From a25bca9f89f77e6e5153c3400c4a27020d8cba9d Mon Sep 17 00:00:00 2001 From: pukkandan Date: Sun, 10 Apr 2022 01:00:21 +0530 Subject: [PATCH] [youtube, cleanup] Minor refactoring Authored by: coletdjnz, pukkandan --- yt_dlp/extractor/youtube.py | 71 ++++++++++++++++++++++++------------- 1 file changed, 47 insertions(+), 24 deletions(-) diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py index 031aa35a1..4ee09ad9a 100644 --- a/yt_dlp/extractor/youtube.py +++ b/yt_dlp/extractor/youtube.py @@ -384,6 +384,9 @@ def _initialize_pref(self): def _real_initialize(self): self._initialize_pref() self._initialize_consent() + self._check_login_required() + + def _check_login_required(self): if (self._LOGIN_REQUIRED and self.get_param('cookiefile') is None and self.get_param('cookiesfrombrowser') is None): @@ -563,6 +566,18 @@ def generate_api_headers( headers['X-Origin'] = origin return {h: v for h, v in headers.items() if v is not None} + def _download_ytcfg(self, client, video_id): + url = { + 'web': 'https://www.youtube.com', + 'web_music': 'https://music.youtube.com', + 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1' + }.get(client) + if not url: + return {} + webpage = self._download_webpage( + url, video_id, fatal=False, note=f'Downloading {client.replace("_", " ").strip()} client config') + return self.extract_ytcfg(video_id, webpage) or {} + @staticmethod def _build_api_continuation_query(continuation, ctp=None): query = { @@ -728,6 +743,7 @@ def extract_relative_time(relative_time_text): return None def _extract_time_text(self, renderer, *path_list): + """@returns (timestamp, time_text)""" text = self._get_text(renderer, *path_list) or '' dt = self.extract_relative_time(text) timestamp = None @@ -2959,16 +2975,6 @@ def _get_requested_clients(self, url, smuggled_data): return orderedSet(requested_clients) - def _extract_player_ytcfg(self, client, video_id): - url = { - 'web_music': 'https://music.youtube.com', - 'web_embedded': f'https://www.youtube.com/embed/{video_id}?html5=1' - }.get(client) - if not url: - return {} - webpage = self._download_webpage(url, video_id, fatal=False, note='Downloading %s config' % client.replace('_', ' ').strip()) - return self.extract_ytcfg(video_id, webpage) or {} - def _extract_player_responses(self, clients, video_id, webpage, master_ytcfg): initial_pr = None if webpage: @@ -3005,8 +3011,8 @@ def append_client(*client_names): while clients: client, base_client, variant = _split_innertube_client(clients.pop()) player_ytcfg = master_ytcfg if client == 'web' else {} - if 'configs' not in self._configuration_arg('player_skip'): - player_ytcfg = self._extract_player_ytcfg(client, video_id) or player_ytcfg + if 'configs' not in self._configuration_arg('player_skip') and client != 'web': + player_ytcfg = self._download_ytcfg(client, video_id) or player_ytcfg player_url = player_url or self._extract_player_url(master_ytcfg, player_ytcfg, webpage=webpage) require_js_player = self._get_default_ytcfg(client).get('REQUIRE_JS_PLAYER') @@ -4347,6 +4353,10 @@ def _reload_with_unavailable_videos(self, item_id, data, ytcfg): check_get_keys='contents', fatal=False, ytcfg=ytcfg, note='Downloading API JSON with unavailable videos') + @property + def skip_webpage(self): + return 'webpage' in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) + def _extract_webpage(self, url, item_id, fatal=True): retries = self.get_param('extractor_retries', 3) count = -1 @@ -4393,9 +4403,21 @@ def _extract_webpage(self, url, item_id, fatal=True): return webpage, data + def _report_playlist_authcheck(self, ytcfg, fatal=True): + """Use if failed to extract ytcfg (and data) from initial webpage""" + if not ytcfg and self.is_authenticated: + msg = 'Playlists that require authentication may not extract correctly without a successful webpage download' + if 'authcheck' not in self._configuration_arg('skip', ie_key=YoutubeTabIE.ie_key()) and fatal: + raise ExtractorError( + f'{msg}. If you are not downloading private content, or ' + 'your cookies are only for the first account and channel,' + ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check', + expected=True) + self.report_warning(msg, only_once=True) + def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=False, default_client='web'): data = None - if 'webpage' not in self._configuration_arg('skip'): + if not self.skip_webpage: webpage, data = self._extract_webpage(url, item_id, fatal=webpage_fatal) ytcfg = ytcfg or self.extract_ytcfg(item_id, webpage) # Reject webpage data if redirected to home page without explicitly requesting @@ -4409,14 +4431,7 @@ def _extract_data(self, url, item_id, ytcfg=None, fatal=True, webpage_fatal=Fals raise ExtractorError(msg, expected=True) self.report_warning(msg, only_once=True) if not data: - if not ytcfg and self.is_authenticated: - msg = 'Playlists that require authentication may not extract correctly without a successful webpage download.' - if 'authcheck' not in self._configuration_arg('skip') and fatal: - raise ExtractorError( - msg + ' If you are not downloading private content, or your cookies are only for the first account and channel,' - ' pass "--extractor-args youtubetab:skip=authcheck" to skip this check', - expected=True) - self.report_warning(msg, only_once=True) + self._report_playlist_authcheck(ytcfg, fatal=fatal) data = self._extract_tab_endpoint(url, item_id, ytcfg, fatal=fatal, default_client=default_client) return data, ytcfg @@ -4454,14 +4469,20 @@ def _search_results(self, query, params=NO_DEFAULT, default_client='web'): ('contents', 'tabbedSearchResultsRenderer', 'tabs', 0, 'tabRenderer', 'content', 'sectionListRenderer', 'contents'), ('continuationContents', ), ) + display_id = f'query "{query}"' check_get_keys = tuple(set(keys[0] for keys in content_keys)) + ytcfg = self._download_ytcfg(default_client, display_id) if not self.skip_webpage else {} + self._report_playlist_authcheck(ytcfg, fatal=False) continuation_list = [None] + search = None for page_num in itertools.count(1): data.update(continuation_list[0] or {}) + headers = self.generate_api_headers( + ytcfg=ytcfg, visitor_data=self._extract_visitor_data(search), default_client=default_client) search = self._extract_response( - item_id='query "%s" page %s' % (query, page_num), ep='search', query=data, - default_client=default_client, check_get_keys=check_get_keys) + item_id=f'{display_id} page {page_num}', ep='search', query=data, + default_client=default_client, check_get_keys=check_get_keys, ytcfg=ytcfg, headers=headers) slr_contents = traverse_obj(search, *content_keys) yield from self._extract_entries({'contents': list(variadic(slr_contents))}, continuation_list) if not continuation_list[0]: @@ -5634,7 +5655,9 @@ class YoutubeFeedsInfoExtractor(InfoExtractor): Subclasses must define the _FEED_NAME property. """ _LOGIN_REQUIRED = True - _TESTS = [] + + def _real_initialize(self): + YoutubeBaseInfoExtractor._check_login_required(self) @property def IE_NAME(self):