finish upgrade, cleanup

yt_dlp/extractor/dailywire.py

@@ -1,9 +1,12 @@
+import itertools
 import json
 from .common import InfoExtractor
 from ..utils import (
     determine_ext,
+    ExtractorError,
     float_or_none,
     join_nonempty,
+    parse_iso8601,
     traverse_obj,
     url_or_none,
 )
@@ -12,48 +15,54 @@
 class DailyWireBaseIE(InfoExtractor):
     _GRAPHQL_API = 'https://v2server.dailywire.com/app/graphql'
     _GRAPHQL_QUERIES = {
-        'getEpisodeBySlug': 'query getEpisodeBySlug($slug: String!) {episode(where: {slug: $slug}) {id,title,status,slug,isLive,description,createdAt,scheduleAt,updatedAt,image,allowedCountryNames,allowedContinents,rating,show {id,name,slug,belongsTo},segments {id,image,title,liveChatAccess,audio,video,duration,watchTime,description,videoAccess,muxAssetId,muxPlaybackId,captions {id}},createdBy {firstName,lastName},discussionId}}',
+        'getClipBySlug': 'query getClipBySlug($slug:String!){clip(where:{slug:$slug}){id,name,slug,description,image,show{id,name,slug},thumbnail,duration,createdBy{firstName,lastName},createdAt,videoURL}}',
-        'getVideoBySlug': 'query getVideoBySlug($slug: String!) {video(where: {slug: $slug}) {id,name,slug,status,description,metadata,image,clips {id,name,slug,image,show {id,name,slug},video {id,name,slug},thumbnail,duration,clipAccess,createdAt,updatedAt},scheduleAt,allowedContinents,allowedCountryNames,rating,thumbnail,videoURL,logoImage,duration,createdBy {firstName,lastName},createdAt,updatedAt,watchTime,liveChatAccess,videoAccess,captions {id}}}',
+        'getEpisodeBySlug': 'query getEpisodeBySlug($slug:String!){episode(where:{slug:$slug}){id,title,slug,description,createdAt,image,show{id,name,slug},segments{audio,video,duration,},createdBy{firstName,lastName}}}',
-        'getClipBySlug': 'query getClipBySlug($slug: String!) {clip(where: {slug: $slug}) {id,name,slug,description,image,show {id,name,slug},video {id,status,name,slug},thumbnail,duration,clipAccess,createdBy {firstName,lastName},createdAt,updatedAt,videoURL,captions {id}}}',
+        'getPodcastEpisodes': 'query getPodcastEpisodes($where: PodcastEpisodeWhereInput, $orderBy: PodcastEpisodeOrderBy, $skip: Int, $first: Int) {listPodcastEpisode(where: $where, orderBy: $orderBy, skip: $skip, first: $first) {...ResPodcastEpisode}}, fragment ResPodcastEpisode on getPodcastEpisodeRes {id,title,description,slug,thumbnail,createdAt,audio,duration,podcast {id,name,slug,author {firstName,lastName}},season {id,name,slug}}',
-        'getShowBySlug': 'query getShowBySlug($slug: String!) {show(where: {slug: $slug}) {id,slug,belongsTo,name,description,image,logoImage,backgroundImage,hostImage,createdAt,updatedAt,author {firstName,lastName},episodes(where: {status_not_in: [DRAFT, UNPUBLISHED]},first: 1,orderBy: createdAt_DESC) {id,image,title,slug,status,createdAt,scheduleAt,description,show {id,name,slug,belongsTo},segments {id,title,muxAssetId,muxPlaybackId,audio}},seasons(orderBy: weight_DESC) {id,name,slug,orderBy}}}',
+        'getSeasonEpisodes': 'query getSeasonEpisodes($where:getSeasonEpisodesInput!,$first:Int,$skip:Int){getSeasonEpisodes(where:$where,first:$first,skip:$skip){episode{slug}}}',
-        'getSeasonEpisodes': 'query getSeasonEpisodes($where: getSeasonEpisodesInput!, $first: Int, $skip: Int) {getSeasonEpisodes(where: $where, first: $first, skip: $skip) {id,episode {id,title,slug,image,description,updatedAt,scheduleAt,createdAt,discussionId,isLive,status}}}',
+        'getShowBySlug': 'query getShowBySlug($slug:String!){show(where:{slug:$slug}){id,name,description,image,seasons(orderBy:weight_DESC){id,name,slug}}}',
+        'getVideoBySlug': 'query getVideoBySlug($slug:String!){video(where:{slug:$slug}){id,name,slug,description,image,thumbnail,videoURL,duration,createdBy{firstName,lastName},createdAt}}',
     }
     _GRAPHQL_VIDEO_QUERIES = {
+        'clips': 'getClipBySlug',
         'episode': 'getEpisodeBySlug',
         'videos': 'getVideoBySlug',
-        'clips': 'getClipBySlug',
     }
     _GRAPHQL_JSON_PATH = {
-        'getEpisodeBySlug': ('data', 'episode'),
-        'getVideoBySlug': ('data', 'video'),
         'getClipBySlug': ('data', 'clip'),
-        'getSeasonEpisodes': ('data', 'getSeasonEpisodes'),
+        'getEpisodeBySlug': ('data', 'episode'),
+        'getPodcastEpisodes': ('data', 'listPodcastEpisode'),
+        'getSeasonEpisodes': ('data', 'getSeasonEpisodes', ..., 'episode', 'slug'),
         'getShowBySlug': ('data', 'show'),
+        'getVideoBySlug': ('data', 'video'),
+    }
+    _API_HEADERS = {
+        'Apollographql-Client-Name': 'DW_WEBSITE',
+        'Content-Type': 'application/json',
+        'Origin': 'https://www.dailywire.com',
+        'Referer': 'https://www.dailywire.com/',
     }
-    _ACCESS_TOKEN = None
 
