X-Git-Url: https://jfr.im/git/yt-dlp.git/blobdiff_plain/26fdfc3704a278acada27cc420d67c6d3f71423b..4cc99d7b6cce8b39506ead01407445d576b63ee4:/yt_dlp/extractor/bilibili.py diff --git a/yt_dlp/extractor/bilibili.py b/yt_dlp/extractor/bilibili.py index 3274a427d..b38c90b1d 100644 --- a/yt_dlp/extractor/bilibili.py +++ b/yt_dlp/extractor/bilibili.py @@ -1,21 +1,33 @@ import base64 import functools +import hashlib import itertools +import json import math -import urllib.error +import re +import time import urllib.parse +import uuid from .common import InfoExtractor, SearchInfoExtractor +from ..dependencies import Cryptodome +from ..networking.exceptions import HTTPError from ..utils import ( ExtractorError, GeoRestrictedError, InAdvancePagedList, OnDemandPagedList, + bool_or_none, + clean_html, + determine_ext, filter_dict, float_or_none, format_field, + get_element_by_class, int_or_none, + join_nonempty, make_archive_id, + merge_dicts, mimetype2ext, parse_count, parse_qs, @@ -24,30 +36,36 @@ srt_subtitles_timecode, str_or_none, traverse_obj, + try_call, + unified_timestamp, unsmuggle_url, url_or_none, urlencode_postdata, + variadic, ) class BilibiliBaseIE(InfoExtractor): + _FORMAT_ID_RE = re.compile(r'-(\d+)\.m4s\?') + def extract_formats(self, play_info): format_names = { r['quality']: traverse_obj(r, 'new_description', 'display_desc') for r in traverse_obj(play_info, ('support_formats', lambda _, v: v['quality'])) } - audios = traverse_obj(play_info, ('dash', 'audio', ...)) + audios = traverse_obj(play_info, ('dash', (None, 'dolby'), 'audio', ..., {dict})) flac_audio = traverse_obj(play_info, ('dash', 'flac', 'audio')) if flac_audio: audios.append(flac_audio) formats = [{ 'url': traverse_obj(audio, 'baseUrl', 'base_url', 'url'), 'ext': mimetype2ext(traverse_obj(audio, 'mimeType', 'mime_type')), - 'acodec': audio.get('codecs'), + 'acodec': traverse_obj(audio, ('codecs', {str.lower})), 'vcodec': 'none', 'tbr': float_or_none(audio.get('bandwidth'), scale=1000), - 'filesize': int_or_none(audio.get('size')) + 'filesize': int_or_none(audio.get('size')), + 'format_id': str_or_none(audio.get('id')), } for audio in audios] formats.extend({ @@ -58,9 +76,13 @@ def extract_formats(self, play_info): 'height': int_or_none(video.get('height')), 'vcodec': video.get('codecs'), 'acodec': 'none' if audios else None, + 'dynamic_range': {126: 'DV', 125: 'HDR10'}.get(int_or_none(video.get('id'))), 'tbr': float_or_none(video.get('bandwidth'), scale=1000), 'filesize': int_or_none(video.get('size')), 'quality': int_or_none(video.get('id')), + 'format_id': traverse_obj( + video, (('baseUrl', 'base_url'), {self._FORMAT_ID_RE.search}, 1), + ('id', {str_or_none}), get_all=False), 'format': format_names.get(video.get('id')), } for video in traverse_obj(play_info, ('dash', 'video', ...))) @@ -71,6 +93,12 @@ def extract_formats(self, play_info): return formats + def _download_playinfo(self, video_id, cid, headers=None): + return self._download_json( + 'https://api.bilibili.com/x/player/playurl', video_id, + query={'bvid': video_id, 'cid': cid, 'fnval': 4048}, + note=f'Downloading video formats for cid {cid}', headers=headers)['data'] + def json2srt(self, json_data): srt_data = '' for idx, line in enumerate(json_data.get('body') or []): @@ -79,7 +107,7 @@ def json2srt(self, json_data): f'{line["content"]}\n\n') return srt_data - def _get_subtitles(self, video_id, initial_state, cid): + def _get_subtitles(self, video_id, cid, aid=None): subtitles = { 'danmaku': [{ 'ext': 'xml', @@ -87,7 +115,15 @@ def _get_subtitles(self, video_id, initial_state, cid): }] } - for s in traverse_obj(initial_state, ('videoData', 'subtitle', 'list')) or []: + subtitle_info = traverse_obj(self._download_json( + 'https://api.bilibili.com/x/player/v2', video_id, + query={'aid': aid, 'cid': cid} if aid else {'bvid': video_id, 'cid': cid}, + note=f'Extracting subtitle info {cid}'), ('data', 'subtitle')) + subs_list = traverse_obj(subtitle_info, ('subtitles', lambda _, v: v['subtitle_url'] and v['lan'])) + if not subs_list and traverse_obj(subtitle_info, 