X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/e037c405ad57fb3276da133a05226451b9be6478..61edf57f8f13f6dfd81154174e647eb5fdd26089:/yt_dlp/extractor/rokfin.py diff --git a/yt_dlp/extractor/rokfin.py b/yt_dlp/extractor/rokfin.py index ea72e9dfc..be174655e 100644 --- a/yt_dlp/extractor/rokfin.py +++ b/yt_dlp/extractor/rokfin.py @@ -1,9 +1,10 @@ +import datetime as dt import itertools +import json import re import urllib.parse -from datetime import datetime -from .common import InfoExtractor +from .common import InfoExtractor, SearchInfoExtractor from ..utils import ( ExtractorError, determine_ext, @@ -37,14 +38,14 @@ class RokfinIE(InfoExtractor): 'upload_date': '20211023', 'timestamp': 1634998029, 'channel': 'Jimmy Dore', - 'channel_id': 65429, + 'channel_id': '65429', 'channel_url': 'https://rokfin.com/TheJimmyDoreShow', - 'duration': 213.0, 'availability': 'public', 'live_status': 'not_live', 'dislike_count': int, 'like_count': int, - } + 'duration': 213, + }, }, { 'url': 'https://rokfin.com/post/223/Julian-Assange-Arrested-Streaming-In-Real-Time', 'info_dict': { @@ -55,14 +56,14 @@ class RokfinIE(InfoExtractor): 'upload_date': '20190412', 'timestamp': 1555052644, 'channel': 'Ron Placone', - 'channel_id': 10, + 'channel_id': '10', 'channel_url': 'https://rokfin.com/RonPlacone', 'availability': 'public', 'live_status': 'not_live', 'dislike_count': int, 'like_count': int, 'tags': ['FreeThinkingMedia^', 'RealProgressives^'], - } + }, }, { 'url': 'https://www.rokfin.com/stream/10543/Its-A-Crazy-Mess-Regional-Director-Blows-Whistle-On-Pfizers-Vaccine-Trial-Data', 'info_dict': { @@ -71,8 +72,8 @@ class RokfinIE(InfoExtractor): 'title': '"It\'s A Crazy Mess" Regional Director Blows Whistle On Pfizer\'s Vaccine Trial Data', 'thumbnail': r're:https://img\.production\.rokfin\.com/.+', 'description': 'md5:324ce2d3e3b62e659506409e458b9d8e', - 'channel': 'Ryan Cristián', - 'channel_id': 53856, + 'channel': 'TLAVagabond', + 'channel_id': '53856', 'channel_url': 'https://rokfin.com/TLAVagabond', 'availability': 'public', 'is_live': False, @@ -85,7 +86,47 @@ class RokfinIE(InfoExtractor): 'dislike_count': int, 'like_count': int, 'tags': ['FreeThinkingMedia^'], - } + }, + }, { + 'url': 'https://rokfin.com/post/126703/Brave-New-World--Aldous-Huxley-DEEPDIVE--Chpts-13--Quite-Frankly--Jay-Dyer', + 'info_dict': { + 'id': 'post/126703', + 'ext': 'mp4', + 'title': 'Brave New World - Aldous Huxley DEEPDIVE! (Chpts 1-3) - Quite Frankly & Jay Dyer', + 'thumbnail': r're:https://img\.production\.rokfin\.com/.+', + 'channel': 'Jay Dyer', + 'channel_id': '186881', + 'channel_url': 'https://rokfin.com/jaydyer', + 'availability': 'premium_only', + 'live_status': 'not_live', + 'dislike_count': int, + 'like_count': int, + 'timestamp': 1678213357, + 'upload_date': '20230307', + 'tags': ['FreeThinkingMedia^', 'OpenMind^'], + 'description': 'md5:cb04e32e68326c9b2b251b297bacff35', + 'duration': 3100, + }, + }, { + 'url': 'https://rokfin.com/stream/31332/The-Grayzone-live-on-Nordstream-blame-game', + 'info_dict': { + 'id': 'stream/31332', + 'ext': 'mp4', + 'title': 'The Grayzone live on Nordstream blame game', + 'thumbnail': r're:https://image\.v\.rokfin\.com/.+', + 'channel': 'Max Blumenthal', + 'channel_id': '248902', + 'channel_url': 'https://rokfin.com/MaxBlumenthal', + 'availability': 'premium_only', + 'live_status': 'was_live', + 'dislike_count': int, + 'like_count': int, + 'timestamp': 1678475166, + 'release_timestamp': 1678475166.0, + 'release_date': '20230310', + 'upload_date': '20230310', + 'tags': ['FreeThinkingMedia^'], + }, }] def _real_extract(self, url): @@ -99,6 +140,12 @@ def _real_extract(self, url): else 'not_live') video_url = traverse_obj(metadata, 'url', ('content', 'contentUrl'), expected_type=url_or_none) + if video_url in (None, 