-    def _get_auth(self):
+    def _real_initialize(self):
-        t = self._get_cookies('https://www.dailywire.com').get('accessToken')
+        if access_token := self._get_cookies('https://www.dailywire.com').get('accessToken'):
-        if t:
+            self._API_HEADERS['Authorization'] = f'Bearer {access_token.value}'
-            self._ACCESS_TOKEN = t.value
 
-    def _call_api(self, slug, query, variables):
+    def _call_api(self, slug, query, variables, message='Downloading JSON from GraphQL API'):
-        headers = {
-                'Apollographql-Client-Name': 'DW_WEBSITE',
-                'Content-Type': 'application/json',
-                'Origin': 'https://www.dailywire.com',
-                'Referer': 'https://www.dailywire.com/'}
-        
-        self._get_auth() ### I doubt this is how you are supposed to set _TOKEN
-        if self._ACCESS_TOKEN:
-            headers['Authorization'] = f'Bearer {self._ACCESS_TOKEN}'
-        
         json_data = self._download_json(
-                self._GRAPHQL_API, slug, 'Downloading JSON from GraphQL API', data=json.dumps({
+            self._GRAPHQL_API, slug, message, data=json.dumps(
-                'query': self._GRAPHQL_QUERIES[query], 'variables': variables}).encode(), headers=headers)
+                {'query': self._GRAPHQL_QUERIES[query], 'variables': variables}).encode(),
+            headers=self._API_HEADERS)
 
         return traverse_obj(json_data, self._GRAPHQL_JSON_PATH.get(query, ()))
 
