From 265e586d96bae2eb86a4f702ee2caef3b0cd78c3 Mon Sep 17 00:00:00 2001 From: "Lesmiscore (Naoya Ozaki)" Date: Mon, 4 Apr 2022 00:41:14 +0900 Subject: [PATCH] [openrec] Download archived livestreams (#3267) Authored by: Lesmiscore --- yt_dlp/extractor/openrec.py | 30 +++++++++++++++++++++++------- 1 file changed, 23 insertions(+), 7 deletions(-) diff --git a/yt_dlp/extractor/openrec.py b/yt_dlp/extractor/openrec.py index b476c0986..5eb1cdbad 100644 --- a/yt_dlp/extractor/openrec.py +++ b/yt_dlp/extractor/openrec.py @@ -7,6 +7,7 @@ get_first, int_or_none, traverse_obj, + try_get, unified_strdate, unified_timestamp, ) @@ -18,6 +19,13 @@ def _extract_pagestore(self, webpage, video_id): return self._parse_json( self._search_regex(r'(?m)window\.pageStore\s*=\s*(\{.+?\});$', webpage, 'window.pageStore'), video_id) + def _expand_media(self, video_id, media): + for name, m3u8_url in (media or {}).items(): + if not m3u8_url: + continue + yield from self._extract_m3u8_formats( + m3u8_url, video_id, ext='mp4', m3u8_id=name) + def _extract_movie(self, webpage, video_id, name, is_live): window_stores = self._extract_pagestore(webpage, video_id) movie_stores = [ @@ -29,13 +37,21 @@ def _extract_movie(self, webpage, video_id, name, is_live): if not any(movie_stores): raise ExtractorError(f'Failed to extract {name} info') - m3u8_playlists = get_first(movie_stores, 'media') or {} - formats = [] - for name, m3u8_url in m3u8_playlists.items(): - if not m3u8_url: - continue - formats.extend(self._extract_m3u8_formats( - m3u8_url, video_id, ext='mp4', live=is_live, m3u8_id=name)) + formats = list(self._expand_media(video_id, get_first(movie_stores, 'media'))) + if not formats and is_live: + # archived livestreams + cookies = self._get_cookies('https://www.openrec.tv/') + detail = self._download_json( + f'https://apiv5.openrec.tv/api/v5/movies/{video_id}/detail', video_id, + headers={ + 'Origin': 'https://www.openrec.tv', + 'Referer': 'https://www.openrec.tv/', + 'access-token': try_get(cookies, lambda x: x.get('access_token').value), + 'uuid': try_get(cookies, lambda x: x.get('uuid').value), + }) + new_media = traverse_obj(detail, ('data', 'items', ..., 'media'), get_all=False) + formats = list(self._expand_media(video_id, new_media)) + is_live = False self._sort_formats(formats)