'allow_submit'): + if not self._get_cookies('https://api.bilibili.com').get('SESSDATA'): # no login session cookie + self.report_warning(f'CC subtitles (if any) are only visible when logged in. {self._login_hint()}', only_once=True) + for s in subs_list: subtitles.setdefault(s['lan'], []).append({ 'ext': 'srt', 'data': self.json2srt(self._download_json(s['subtitle_url'], video_id)) @@ -128,9 +164,67 @@ def _get_all_children(self, reply): for children in map(self._get_all_children, traverse_obj(reply, ('replies', ...))): yield from children + def _get_episodes_from_season(self, ss_id, url): + season_info = self._download_json( + 'https://api.bilibili.com/pgc/web/season/section', ss_id, + note='Downloading season info', query={'season_id': ss_id}, + headers={'Referer': url, **self.geo_verification_headers()}) + + for entry in traverse_obj(season_info, ( + 'result', 'main_section', 'episodes', + lambda _, v: url_or_none(v['share_url']) and v['id'])): + yield self.url_result(entry['share_url'], BiliBiliBangumiIE, str_or_none(entry.get('id'))) + + def _get_divisions(self, video_id, graph_version, edges, edge_id, cid_edges=None): + cid_edges = cid_edges or {} + division_data = self._download_json( + 'https://api.bilibili.com/x/stein/edgeinfo_v2', video_id, + query={'graph_version': graph_version, 'edge_id': edge_id, 'bvid': video_id}, + note=f'Extracting divisions from edge {edge_id}') + edges.setdefault(edge_id, {}).update( + traverse_obj(division_data, ('data', 'story_list', lambda _, v: v['edge_id'] == edge_id, { + 'title': ('title', {str}), + 'cid': ('cid', {int_or_none}), + }), get_all=False)) + + edges[edge_id].update(traverse_obj(division_data, ('data', { + 'title': ('title', {str}), + 'choices': ('edges', 'questions', ..., 'choices', ..., { + 'edge_id': ('id', {int_or_none}), + 'cid': ('cid', {int_or_none}), + 'text': ('option', {str}), + }), + }))) + # use dict to combine edges that use the same video section (same cid) + cid_edges.setdefault(edges[edge_id]['cid'], {})[edge_id] = edges[edge_id] + for choice in traverse_obj(edges, (edge_id, 'choices', ...)): + if choice['edge_id'] not in edges: + edges[choice['edge_id']] = {'cid': choice['cid']} + self._get_divisions(video_id, graph_version, edges, choice['edge_id'], cid_edges=cid_edges) + return cid_edges + + def _get_interactive_entries(self, video_id, cid, metainfo): + graph_version = traverse_obj( + self._download_json( + 'https://api.bilibili.com/x/player/wbi/v2', video_id, + 'Extracting graph version', query={'bvid': video_id, 'cid': cid}), + ('data', 'interaction', 'graph_version', {int_or_none})) + cid_edges = self._get_divisions(video_id, graph_version, {1: {'cid': cid}}, 1) + for cid, edges in cid_edges.items(): + play_info = self._download_playinfo(video_id, cid) + yield { + **metainfo, + 'id': f'{video_id}_{cid}', + 'title': f'{metainfo.get("title")} - {list(edges.values())[0].get("title")}', + 'formats': self.extract_formats(play_info), + 'description': f'{json.dumps(edges, ensure_ascii=False)}\n{metainfo.get("description", "")}', + 'duration': float_or_none(play_info.get('timelength'), scale=1000), + 'subtitles': self.extract_subtitles(video_id, cid), + } + class BiliBiliIE(BilibiliBaseIE): - _VALID_URL = r'https?://www\.bilibili\.com/video/[aAbB][vV](?P[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/(?:video/|festival/\w+\?(?:[^#]*&)?bvid=)[aAbB][vV](?P[^/?#&]+)' _TESTS = [{ 'url': 'https://www.bilibili.com/video/BV13x41117TL', @@ -151,7 +245,7 @@ class BiliBiliIE(BilibiliBaseIE): 'view_count': int, }, }, { - # old av URL version + 'note': 'old av URL version', 'url': 'http://www.bilibili.com/video/av1074402/', 'info_dict': { 'thumbnail': r're:^https?://.*\.