+    def _paginate(self, slug, query, where):
+        for i in itertools.count(0):
+            page = self._call_api(slug,
+                query, {'where': where, 'first': 10, 'skip': i * 10},
+                message=f'Downloading page {i + 1}')
+            if not page:
+                break
+            yield page
+
 
 class DailyWireIE(DailyWireBaseIE):
     _VALID_URL = r'https?://(?:www\.)dailywire(?:\.com)/(?P<sites_type>episode|videos|clips)/(?P<id>[\w-]+)'
@@ -66,39 +75,79 @@ class DailyWireIE(DailyWireBaseIE):
             'title': '1. Fauci',
             'description': 'md5:9df630347ef85081b7e97dd30bc22853',
             'thumbnail': 'https://daily-wire-production.imgix.net/episodes/ckzsl50xnqpy30850in3v4bu7/ckzsl50xnqpy30850in3v4bu7-1648237399554.jpg',
-            'creator': 'Caroline Roberts',
             'series_id': 'ckzplm0a097fn0826r2vc3j7h',
             'series': 'China: The Enemy Within',
-        }
+            'upload_date': '20220218',
-    }, {
+            'creators': ['Caroline Roberts'],
-        'url': 'https://www.dailywire.com/episode/ep-124-bill-maher',
+            'timestamp': 1645182003,
-        'info_dict': {
-            'id': 'cl0ngbaalplc80894sfdo9edf',
-            'ext': 'mp3',
-            'display_id': 'ep-124-bill-maher',
-            'title': 'Ep. 124 - Bill Maher',
-            'thumbnail': 'https://daily-wire-production.imgix.net/episodes/cl0ngbaalplc80894sfdo9edf/cl0ngbaalplc80894sfdo9edf-1647065568518.jpg',
-            'creator': 'Caroline Roberts',
-            'description': 'md5:adb0de584bcfa9c41374999d9e324e98',
-            'series_id': 'cjzvep7270hp00786l9hwccob',
-            'series': 'The Sunday Special',
         }
     }, {
         'url': 'https://www.dailywire.com/videos/the-hyperions',
         'only_matching': True,
+    }, {
+        'skip': 'premium only',
+        'url': 'https://www.dailywire.com/episode/ep-3-avery-s-niece-new',
+        'info_dict': {
+            'id': 'clm8geguv3qku0870ewvcu0ed',
+            'display_id': 'ep-3-avery-s-niece-new',
+            'title': 'Ep 3 - Avery’s Niece',
+            'description': 'md5:861ab336bd2bab2abebc25a1479a42e0',
+            'thumbnail': 'https://daily-wire-production.imgix.net/episodes/clm8geguv3qku0870ewvcu0ed/clm8geguv3qku0870ewvcu0ed-1694047935734.png',
+            'series_id': 'clim20ue5f8160838ecz7ba8q',
+            'ext': 'mp4',
+            'subtitles': {'en-US': [{'ext': 'vtt'}]},
+            'timestamp': 1694062826,
+            'series': 'Convicting a Murderer',
+            'creators': ['Scott Bowler '],
+            'upload_date': '20230907',
+        },
+    }, {
+        'skip': 'premium only',
+        'url': 'https://www.dailywire.com/clips/the-making-of-run-hide-fight',
+        'info_dict': {
+            'id': 'ckjutyd6810dd0806ivcq2526',
+            'display_id': 'the-making-of-run-hide-fight',
+            'title': 'The Making of Run Hide Fight',
+            'description': 'md5:085297d753b73ad87bdd8b050cc10d2c',
+            'thumbnail': 'https://image.media.dailywire.com/K7OqsPwWH5c9hpWT68CHeZ4vRUtoz5Le/thumbnail.png',
+            'duration': 916.790889,
+            'creators': ['Paul Snyder'],
+            'upload_date': '20210113',
+            'timestamp': 1610506443,
+            'ext': 'mp4',
+        },
+    }, {
+        'skip': 'premium only',
+        'url': 'https://www.dailywire.com/videos/choosing-death-the-legacy-of-roe',
+        'info_dict': {
+            'id': 'cl3260dva6pjr097819zw506s',
+            'display_id': 'choosing-death-the-legacy-of-roe',
+            'title': 'Choosing Death [The Legacy of Roe]',
+            'description': 'md5:b07597f0ef32130365427a05fd1ccd25',
+            'duration': 2618.0738,
+            'timestamp': 1652308821,
+            'upload_date': '20220511',
+            'thumbnail': 'https://image.media.dailywire.com/FBgIBgmq635VuqTgWKjcGviEjJ2vJ02Zz/thumbnail.png',
+            'subtitles': {'en-US': [{'ext': 'vtt'}]},
+            'ext': 'mp4',
+        },
     }]
 