'fake.m3u8'): + video_url = format_field(self._search_regex( + r'https?://[^/]+/([^/]+)/storyboard.vtt', + traverse_obj(metadata, 'timelineUrl', ('content', 'timelineUrl'), expected_type=url_or_none), + video_id, default=None), None, 'https://stream.v.rokfin.com/%s.m3u8') + formats, subtitles = [{'url': video_url}] if video_url else [], {} if determine_ext(video_url) == 'm3u8': formats, subtitles = self._extract_m3u8_formats_and_subtitles( @@ -109,9 +156,8 @@ def _real_extract(self, url): self.raise_login_required('This video is only available to premium users', True, method='cookies') elif scheduled: self.raise_no_formats( - f'Stream is offline; sheduled for {datetime.fromtimestamp(scheduled).strftime("%Y-%m-%d %H:%M:%S")}', + f'Stream is offline; scheduled for {dt.datetime.fromtimestamp(scheduled).strftime("%Y-%m-%d %H:%M:%S")}', video_id=video_id, expected=True) - self._sort_formats(formats) uploader = traverse_obj(metadata, ('createdBy', 'username'), ('creator', 'username')) timestamp = (scheduled or float_or_none(metadata.get('postedAtMilli'), 1000) @@ -127,7 +173,7 @@ def _real_extract(self, url): 'like_count': int_or_none(metadata.get('likeCount')), 'dislike_count': int_or_none(metadata.get('dislikeCount')), 'channel': str_or_none(traverse_obj(metadata, ('createdBy', 'name'), ('creator', 'name'))), - 'channel_id': traverse_obj(metadata, ('createdBy', 'id'), ('creator', 'id')), + 'channel_id': str_or_none(traverse_obj(metadata, ('createdBy', 'id'), ('creator', 'id'))), 'channel_url': url_or_none(f'https://rokfin.com/{uploader}') if uploader else None, 'timestamp': timestamp, 'release_timestamp': timestamp if live_status != 'not_live' else None, @@ -145,7 +191,7 @@ def _get_comments(self, video_id): for page_n in itertools.count(): raw_comments = self._download_json( f'{_API_BASE_URL}comment?postId={video_id[5:]}&page={page_n}&size=50', - video_id, note=f'Downloading viewer comments page {page_n + 1}{format_field(pages_total, template=" of %s")}', + video_id, note=f'Downloading viewer comments page {page_n + 1}{format_field(pages_total, None, " of %s")}', fatal=False) or {} for comment in raw_comments.get('content') or []: @@ -157,7 +203,7 @@ def _get_comments(self, video_id): 'parent': 'root', 'like_count': int_or_none(comment.get('numLikes')), 'dislike_count': int_or_none(comment.get('numDislikes')), - 'timestamp': unified_timestamp(comment.get('postedAt')) + 'timestamp': unified_timestamp(comment.get('postedAt')), } pages_total = int_or_none(raw_comments.get('totalPages')) or None @@ -197,10 +243,10 @@ def _perform_login(self, username, password): f'{self._AUTH_BASE}/token', None, note='getting access credentials', errnote='error getting access credentials', data=urlencode_postdata({ - 'code': urllib.parse.parse_qs(urllib.parse.urldefrag(urlh.geturl()).fragment).get('code')[0], + 'code': urllib.parse.parse_qs(urllib.parse.urldefrag(urlh.url).fragment).get('code')[0], 'client_id': 'web', 'grant_type': 'authorization_code', - 'redirect_uri': 'https://rokfin.com/silent-check-sso.html' + 'redirect_uri': 'https://rokfin.com/silent-check-sso.html', })) def _authentication_active(self): @@ -221,7 +267,7 @@ def _download_json_using_access_token(self, url_or_request, video_id, headers={} json_string, urlh = self._download_webpage_handle( url_or_request, video_id, headers=headers, query=query, expected_status=401) - if not auth_token or urlh.code != 401 or refresh_token is None: + if not auth_token or urlh.status != 401 or refresh_token is None: return self._parse_json(json_string, video_id) self._access_mgmt_tokens = self._download_json( @@ -230,7 +276,7 @@ def _download_json_using_access_token(self, url_or_request, video_id, headers={} data=urlencode_postdata({ 'grant_type': 'refresh_token', 'refresh_token': refresh_token, - 'client_id': 'web' + 