From 7eaf7f9abafb287e7c0e0e019430f4f8ed860e05 Mon Sep 17 00:00:00 2001 From: pukkandan Date: Wed, 2 Mar 2022 21:16:25 +0530 Subject: [PATCH] [rokfin] Add stack and channel extractors (#1534) Authored by: P-reducible, pukkandan --- README.md | 4 ++ yt_dlp/extractor/extractors.py | 6 +- yt_dlp/extractor/rokfin.py | 110 +++++++++++++++++++++++++++++++-- 3 files changed, 113 insertions(+), 7 deletions(-) diff --git a/README.md b/README.md index 70b2e202f..1aab0ba94 100644 --- a/README.md +++ b/README.md @@ -1698,6 +1698,10 @@ #### tiktok * `app_version`: App version to call mobile APIs with - should be set along with `manifest_app_version`. (e.g. `20.2.1`) * `manifest_app_version`: Numeric app version to call mobile APIs with. (e.g. `221`) +#### rokfinchannel +* `tab`: Which tab to download. One of `new`, `top`, `videos`, `podcasts`, `streams`, `stacks`. (E.g. `rokfinchannel:tab=streams`) + + NOTE: These options may be changed/removed in the future without concern for backward compatibility diff --git a/yt_dlp/extractor/extractors.py b/yt_dlp/extractor/extractors.py index 9cb1702f1..1a94c8b75 100644 --- a/yt_dlp/extractor/extractors.py +++ b/yt_dlp/extractor/extractors.py @@ -1333,7 +1333,11 @@ from .rice import RICEIE from .rmcdecouverte import RMCDecouverteIE from .rockstargames import RockstarGamesIE -from .rokfin import RokfinIE +from .rokfin import ( + RokfinIE, + RokfinStackIE, + RokfinChannelIE, +) from .roosterteeth import RoosterTeethIE, RoosterTeethSeriesIE from .rottentomatoes import RottenTomatoesIE from .rozhlas import RozhlasIE diff --git a/yt_dlp/extractor/rokfin.py b/yt_dlp/extractor/rokfin.py index 8b609cfb9..79a5b2336 100644 --- a/yt_dlp/extractor/rokfin.py +++ b/yt_dlp/extractor/rokfin.py @@ -1,11 +1,11 @@ # coding: utf-8 - import itertools from datetime import datetime from .common import InfoExtractor from ..utils import ( determine_ext, + ExtractorError, float_or_none, format_field, int_or_none, @@ -16,6 +16,9 @@ ) +_API_BASE_URL = 'https://prod-api-v2.production.rokfin.com/api/v2/public/' + + class RokfinIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?rokfin\.com/(?P(?Ppost|stream)/\d+)' _TESTS = [{ @@ -82,8 +85,7 @@ class RokfinIE(InfoExtractor): def _real_extract(self, url): video_id, video_type = self._match_valid_url(url).group('id', 'type') - metadata = self._download_json(f'https://prod-api-v2.production.rokfin.com/api/v2/public/{video_id}', - video_id, fatal=False) or {} + metadata = self._download_json(f'{_API_BASE_URL}{video_id}', video_id) scheduled = unified_timestamp(metadata.get('scheduledAt')) live_status = ('was_live' if metadata.get('stoppedAt') @@ -137,7 +139,7 @@ def _get_comments(self, video_id): pages_total = None for page_n in itertools.count(): raw_comments = self._download_json( - f'https://prod-api-v2.production.rokfin.com/api/v2/public/comment?postId={video_id[5:]}&page={page_n}&size=50', + 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")}', fatal=False) or {} @@ -153,6 +155,102 @@ def _get_comments(self, video_id): 'timestamp': unified_timestamp(comment.get('postedAt')) } - pages_total = int_or_none(raw_comments.get('totalPages')) - if not raw_comments.get('content') or raw_comments.get('last') is not False or page_n > (pages_total or 0): + pages_total = int_or_none(raw_comments.get('totalPages')) or None + is_last = raw_comments.get('last') + if not raw_comments.get('content') or is_last or (page_n > pages_total if pages_total else is_last is not False): return + + +class RokfinPlaylistBaseIE(InfoExtractor): + _TYPES = { + 'video': 'post', + 'audio': 'post', + 'stream': 'stream', + 'dead_stream': 'stream', + 'stack': 'stack', + } + + def _get_video_data(self, metadata): + for content in metadata.get('content') or []: + media_type = self._TYPES.get(content.get('mediaType')) + video_id = content.get('id') if media_type == 'post' else content.get('mediaId') + if not media_type or not video_id: + continue + + yield self.url_result(f'https://rokfin.com/{media_type}/{video_id}', video_id=f'{media_type}/{video_id}', + video_title=str_or_none(traverse_obj(content, ('content', 'contentTitle')))) + + +class RokfinStackIE(RokfinPlaylistBaseIE): + IE_NAME = 'rokfin:stack' + _VALID_URL = r'https?://(?:www\.)?rokfin\.com/stack/(?P[^/]+)' + _TESTS = [{ + 'url': 'https://www.rokfin.com/stack/271/Tulsi-Gabbard-Portsmouth-Townhall-FULL--Feb-9-2020', + 'playlist_count': 8, + 'info_dict': { + 'id': '271', + }, + }] + + def _real_extract(self, url): + list_id = self._match_id(url) + return self.playlist_result(self._get_video_data( + self._download_json(f'{_API_BASE_URL}stack/{list_id}', list_id)), list_id) + + +class RokfinChannelIE(RokfinPlaylistBaseIE): + IE_NAME = 'rokfin:channel' + _VALID_URL = r'https?://(?:www\.)?rokfin\.com/(?!((feed/?)|(discover/?)|(channels/?))$)(?P[^/]+)/?$' + _TESTS = [{ + 'url': 'https://rokfin.com/TheConvoCouch', + 'playlist_mincount': 100, + 'info_dict': { + 'id': '12071-new', + 'title': 'TheConvoCouch - New', + 'description': 'md5:bb622b1bca100209b91cd685f7847f06', + }, + }] + + _TABS = { + 'new': 'posts', + 'top': 'top', + 'videos': 'video', + 'podcasts': 'audio', + 'streams': 'stream', + 'stacks': 'stack', + } + + def _real_initialize(self): + self._validate_extractor_args() + + def _validate_extractor_args(self): + requested_tabs = self._configuration_arg('tab', None) + if requested_tabs is not None and (len(requested_tabs) > 1 or requested_tabs[0] not in self._TABS): + raise ExtractorError(f'Invalid extractor-arg "tab". Must be one of {", ".join(self._TABS)}', expected=True) + + def _entries(self, channel_id, channel_name, tab): + pages_total = None + for page_n in itertools.count(0): + if tab in ('posts', 'top'): + data_url = f'{_API_BASE_URL}user/{channel_name}/{tab}?page={page_n}&size=50' + else: + 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")}') + + yield from self._get_video_data(metadata) + pages_total = int_or_none(metadata.get('totalPages')) or None + is_last = metadata.get('last') + if is_last or (page_n > pages_total if pages_total else is_last is not False): + return + + def _real_extract(self, url): + channel_name = self._match_id(url) + channel_info = self._download_json(f'{_API_BASE_URL}user/{channel_name}', channel_name) + channel_id = channel_info['id'] + tab = self._configuration_arg('tab', default=['new'])[0] + + 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')))