     def _real_extract(self, url):
         sites_type, slug = self._match_valid_url(url).group('sites_type', 'id')
-        episode_info = self._call_api(slug, self._GRAPHQL_VIDEO_QUERIES[sites_type], {'slug': slug})
+        episode_data = self._call_api(slug, self._GRAPHQL_VIDEO_QUERIES[sites_type], {'slug': slug})
 
-        urls = traverse_obj(episode_info,
+        if not episode_data:
-            (('segments', 'clips'), ..., ('video', 'audio')), expected_type=url_or_none) or [
+            raise ExtractorError('video not found')
-            episode_info.get('videoURL') if episode_info.get('videoURL') != 'Access Denied' else None]
 
-        ### print(urls)
+        urls = traverse_obj(episode_data,
+                            (('segments', 'clips'), ..., ('video', 'audio'))
+                            ) or [episode_data.get('videoURL')]
 
-        ### add warning about auth if no formats and no auth ?
+        if 'Access Denied' in urls:
+            self.report_warning(f'It looks like {slug} requires a login. Try passing cookies and try again.')
+
+        urls = [url_or_none(u) for u in urls if url_or_none(u)]
 
         formats, subtitles = [], {}
         for url in urls:
@@ -109,24 +158,27 @@ def _real_extract(self, url):
             formats.extend(format_)
             self._merge_subtitles(subs_, target=subtitles)
         return {
-            'id': episode_info['id'],
+            'id': episode_data.get('id'),
             'display_id': slug,
-            'title': traverse_obj(episode_info, 'title', 'name'),
+            'title': traverse_obj(episode_data, 'title', 'name'),
-            'description': episode_info.get('description'),
+            'description': episode_data.get('description'),
-            'creator': join_nonempty(('createdBy', 'firstName'), ('createdBy', 'lastName'), from_dict=episode_info, delim=' '),
+            'creator': join_nonempty(('createdBy', 'firstName'), ('createdBy', 'lastName'),
-            'duration': float_or_none(episode_info.get('duration')),
+                                     from_dict=episode_data, delim=' '),
-            'is_live': episode_info.get('isLive'),
+            'duration': float_or_none(episode_data.get('duration')),
-            'thumbnail': traverse_obj(episode_info, 'thumbnail', 'image', expected_type=url_or_none),
+            'timestamp': parse_iso8601(episode_data.get('createdAt')),
+            'is_live': episode_data.get('isLive'),
+            'thumbnail': traverse_obj(episode_data, 'thumbnail', 'image', expected_type=url_or_none),
             'formats': formats,
             'subtitles': subtitles,
-            'series_id': traverse_obj(episode_info, ('show', 'id')),
+            'series_id': traverse_obj(episode_data, ('show', 'id')),
-            'series': traverse_obj(episode_info, ('show', 'name')),
+            'series': traverse_obj(episode_data, ('show', 'name')),
-        } ### test to make sure all values are properly extracted
+        }
 