(jpg|jpeg)$', @@ -183,7 +277,7 @@ class BiliBiliIE(BilibiliBaseIE): 'id': 'BV1bK411W797_p1', 'ext': 'mp4', 'title': '物语中的人物是如何吐槽自己的OP的 p01 Staple Stable/战场原+羽川', - 'tags': 'count:11', + 'tags': 'count:10', 'timestamp': 1589601697, 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', 'uploader': '打牌还是打桩', @@ -203,7 +297,7 @@ class BiliBiliIE(BilibiliBaseIE): 'id': 'BV1bK411W797_p1', 'ext': 'mp4', 'title': '物语中的人物是如何吐槽自己的OP的 p01 Staple Stable/战场原+羽川', - 'tags': 'count:11', + 'tags': 'count:10', 'timestamp': 1589601697, 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', 'uploader': '打牌还是打桩', @@ -226,7 +320,7 @@ class BiliBiliIE(BilibiliBaseIE): 'description': 'md5:afde2b7ba9025c01d9e3dde10de221e4', 'duration': 313.557, 'upload_date': '20220709', - 'uploader': '小夫Tech', + 'uploader': '小夫太渴', 'timestamp': 1657347907, 'uploader_id': '1326814124', 'comment_count': int, @@ -278,23 +372,167 @@ class BiliBiliIE(BilibiliBaseIE): 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', }, 'params': {'skip_download': True}, + }, { + 'note': 'video redirects to festival page', + 'url': 'https://www.bilibili.com/video/BV1wP4y1P72h', + 'info_dict': { + 'id': 'BV1wP4y1P72h', + 'ext': 'mp4', + 'title': '牛虎年相交之际,一首传统民族打击乐《牛斗虎》祝大家新春快乐,虎年大吉!【bilibili音乐虎闹新春】', + 'timestamp': 1643947497, + 'upload_date': '20220204', + 'description': 'md5:8681a0d4d2c06b4ae27e59c8080a7fe6', + 'uploader': '叨叨冯聊音乐', + 'duration': 246.719, + 'uploader_id': '528182630', + 'view_count': int, + 'like_count': int, + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + }, + 'params': {'skip_download': True}, + }, { + 'note': 'newer festival video', + 'url': 'https://www.bilibili.com/festival/2023honkaiimpact3gala?bvid=BV1ay4y1d77f', + 'info_dict': { + 'id': 'BV1ay4y1d77f', + 'ext': 'mp4', + 'title': '【崩坏3新春剧场】为特别的你送上祝福!', + 'timestamp': 1674273600, + 'upload_date': '20230121', + 'description': 'md5:58af66d15c6a0122dc30c8adfd828dd8', + 'uploader': '果蝇轰', + 'duration': 1111.722, + 'uploader_id': '8469526', + 'view_count': int, + 'like_count': int, + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + }, + 'params': {'skip_download': True}, + }, { + 'note': 'interactive/split-path video', + 'url': 'https://www.bilibili.com/video/BV1af4y1H7ga/', + 'info_dict': { + 'id': 'BV1af4y1H7ga', + 'title': '【互动游戏】花了大半年时间做的自我介绍~请查收!!', + 'timestamp': 1630500414, + 'upload_date': '20210901', + 'description': 'md5:01113e39ab06e28042d74ac356a08786', + 'tags': list, + 'uploader': '钉宫妮妮Ninico', + 'duration': 1503, + 'uploader_id': '8881297', + 'comment_count': int, + 'view_count': int, + 'like_count': int, + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + }, + 'playlist_count': 33, + 'playlist': [{ + 'info_dict': { + 'id': 'BV1af4y1H7ga_400950101', + 'ext': 'mp4', + 'title': '【互动游戏】花了大半年时间做的自我介绍~请查收!! - 听见猫猫叫~', + 'timestamp': 1630500414, + 'upload_date': '20210901', + 'description': 'md5:db66ac7a2813a94b8291dbce990cc5b2', + 'tags': list, + 'uploader': '钉宫妮妮Ninico', + 'duration': 11.605, + 'uploader_id': '8881297', + 'comment_count': int, + 'view_count': int, + 'like_count': int, + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + }, + }], + }, { + 'note': '301 redirect to bangumi link', + 'url': 'https://www.bilibili.com/video/BV1TE411f7f1', + 'info_dict': { + 'id': '288525', + 'title': '李永乐老师 钱学森弹道和乘波体飞行器是什么?', + 'ext': 'mp4', + 'series': '我和我的祖国', + 'series_id': '4780', + 'season': '幕后纪实', + 'season_id': '28609', + 'season_number': 1, + 'episode': '钱学森弹道和乘波体飞行器是什么?', + 'episode_id': '288525', + 'episode_number': 105, + 'duration': 1183.957, + 'timestamp': 1571648124, + 'upload_date': '20191021', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + }, + }, { + 'url': 'https://www.bilibili.com/video/BV1jL41167ZG/', + 'info_dict': { + 'id': 'BV1jL41167ZG', + 'title': '一场大火引发的离奇死亡!古典推理经典短篇集《不可能犯罪诊断书》!', + 'ext': 'mp4', + }, + 'skip': 'supporter-only video', + }, { + 'url': 'https://www.bilibili.com/video/BV1Ks411f7aQ/', + 'info_dict': { + 'id': 'BV1Ks411f7aQ', + 'title': '【BD1080P】狼与香辛料I【华盟】', + 'ext': 'mp4', + }, + 'skip': 'login required', + }, { + 'url': 'https://www.bilibili.com/video/BV1GJ411x7h7/', + 'info_dict': { + 'id': 'BV1GJ411x7h7', + 'title': '【官方 MV】Never Gonna Give You Up - Rick Astley', + 'ext': 'mp4', + }, + 'skip': 'geo-restricted', }] def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + headers = self.geo_verification_headers() + webpage, urlh = self._download_webpage_handle(url, video_id, headers=headers) + if not self._match_valid_url(urlh.url): + return self.url_result(urlh.url) + initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', video_id) - play_info = self._search_json(r'window\.