',  # noqa: UP031
+            'password: "{}"'.format(self.get_param('videopassword')) if password else None)
        xid = media['xid']

        metadata = self._download_json(
@@ -261,11 +406,13 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                allowed_countries = try_get(media, lambda x: x['geoblockedCountries']['allowed'], list)
                self.raise_geo_restricted(msg=title, countries=allowed_countries)
            raise ExtractorError(
-                '%s said: %s' % (self.IE_NAME, title), expected=True)
+                f'{self.IE_NAME} said: {title}', expected=True)

        title = metadata['title']
        is_live = media.get('isOnAir')
        formats = []
+        subtitles = {}
+
        for quality, media_list in metadata['qualities'].items():
            for m in media_list:
                media_url = m.get('url')
@@ -273,8 +420,10 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
                if not media_url or media_type == 'application/vnd.lumberjack.manifest':
                    continue
                if media_type == 'application/x-mpegURL':
-                    formats.extend(self._extract_m3u8_formats(
-                        media_url, video_id, 'mp4', live=is_live, m3u8_id='hls', fatal=False))
+                    fmt, subs = self._extract_m3u8_formats_and_subtitles(
+                        media_url, video_id, 'mp4', live=is_live, m3u8_id='hls', fatal=False)
+                    formats.extend(fmt)
+                    self._merge_subtitles(subs, target=subtitles)
                else:
                    f = {
                        'url': media_url,
@@ -294,20 +443,18 @@ class DailymotionIE(DailymotionBaseInfoExtractor):
            if not f.get('fps') and f['format_id'].endswith('@60'):
                f['fps'] = 60

-        subtitles = {}
        subtitles_data = try_get(metadata, lambda x: x['subtitles']['data'], dict) or {}
        for subtitle_lang, subtitle in subtitles_data.items():
            subtitles[subtitle_lang] = [{
                'url': subtitle_url,
            } for subtitle_url in subtitle.get('urls', [])]

-        thumbnails = []
-        for height, poster_url in metadata.get('posters', {}).items():
-            thumbnails.append({
-                'height': int_or_none(height),
-                'id': height,
-                'url': poster_url,
-            })
+        thumbnails = traverse_obj(metadata, (
+            ('posters', 'thumbnails'), {dict.items}, lambda _, v: url_or_none(v[1]), {
+                'height': (0, {int_or_none}),
+                'id': (0, {str}),
+                'url': 1,
+            }))

        owner = metadata.get('owner') or {}
        stats = media.get('stats') or {}
@@ -347,7 +494,7 @@ class DailymotionPlaylistBaseIE(DailymotionBaseInfoExtractor):
        }
      }
    }''' % ('false' if self._FAMILY_FILTER else 'true', self._PAGE_SIZE, page),
-            'Downloading page %d' % page)['videos']
+            f'Downloading page {page}')['videos']
        for edge in videos['edges']:
            node = edge['node']
            yield self.url_result(
@@ -380,12 +527,58 @@ class DailymotionPlaylistIE(DailymotionPlaylistBaseIE):
                r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//(?:www\.)?dailymotion\.[a-z]{2,3}/widget/jukebox\?.+?)\1',
                webpage):
            for p in re.findall(r'list\[\]=/playlist/([^/]+)/', unescapeHTML(mobj.group('url'))):
-                yield '//dailymotion.com/playlist/%s' % p
+                yield f'//dailymotion.com/playlist/{p}'
+
+
+class DailymotionSearchIE(DailymotionPlaylistBaseIE):
+    IE_NAME = 'dailymotion:search'
+    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/search/(?P<id>[^/?#]+)/videos'
+    _PAGE_SIZE = 20
+    _TESTS = [{
+        'url': 'http://www.dailymotion.com/search/king of turtles/videos',
+        'info_dict': {
+            'id': 'king of turtles',
+            'title': 'king of turtles',
+        },
+        'playlist_mincount': 90,
+    }]
+    _SEARCH_QUERY = 'query SEARCH_QUERY( $query: String! $page: Int $limit: Int ) { search { videos( query: $query first: $limit page: $page ) { edges { node { xid } } } } } '
+
+    def _call_search_api(self, term, page, note):
+        if not self._HEADERS.get('Authorization'):
+            self._HEADERS['Authorization'] = f'Bearer {self._get_token(term)}'
+        resp = self._download_json(
+            'https://graphql.api.dailymotion.com/', None, note, data=json.dumps({
+                'operationName': 'SEARCH_QUERY',
+                'query': self._SEARCH_QUERY,
+                'variables': {
+                    'limit': 20,
+                    'page': page,
+                    'query': term,
+                },
+            }).encode(), headers=self._HEADERS)
+        obj = traverse_obj(resp, ('data', 'search', {dict}))
+        if not obj:
+            raise ExtractorError(
+                traverse_obj(resp, ('errors', 0, 'message', {str})) or 'Could not fetch search data')
+
+        return obj
+
+    def _fetch_page(self, term, page):
+        page += 1
+        response = self._call_search_api(term, page, f'Searching "{term}" page {page}')
+        for xid in traverse_obj(response, ('videos', 'edges', ..., 'node', 'xid')):
+            yield self.url_result(f'https://www.dailymotion.com/video/{xid}', DailymotionIE, xid)
+
+    def _real_extract(self, url):
+        term = urllib.parse.unquote_plus(self._match_id(url))
+        return self.playlist_result(
+            OnDemandPagedList(functools.partial(self._fetch_page, term), self._PAGE_SIZE), term, term)


 class DailymotionUserIE(DailymotionPlaylistBaseIE):
    IE_NAME = 'dailymotion:user'
-    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist)/)(?:(?:old/)?user/)?(?P<id>[^/]+)'
+    _VALID_URL = r'https?://(?:www\.)?dailymotion\.[a-z]{2,3}/(?!(?:embed|swf|#|video|playlist|search|crawler)/)(?:(?:old/)?user/)?(?P<id>[^/?#]+)'
    _TESTS = [{
        'url': 'https://www.dailymotion.com/user/nqtv',
        'info_dict': {