 
-class DailyWirePodcastIE(DailyWireBaseIE): ### need to rewrite this
+class DailyWirePodcastIE(DailyWireBaseIE):
-    _VALID_URL = r'https?://(?:www\.)dailywire(?:\.com)/(?P<sites_type>podcasts)/(?P<podcaster>[\w-]+/(?P<id>[\w-]+))'
+    _VALID_URL = r'https?://(?:www\.)dailywire(?:\.com)/(?P<sites_type>podcasts)/(?P<podcaster>[\w-]+)/?(?P<id>[\w-]+)?'
     _TESTS = [{
+        'note': 'serves shorter ad-free stream with paid cookies',
         'url': 'https://www.dailywire.com/podcasts/morning-wire/get-ready-for-recession-6-15-22',
         'info_dict': {
             'id': 'cl4f01d0w8pbe0a98ydd0cfn1',
@@ -134,45 +186,120 @@ class DailyWirePodcastIE(DailyWireBaseIE): ### need to rewrite this
             'display_id': 'get-ready-for-recession-6-15-22',
             'title': 'Get Ready for Recession | 6.15.22',
             'description': 'md5:c4afbadda4e1c38a4496f6d62be55634',
-            'thumbnail': 'https://daily-wire-production.imgix.net/podcasts/ckx4otgd71jm508699tzb6hf4-1639506575562.jpg',
+            'thumbnail': 'https://daily-wire-production.imgix.net/podcasts/ckx4otgd71jm508699tzb6hf4-1667859984424.jpg',
             'duration': 900.117667,
+            'timestamp': 1655261631,
+            'season_id': 'morning-wire-morning-wire-podcast-season',
+            'series_id': 'morning-wire',
+            'creators': ['Georgia Howe'],
+            'season': '2022',
+            'series': 'Morning Wire',
+            'upload_date': '20220615',
         }
+    }, {
+        'url': 'https://www.dailywire.com/podcasts/enough',
+        'info_dict': {
+            'id': 'ckx4kvm8710i80869lvuu1b8z',
+            'title': 'Enough',
+            'display_id': 'enough',
+        },
+        'playlist_mincount': 7,
     }]
 
     def _real_extract(self, url):
-        raise Exception('currently broken') ###
+        podcaster, slug = self._match_valid_url(url).group('podcaster', 'id')
-        slug, episode_info = self._get_json(url)
-        audio_id = traverse_obj(episode_info, 'audioMuxPlaybackId', 'VUsAipTrBVSgzw73SpC2DAJD401TYYwEp')
 
-        return {
+        def _extract_pod_ep_info(episode_data):
-            'id': episode_info['id'],
+            print(episode_data)
-            'url': f'https://stream.media.dailywire.com/{audio_id}/audio.m4a',
+            return {
-            'display_id': slug,
+                'id': episode_data.get('id'),
-            'title': episode_info.get('title'),
+                'url': episode_data.get('audio'),
-            'duration': float_or_none(episode_info.get('duration')),
+                'display_id': episode_data.get('slug'),
-            'thumbnail': episode_info.get('thumbnail'),
+                'title': episode_data.get('title'),
-            'description': episode_info.get('description'),
+                'duration': float_or_none(episode_data.get('duration')),
-        }
+                'timestamp': parse_iso8601(episode_data.get('createdAt')),
+                'thumbnail': episode_data.get('thumbnail'),
+                'description': episode_data.get('description'),
+                'creator': join_nonempty(('podcast', 'author', 'firstName'),
+                                         ('podcast', 'author', 'lastName'),
+                                         from_dict=episode_data, delim=' '),
+                'season': traverse_obj(episode_data, ('season', 'name')),
+                'season_id': traverse_obj(episode_data, ('season', 'slug')),
+                'series': traverse_obj(episode_data, ('podcast', 'name')),
+                'series_id': traverse_obj(episode_data, ('podcast', 'slug')),
+            }
+
+        if slug:
+            episodes = self._call_api(slug, 'getPodcastEpisodes', {'where': {'slug': slug}})
+            if episode_data := traverse_obj(episodes, ..., get_all=False):
+                return _extract_pod_ep_info(episode_data)
+        else:
+            episodes = [
+                episode for page in
+                self._paginate(podcaster, 'getPodcastEpisodes', {'podcast': {'slug': podcaster}})
+                for episode in page
+            ]
+
+            if episodes:
+                podcast_data = traverse_obj(episodes, (..., 'podcast'), {}, get_all=False)
+                return self.playlist_result(
+                    [_extract_pod_ep_info(e) for e in episodes],
+                    podcast_data.get('id'), podcast_data.get('name'), podcast_data.get('description'),
+                    display_id=podcast_data.get('slug'), thumbnail=podcast_data.get('coverImage'))
+
+        raise ExtractorError('Podcast not found')
 