__playinfo__\s*=', webpage, 'play info', video_id)['data'] - video_data = initial_state['videoData'] + 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'] + 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. - page_list_json = traverse_obj( + page_list_json = not is_festival and traverse_obj( self._download_json( 'https://api.bilibili.com/x/player/pagelist', video_id, fatal=False, query={'bvid': video_id, 'jsonp': 'jsonp'}, - note='Extracting videos in anthology'), + note='Extracting videos in anthology', headers=headers), 'data', expected_type=list) or [] is_anthology = len(page_list_json) > 1 @@ -313,121 +551,422 @@ def _real_extract(self, url): cid = traverse_obj(video_data, ('pages', part_id - 1, 'cid')) if part_id else video_data.get('cid') - return { + 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}), + 'like_count': ('videoStatus', 'like', {int_or_none}), + 'thumbnail': ('sectionEpisodes', lambda _, v: v['bvid'] == video_id, 'cover'), + }, get_all=False) + + metainfo = { + **traverse_obj(initial_state, { + 'uploader': ('upData', 'name'), + 'uploader_id': ('upData', 'mid', {str_or_none}), + 'like_count': ('videoData', 'stat', 'like', {int_or_none}), + 'tags': ('tags', ..., 'tag_name'), + 'thumbnail': ('videoData', 'pic', {url_or_none}), + }), + **festival_info, + **traverse_obj(video_data, { + 'description': 'desc', + 'timestamp': ('pubdate', {int_or_none}), + 'view_count': (('viewCount', ('stat', 'view')), {int_or_none}), + 'comment_count': ('stat', 'reply', {int_or_none}), + }, get_all=False), 'id': f'{video_id}{format_field(part_id, None, "_p%d")}', - 'formats': self.extract_formats(play_info), '_old_archive_ids': [make_archive_id(self, old_video_id)] if old_video_id else None, 'title': title, - 'description': traverse_obj(initial_state, ('videoData', 'desc')), - 'view_count': traverse_obj(initial_state, ('videoData', 'stat', 'view')), - 'uploader': traverse_obj(initial_state, ('upData', 'name')), - 'uploader_id': traverse_obj(initial_state, ('upData', 'mid')), - 'like_count': traverse_obj(initial_state, ('videoData', 'stat', 'like')), - 'comment_count': traverse_obj(initial_state, ('videoData', 'stat', 'reply')), - 'tags': traverse_obj(initial_state, ('tags', ..., 'tag_name')), - 'thumbnail': traverse_obj(initial_state, ('videoData', 'pic')), - 'timestamp': traverse_obj(initial_state, ('videoData', 'pubdate')), - 'duration': float_or_none(play_info.get('timelength'), scale=1000), - 'chapters': self._get_chapters(aid, cid), - 'subtitles': self.extract_subtitles(video_id, initial_state, cid), - '__post_extractor': self.extract_comments(aid), 'http_headers': {'Referer': url}, } + is_interactive = traverse_obj(video_data, ('rights', 'is_stein_gate')) + if is_interactive: + return self.playlist_result( + self._get_interactive_entries(video_id, cid, metainfo), **metainfo, **{ + 'duration': traverse_obj(initial_state, ('videoData', 'duration', {int_or_none})), + '__post_extractor': self.extract_comments(aid), + }) + else: + return { + **metainfo, + 'duration': float_or_none(play_info.get('timelength'), scale=1000), + 'chapters': self._get_chapters(aid, cid), + 'subtitles': self.extract_subtitles(video_id, cid), + 'formats': self.extract_formats(play_info), + '__post_extractor': self.extract_comments(aid), + } + class BiliBiliBangumiIE(BilibiliBaseIE): - _VALID_URL = r'(?x)https?://www\.bilibili\.com/bangumi/play/(?P(?:ss|ep)\d+)' + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/bangumi/play/ep(?P\d+)' _TESTS = [{ - 'url': 'https://www.bilibili.com/bangumi/play/ss897', + 'url': 'https://www.bilibili.com/bangumi/play/ep21495/', 'info_dict': { - 'id': 'ss897', + 'id': '21495', 'ext': 'mp4', - 'series': '神的记事本', - 'season': '神的记事本', - 'season_id': 897, - 'season_number': 1, - 'episode': '你与旅行包', - 'episode_number': 