[misc] Add `hatch`, `ruff`, `pre-commit` and improve dev docs (#7409)

[yt-dlp.git] / yt_dlp / extractor / openrec.py

diff --git a/yt_dlp/extractor/openrec.py b/yt_dlp/extractor/openrec.py
index b476c0986e20528d2125c9f20850a06fa11d3545..c9a96aeb4dececa89b7cd3207012231a24d58bbc 100644 (file)
--- a/yt_dlp/extractor/openrec.py
+++ b/yt_dlp/extractor/openrec.py
@@ -1,23 +1,30 @@
-# coding: utf-8
-from __future__ import unicode_literals
-
 from .common import InfoExtractor
+from ..compat import compat_str
 from ..utils import (
     ExtractorError,
     get_first,
     int_or_none,
     traverse_obj,
+    try_get,
     unified_strdate,
     unified_timestamp,
 )
-from ..compat import compat_str
 
 
 class OpenRecBaseIE(InfoExtractor):
+    _M3U8_HEADERS = {'Referer': 'https://www.openrec.tv/'}
+
     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, headers=self._M3U8_HEADERS)
+
     def _extract_movie(self, webpage, video_id, name, is_live):
         window_stores = self._extract_pagestore(webpage, video_id)
         movie_stores = [
@@ -29,15 +36,21 @@ 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))
-
-        self._sort_formats(formats)
+        formats = list(self._expand_media(video_id, get_first(movie_stores, 'media')))
+        if not formats:
+            # archived livestreams or subscriber-only videos
+            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
 
         return {
             'id': video_id,
@@ -49,6 +62,7 @@ def _extract_movie(self, webpage, video_id, name, is_live):
             'uploader_id': get_first(movie_stores, ('channel', 'user', 'id')),
             'timestamp': int_or_none(get_first(movie_stores, ['publishedAt', 'time']), scale=1000) or unified_timestamp(get_first(movie_stores, 'publishedAt')),
             'is_live': is_live,
+            'http_headers': self._M3U8_HEADERS,
         }
 
 
@@ -99,8 +113,7 @@ def _real_extract(self, url):
             raise ExtractorError('Cannot extract title')
 
         formats = self._extract_m3u8_formats(
-            capture_data.get('source'), video_id, ext='mp4')
-        self._sort_formats(formats)
+            capture_data.get('source'), video_id, ext='mp4', headers=self._M3U8_HEADERS)
 
         return {
             'id': video_id,
@@ -111,6 +124,7 @@ def _real_extract(self, url):
             'uploader': traverse_obj(movie_store, ('channel', 'name'), expected_type=compat_str),
             'uploader_id': traverse_obj(movie_store, ('channel', 'id'), expected_type=compat_str),
             'upload_date': unified_strdate(capture_data.get('createdAt')),
+            'http_headers': self._M3U8_HEADERS,
         }