 
 class DailyWireShowIE(DailyWireBaseIE):
     _VALID_URL = r'https?://(?:www\.)dailywire(?:\.com)/(?P<sites_type>show)/(?P<id>[\w-]+)'
-    _TESTS = [
+    _TESTS = [{
-    ]
+        'skip': 'premium only',
-    
+        'url': 'https://www.dailywire.com/show/apollo-11-what-we-saw',
-    def _real_extract(self, url):
+        'playlist_mincount': 28,
-        sites_type, slug = self._match_valid_url(url).group('sites_type', 'id')
+        'info_dict': {
-        
+            'id': 'ckixsvamonvl40862ysxve50i',
-        show = self._call_api(slug, 'getShowBySlug', {'slug': slug})
+            'thumbnail': 'https://daily-wire-production.imgix.net/shows/ckixsvamonvl40862ysxve50i-1679082975554.jpg',
-        
+            'title': 'What We Saw',
-        for season in show.get('seasons', []):
+            'description': 'md5:98d2a7d5cc8175494a4ca611058ed440',
-            season['episodes'] = []
+        },
-            while episode_page := self._call_api(season['slug'], 'getSeasonEpisodes', {'where': {'season': {'id': season['id']}},
+        'params': {
-                                                 'first': 10, 'skip': len(season['episodes']) or None}):
+            'skip_download': True,
-                season['episodes'] += [episode['episode'] for episode in episode_page]
+        },
-        
+        'playlist': [{
-        return self.playlist_result(
+            'info_dict': {
-            [self.url_result(f'https://www.dailywire.com/episode/{e["slug"]}')
+                'id': 'cltf80tk79fxi0942c7h394b5',
-             for season in show['seasons'] for e in season['episodes']],
+                'season_id': 'what-we-saw-season-3-an-empire-of-terror-season',
-            show.get('id'), show.get('name'), show.get('description'))
+                'ext': 'mp4',
+                'display_id': 'season-3-an-empire-of-terror',
+                'display_id': 'season-3-an-empire-of-terror',
+                'series_id': 'ckixsvamonvl40862ysxve50i',
+                'title': 'Season 3: An Empire of Terror',
+                'description': 'What We Saw: An Empire of Terror premieres on March 6, 2024.',
+                'creators': ['Scott Bowler '],
+                'upload_date': '20240306',
+                'timestamp': 1709704832,
+                'thumbnail': 'https://daily-wire-production.imgix.net/episodes/cltf80tk79fxi0942c7h394b5/cltf80tk79fxi0942c7h394b5-1709694601671.png',
+                'series': 'What We Saw',
+            }}]
+    }]
 
+    def _real_extract(self, url):
+        slug = self._match_valid_url(url).group('id')
+
+        show_data = self._call_api(slug, 'getShowBySlug', {'slug': slug})
+        if not show_data:
+            raise ExtractorError('Show not found')
+
+        for season_data in show_data.get('seasons', []):
+            season_data['episodes'] = [
+                episode for page in
+                self._paginate(season_data.get('slug'), 'getSeasonEpisodes', {'season': {'id': season_data.get('id')}})
+                for episode in page
+            ]
+
+        return self.playlist_result(
+            [self.url_result(f'https://www.dailywire.com/episode/{episode_slug}',
+             season_id=season_data.get('slug'), season=season_data.get('title'), url_transparent=True)
+             for season_data in show_data.get('seasons', []) for episode_slug in season_data['episodes']],
+            show_data.get('id'), show_data.get('name'), show_data.get('description'),
+            thumbnail=show_data.get('image'))