2, - 'title': '神的记事本:第2话 你与旅行包', - 'duration': 1428.487, - 'timestamp': 1310809380, - 'upload_date': '20110716', + 'series': '悠久之翼', + 'series_id': '774', + 'season': '第二季', + 'season_id': '1182', + 'season_number': 2, + 'episode': 'forever/ef', + 'episode_id': '21495', + 'episode_number': 12, + 'title': '12 forever/ef', + 'duration': 1420.791, + 'timestamp': 1320412200, + 'upload_date': '20111104', 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', }, }, { - 'url': 'https://www.bilibili.com/bangumi/play/ep508406', - 'only_matching': True, + 'url': 'https://www.bilibili.com/bangumi/play/ep267851', + 'info_dict': { + 'id': '267851', + 'ext': 'mp4', + 'series': '鬼灭之刃', + 'series_id': '4358', + 'season': '立志篇', + 'season_id': '26801', + 'season_number': 1, + 'episode': '残酷', + 'episode_id': '267851', + 'episode_number': 1, + 'title': '1 残酷', + 'duration': 1425.256, + 'timestamp': 1554566400, + 'upload_date': '20190406', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$' + }, + 'skip': 'Geo-restricted', + }, { + 'note': 'a making-of which falls outside main section', + 'url': 'https://www.bilibili.com/bangumi/play/ep345120', + 'info_dict': { + 'id': '345120', + 'ext': 'mp4', + 'series': '鬼灭之刃', + 'series_id': '4358', + 'season': '立志篇', + 'season_id': '26801', + 'season_number': 1, + 'episode': '炭治郎篇', + 'episode_id': '345120', + 'episode_number': 27, + 'title': '#1 炭治郎篇', + 'duration': 1922.129, + 'timestamp': 1602853860, + 'upload_date': '20201016', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$' + }, }] def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + episode_id = self._match_id(url) + headers = self.geo_verification_headers() + webpage = self._download_webpage(url, episode_id, headers=headers) if '您所在的地区无法观看本片' in webpage: raise GeoRestrictedError('This video is restricted') - elif ('开通大会员观看' in webpage and '__playinfo__' not in webpage - or '正在观看预览,大会员免费看全片' in webpage): + elif '正在观看预览,大会员免费看全片' in webpage: self.raise_login_required('This video is for premium members only') - play_info = self._search_json(r'window\.__playinfo__\s*=', webpage, 'play info', video_id)['data'] + headers['Referer'] = url + play_info = self._download_json( + 'https://api.bilibili.com/pgc/player/web/v2/playurl', episode_id, + 'Extracting episode', query={'fnval': '4048', 'ep_id': episode_id}, + headers=headers) + premium_only = play_info.get('code') == -10403 + play_info = traverse_obj(play_info, ('result', 'video_info', {dict})) or {} + formats = self.extract_formats(play_info) - if (not formats and '成为大会员抢先看' in webpage - and play_info.get('durl') and not play_info.get('dash')): + if not formats and (premium_only or '成为大会员抢先看' in webpage or '开通大会员观看' in webpage): self.raise_login_required('This video is for premium members only') - initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', video_id) + bangumi_info = self._download_json( + 'https://api.bilibili.com/pgc/view/web/season', episode_id, 'Get episode details', + query={'ep_id': episode_id}, headers=headers)['result'] - season_id = traverse_obj(initial_state, ('mediaInfo', 'season_id')) - season_number = season_id and next(( - idx + 1 for idx, e in enumerate( - traverse_obj(initial_state, ('mediaInfo', 'seasons', ...))) + episode_number, episode_info = next(( + (idx, ep) for idx, ep in enumerate(traverse_obj( + bangumi_info, (('episodes', ('section', ..., 'episodes')), ..., {dict})), 1) + if str_or_none(ep.get('id')) == episode_id), (1, {})) + + season_id = bangumi_info.get('season_id') + season_number, season_title = season_id and next(( + (idx + 1, e.get('season_title')) for idx, e in enumerate( + traverse_obj(bangumi_info, ('seasons', ...))) if e.get('season_id') == season_id - ), None) + ), (None, None)) + + aid = episode_info.get('aid') return { - 'id': video_id, + 'id': episode_id, 'formats': formats, - 'title': traverse_obj(initial_state, 'h1Title'), - 'episode': traverse_obj(initial_state, ('epInfo', 'long_title')), - 'episode_number': int_or_none(traverse_obj(initial_state, ('epInfo', 'title'))), - 