'client_id': 'web', })) headers['authorization'] = self._get_auth_token() if headers['authorization'] is None: @@ -317,7 +363,7 @@ def _entries(self, channel_id, channel_name, tab): data_url = f'{_API_BASE_URL}post/search/{tab}?page={page_n}&size=50&creator={channel_id}' metadata = self._download_json( data_url, channel_name, - note=f'Downloading video metadata page {page_n + 1}{format_field(pages_total, template=" of %s")}') + note=f'Downloading video metadata page {page_n + 1}{format_field(pages_total, None, " of %s")}') yield from self._get_video_data(metadata) pages_total = int_or_none(metadata.get('totalPages')) or None @@ -334,3 +380,76 @@ def _real_extract(self, url): return self.playlist_result( self._entries(channel_id, channel_name, self._TABS[tab]), f'{channel_id}-{tab}', f'{channel_name} - {tab.title()}', str_or_none(channel_info.get('description'))) + + +class RokfinSearchIE(SearchInfoExtractor): + IE_NAME = 'rokfin:search' + IE_DESC = 'Rokfin Search' + _SEARCH_KEY = 'rkfnsearch' + _TYPES = { + 'video': (('id', 'raw'), 'post'), + 'audio': (('id', 'raw'), 'post'), + 'stream': (('content_id', 'raw'), 'stream'), + 'dead_stream': (('content_id', 'raw'), 'stream'), + 'stack': (('content_id', 'raw'), 'stack'), + } + _TESTS = [{ + 'url': 'rkfnsearch5:"zelenko"', + 'playlist_count': 5, + 'info_dict': { + 'id': '"zelenko"', + 'title': '"zelenko"', + }, + }] + _db_url = None + _db_access_key = None + + def _real_initialize(self): + self._db_url, self._db_access_key = self.cache.load(self.ie_key(), 'auth', default=(None, None)) + if not self._db_url: + self._get_db_access_credentials() + + def _search_results(self, query): + total_pages = None + for page_number in itertools.count(1): + search_results = self._run_search_query( + query, data={'query': query, 'page': {'size': 100, 'current': page_number}}, + note=f'Downloading page {page_number}{format_field(total_pages, None, " of ~%s")}') + total_pages = traverse_obj(search_results, ('meta', 'page', 'total_pages'), expected_type=int_or_none) + + for result in search_results.get('results') or []: + video_id_key, video_type = self._TYPES.get(traverse_obj(result, ('content_type', 'raw')), (None, None)) + video_id = traverse_obj(result, video_id_key, expected_type=int_or_none) + if video_id and video_type: + yield self.url_result(url=f'https://rokfin.com/{video_type}/{video_id}') + if not search_results.get('results'): + return + + def _run_search_query(self, video_id, data, **kwargs): + data = json.dumps(data).encode() + for attempt in range(2): + search_results = self._download_json( + self._db_url, video_id, data=data, fatal=(attempt == 1), + headers={'authorization': self._db_access_key}, **kwargs) + if search_results: + return search_results + self.write_debug('Updating access credentials') + self._get_db_access_credentials(video_id) + + def _get_db_access_credentials(self, video_id=None): + auth_data = {'SEARCH_KEY': None, 'ENDPOINT_BASE': None} + notfound_err_page = self._download_webpage( + 'https://rokfin.com/discover', video_id, expected_status=404, note='Downloading home page') + for js_file_path in re.findall(r']*\ssrc\s*=\s*"(/static/js/[^">]+)"', notfound_err_page): + js_content = self._download_webpage( + f'https://rokfin.com{js_file_path}', video_id, note='Downloading JavaScript file', fatal=False) + auth_data.update(re.findall( + rf'REACT_APP_({"|".join(auth_data.keys())})\s*:\s*"([^"]+)"', js_content or '')) + if not all(auth_data.values()): + continue + + self._db_url = url_or_none(f'{auth_data["ENDPOINT_BASE"]}/api/as/v1/engines/rokfin-search/search.json') + self._db_access_key = f'Bearer {auth_data["SEARCH_KEY"]}' + self.cache.store(self.ie_key(), 'auth', (self._db_url, self._db_access_key)) + return + raise ExtractorError('Unable to extract access credentials')