From 239f5f36fe04603bec59c8b975f6a792f10246db Mon Sep 17 00:00:00 2001 From: N/Ame <173015200+grqz@users.noreply.github.com> Date: Mon, 2 Dec 2024 10:55:18 +1300 Subject: [PATCH] [ie/bilibili] Fix extractor (#11667) Closes #11665 Authored by: grqz --- yt_dlp/extractor/bilibili.py | 43 ++++++++++++++++++------------------ 1 file changed, 21 insertions(+), 22 deletions(-) diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py index 02ea67707..f01befcc0 100644 --- a/yt_dlp/extractor/bilibili.py +++ b/yt_dlp/extractor/bilibili.py @@ -18,7 +18,6 @@ from ..utils import ( InAdvancePagedList, OnDemandPagedList, bool_or_none, - clean_html, determine_ext, filter_dict, float_or_none, @@ -639,31 +638,27 @@ class BiliBiliIE(BilibiliBaseIE): headers['Referer'] = url initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', video_id) + + if traverse_obj(initial_state, ('error', 'trueCode')) == -403: + self.raise_login_required() + if traverse_obj(initial_state, ('error', 'trueCode')) == -404: + raise ExtractorError( + 'This video may be deleted or geo-restricted. ' + 'You might want to try a VPN or a proxy server (with --proxy)', expected=True) + is_festival = 'videoData' not in initial_state if is_festival: video_data = initial_state['videoInfo'] else: - play_info_obj = self._search_json( - r'window\.__playinfo__\s*=', webpage, 'play info', video_id, fatal=False) - if not play_info_obj: - if traverse_obj(initial_state, ('error', 'trueCode')) == -403: - self.raise_login_required() - if traverse_obj(initial_state, ('error', 'trueCode')) == -404: - raise ExtractorError( - 'This video may be deleted or geo-restricted. ' - 'You might want to try a VPN or a proxy server (with --proxy)', expected=True) - play_info = traverse_obj(play_info_obj, ('data', {dict})) - if not play_info: - if traverse_obj(play_info_obj, 'code') == 87007: - toast = get_element_by_class('tips-toast', webpage) or '' - msg = clean_html( - f'{get_element_by_class("belongs-to", toast) or ""},' - + (get_element_by_class('level', toast) or '')) - raise ExtractorError( - f'This is a supporter-only video: {msg}. {self._login_hint()}', expected=True) - raise ExtractorError('Failed to extract play info') video_data = initial_state['videoData'] + if video_data.get('is_upower_exclusive'): + high_level = traverse_obj(initial_state, ('elecFullInfo', 'show_info', 'high_level', {dict})) or {} + raise ExtractorError( + 'This is a supporter-only video: ' + f'{join_nonempty("title", "sub_title", from_dict=high_level, delim=",")}. ' + f'{self._login_hint()}', expected=True) + video_id, title = video_data['bvid'], video_data.get('title') # Bilibili anthologies are similar to playlists but all videos share the same video ID as the anthology itself. @@ -689,10 +684,14 @@ class BiliBiliIE(BilibiliBaseIE): old_video_id = format_field(aid, None, f'%s_part{part_id or 1}') cid = traverse_obj(video_data, ('pages', part_id - 1, 'cid')) if part_id else video_data.get('cid') + play_info = ( + traverse_obj( + self._search_json(r'window\.__playinfo__\s*=', webpage, 'play info', video_id, default=None), + ('data', {dict})) + or self._download_playinfo(video_id, cid, headers=headers)) + festival_info = {} if is_festival: - play_info = self._download_playinfo(video_id, cid, headers=headers) - festival_info = traverse_obj(initial_state, { 'uploader': ('videoInfo', 'upName'), 'uploader_id': ('videoInfo', 'upMid', {str_or_none}),