'series': traverse_obj(initial_state, ('mediaInfo', 'series')), - 'season': traverse_obj(initial_state, ('mediaInfo', 'season_title')), - 'season_id': season_id, + **traverse_obj(bangumi_info, { + 'series': ('series', 'series_title', {str}), + 'series_id': ('series', 'series_id', {str_or_none}), + 'thumbnail': ('square_cover', {url_or_none}), + }), + **traverse_obj(episode_info, { + 'episode': ('long_title', {str}), + 'episode_number': ('title', {int_or_none}, {lambda x: x or episode_number}), + 'timestamp': ('pub_time', {int_or_none}), + 'title': {lambda v: v and join_nonempty('title', 'long_title', delim=' ', from_dict=v)}, + }), + 'episode_id': episode_id, + 'season': str_or_none(season_title), + 'season_id': str_or_none(season_id), 'season_number': season_number, - 'thumbnail': traverse_obj(initial_state, ('epInfo', 'cover')), - 'timestamp': traverse_obj(initial_state, ('epInfo', 'pub_time')), 'duration': float_or_none(play_info.get('timelength'), scale=1000), - 'subtitles': self.extract_subtitles( - video_id, initial_state, traverse_obj(initial_state, ('epInfo', 'cid'))), - '__post_extractor': self.extract_comments(traverse_obj(initial_state, ('epInfo', 'aid'))), - 'http_headers': {'Referer': url, **self.geo_verification_headers()}, + 'subtitles': self.extract_subtitles(episode_id, episode_info.get('cid'), aid=aid), + '__post_extractor': self.extract_comments(aid), + 'http_headers': {'Referer': url}, } -class BiliBiliBangumiMediaIE(InfoExtractor): - _VALID_URL = r'https?://www\.bilibili\.com/bangumi/media/md(?P\d+)' +class BiliBiliBangumiMediaIE(BilibiliBaseIE): + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/bangumi/media/md(?P\d+)' _TESTS = [{ 'url': 'https://www.bilibili.com/bangumi/media/md24097891', 'info_dict': { 'id': '24097891', + 'title': 'CAROLE & TUESDAY', + 'description': 'md5:42417ad33d1eaa1c93bfd2dd1626b829', }, 'playlist_mincount': 25, + }, { + 'url': 'https://www.bilibili.com/bangumi/media/md1565/', + 'info_dict': { + 'id': '1565', + 'title': '攻壳机动队 S.A.C. 2nd GIG', + 'description': 'md5:46cac00bafd645b97f4d6df616fc576d', + }, + 'playlist_count': 26, + 'playlist': [{ + 'info_dict': { + 'id': '68540', + 'ext': 'mp4', + 'series': '攻壳机动队', + 'series_id': '1077', + 'season': '第二季', + 'season_id': '1565', + 'season_number': 2, + 'episode': '再启动 REEMBODY', + 'episode_id': '68540', + 'episode_number': 1, + 'title': '1 再启动 REEMBODY', + 'duration': 1525.777, + 'timestamp': 1425074413, + 'upload_date': '20150227', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$' + }, + }], }] def _real_extract(self, url): media_id = self._match_id(url) webpage = self._download_webpage(url, media_id) - initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial_state', media_id) - episode_list = self._download_json( - 'https://api.bilibili.com/pgc/web/season/section', media_id, - query={'season_id': initial_state['mediaInfo']['season_id']}, - note='Downloading season info')['result']['main_section']['episodes'] + initial_state = self._search_json( + r'window\.__INITIAL_STATE__\s*=', webpage, 'initial_state', media_id) + ss_id = initial_state['mediaInfo']['season_id'] + + return self.playlist_result( + self._get_episodes_from_season(ss_id, url), media_id, + **traverse_obj(initial_state, ('mediaInfo', { + 'title': ('title', {str}), + 'description': ('evaluate', {str}), + }))) + + +class BiliBiliBangumiSeasonIE(BilibiliBaseIE): + _VALID_URL = r'(?x)https?://(?:www\.)?bilibili\.com/bangumi/play/ss(?P\d+)' + _TESTS = [{ + 'url': 'https://www.bilibili.com/bangumi/play/ss26801', + 'info_dict': { + 'id': '26801', + 'title': '鬼灭之刃', + 'description': 'md5:e2cc9848b6f69be6db79fc2a82d9661b', + }, + 'playlist_mincount': 26 + }, { + 'url': 'https://www.bilibili.com/bangumi/play/ss2251', + 'info_dict': { + 'id': '2251', + 'title': '玲音', + 'description': 'md5:1fd40e3df4c08d4d9d89a6a34844bdc4', + }, + 'playlist_count': 13, + 'playlist': [{ + 'info_dict': { + 'id': '50188', + 'ext': 'mp4', + 'series': '玲音', + 'series_id': '1526', + 'season': 'TV', + 'season_id': '2251', + 'season_number': 1, + 'episode': 'WEIRD', + 'episode_id': '50188', + 'episode_number': 1, + 'title': '1 WEIRD', + 'duration': 1436.992, + 'timestamp': 1343185080, + 'upload_date': '20120725', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$' + }, + }], + }] + + def _real_extract(self, url): + ss_id = self._match_id(url) + webpage = self._download_webpage(url, ss_id) + metainfo = traverse_obj( + self._search_json(r']+type="application/ld\+json"[^>]*>', webpage, 'info', ss_id), + ('itemListElement', ..., { + 'title': ('name', {str}), + 'description': ('description', {str}), + }), get_all=False) + + return self.playlist_result(self._get_episodes_from_season(ss_id, url), ss_id, **metainfo) + + +class BilibiliCheeseBaseIE(BilibiliBaseIE): + _HEADERS = {'Referer': 'https://www.bilibili.com/'} + + def _extract_episode(self, season_info, ep_id): + episode_info = traverse_obj(season_info, ( + 'episodes', lambda _, v: v['id'] == int(ep_id)), get_all=False) + aid, cid = episode_info['aid'], episode_info['cid'] + + if traverse_obj(episode_info, 'ep_status') == -1: + raise ExtractorError('This course episode is not yet available.', expected=True) + if not traverse_obj(episode_info, 'playable'): + self.raise_login_required('You need to purchase the course to download this episode') + + play_info = self._download_json( + 'https://api.bilibili.com/pugv/player/web/playurl', ep_id, + query={'avid': aid, 'cid': cid, 'ep_id': ep_id, 'fnval': 16, 'fourk': 1}, + headers=self._HEADERS, note='Downloading playinfo')['data'] + + return { + 'id': str_or_none(ep_id), + 'episode_id': str_or_none(ep_id), + 'formats': self.extract_formats(play_info), + 'extractor_key': BilibiliCheeseIE.ie_key(), + 'extractor': BilibiliCheeseIE.IE_NAME, + 'webpage_url': f'https://www.bilibili.com/cheese/play/ep{ep_id}', + **traverse_obj(episode_info, { + 'episode': ('title', {str}), + 'title': {lambda v: v and join_nonempty('index', 'title', delim=' - ', from_dict=v)}, + 'alt_title': ('subtitle', {str}), + 'duration': ('duration', {int_or_none}), + 'episode_number': ('index', {int_or_none}), + 'thumbnail': ('cover', {url_or_none}), + 'timestamp': ('release_date', {int_or_none}), + 'view_count': ('play', {int_or_none}), + }), + **traverse_obj(season_info, { + 'uploader': ('up_info', 'uname', {str}), + 'uploader_id': ('up_info', 'mid', {str_or_none}), + }), + 'subtitles': self.extract_subtitles(ep_id, cid, aid=aid), + '__post_extractor': self.extract_comments(aid), + 'http_headers': self._HEADERS, + } + + def _download_season_info(self, query_key, video_id): + return self._download_json( + f'https://api.bilibili.com/pugv/view/web/season?{query_key}={video_id}', video_id, + headers=self._HEADERS, note='Downloading season info')['data'] + + +class BilibiliCheeseIE(BilibiliCheeseBaseIE): + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/cheese/play/ep(?P\d+)' + _TESTS = [{ + 'url': 'https://www.bilibili.com/cheese/play/ep229832', + 'info_dict': { + 'id': '229832', + 'ext': 'mp4', + 'title': '1 - 课程先导片', + 'alt_title': '视频课 · 3分41秒', + 'uploader': '马督工', + 'uploader_id': '316568752', + 'episode': '课程先导片', + 'episode_id': '229832', + 'episode_number': 1, + 'duration': 221, + 'timestamp': 1695549606, + 'upload_date': '20230924', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + 'view_count': int, + } + }] + + def _real_extract(self, url): + ep_id = self._match_id(url) + return self._extract_episode(self._download_season_info('ep_id', ep_id), ep_id) + + +class BilibiliCheeseSeasonIE(BilibiliCheeseBaseIE): + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/cheese/play/ss(?P\d+)' + _TESTS = [{ + 'url': 'https://www.bilibili.com/cheese/play/ss5918', + 'info_dict': { + 'id': '5918', + 'title': '【限时五折】新闻系学不到:马督工教你做自媒体', + 'description': '帮普通人建立世界模型,降低人与人的沟通门槛', + }, + 'playlist': [{ + 'info_dict': { + 'id': '229832', + 'ext': 'mp4', + 'title': '1 - 课程先导片', + 'alt_title': '视频课 · 3分41秒', + 'uploader': '马督工', + 'uploader_id': '316568752', + 'episode': '课程先导片', + 'episode_id': '229832', + 'episode_number': 1, + 'duration': 221, + 'timestamp': 1695549606, + 'upload_date': '20230924', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + 'view_count': int, + } + }], + 'params': {'playlist_items': '1'}, + }, { + 'url': 'https://www.bilibili.com/cheese/play/ss5918', + 'info_dict': { + 'id': '5918', + 'title': '【限时五折】新闻系学不到:马督工教你做自媒体', + 'description': '帮普通人建立世界模型,降低人与人的沟通门槛', + }, + 'playlist_mincount': 5, + 'skip': 'paid video in list', + }] + + def _get_cheese_entries(self, season_info): + for ep_id in traverse_obj(season_info, ('episodes', lambda _, v: v['episode_can_view'], 'id')): + yield self._extract_episode(season_info, ep_id) - return self.playlist_result(( - self.url_result(entry['share_url'], BiliBiliBangumiIE, entry['aid']) - for entry in episode_list), media_id) + def _real_extract(self, url): + season_id = self._match_id(url) + season_info = self._download_season_info('season_id', season_id) + + return self.playlist_result( + self._get_cheese_entries(season_info), season_id, + **traverse_obj(season_info, { + 'title': ('title', {str}), + 'description': ('subtitle', {str}), + })) class BilibiliSpaceBaseIE(InfoExtractor): @@ -450,27 +989,73 @@ class BilibiliSpaceVideoIE(BilibiliSpaceBaseIE): 'id': '3985676', }, 'playlist_mincount': 178, + }, { + 'url': 'https://space.bilibili.com/313580179/video', + 'info_dict': { + 'id': '313580179', + }, + 'playlist_mincount': 92, }] + def _extract_signature(self, playlist_id): + session_data = self._download_json('https://api.bilibili.com/x/web-interface/nav', playlist_id, fatal=False) + + key_from_url = lambda x: x[x.rfind('/') + 1:].split('.')[0] + img_key = traverse_obj( + session_data, ('data', 'wbi_img', 'img_url', {key_from_url})) or '34478ba821254d9d93542680e3b86100' + sub_key = traverse_obj( + session_data, ('data', 'wbi_img', 'sub_url', {key_from_url})) or '7e16a90d190a4355a78fd00b32a38de6' + + session_key = img_key + sub_key + + signature_values = [] + for position in ( + 46, 47, 18, 2, 53, 8, 23, 32, 15, 50, 10, 31, 58, 3, 45, 35, 27, 43, 5, 49, 33, 9, 42, 19, 29, 28, 14, 39, + 12, 38, 41, 13, 37, 48, 7, 16, 24, 55, 40, 61, 26, 17, 0, 1, 60, 51, 30, 4, 22, 25, 54, 21, 56, 59, 6, 63, + 57, 62, 11, 36, 20, 34, 44, 52 + ): + char_at_position = try_call(lambda: session_key[position]) + if char_at_position: + signature_values.append(char_at_position) + + return ''.join(signature_values)[:32] + def _real_extract(self, url): playlist_id, is_video_url = self._match_valid_url(url).group('id', 'video') if not is_video_url: self.to_screen('A channel URL was given. Only the channel\'s videos will be downloaded. ' 'To download audios, add a "/audio" to the URL') + signature = self._extract_signature(playlist_id) + def fetch_page(page_idx): + query = { + 'keyword': '', + 'mid': playlist_id, + 'order': 'pubdate', + 'order_avoided': 'true', + 'platform': 'web', + 'pn': page_idx + 1, + 'ps': 30, + 'tid': 0, + 'web_location': 1550101, + 'wts': int(time.time()), + } + query['w_rid'] = hashlib.md5(f'{urllib.parse.urlencode(query)}{signature}'.encode()).hexdigest() + try: - response = self._download_json('https://api.bilibili.com/x/space/arc/search', - playlist_id, note=f'Downloading page {page_idx}', - query={'mid': playlist_id, 'pn': page_idx + 1, 'jsonp': 'jsonp'}) + response = self._download_json('https://api.bilibili.com/x/space/wbi/arc/search', + playlist_id, note=f'Downloading page {page_idx}', query=query, + headers={'referer': url}) except ExtractorError as e: - if isinstance(e.cause, urllib.error.HTTPError) and e.cause.code == 412: + if isinstance(e.cause, HTTPError) and e.cause.status == 412: raise ExtractorError( 'Request is blocked by server (412), please add cookies, wait and try later.', expected=True) raise - if response['code'] == -401: + if response['code'] in (-352, -401): raise ExtractorError( - 'Request is blocked by server (401), please add cookies, wait and try later.', expected=True) + f'Request is blocked by server ({-response["code"]}), ' + 'please add cookies, wait and try later.', expected=True) return response['data'] def get_metadata(page_data): @@ -492,9 +1077,9 @@ def get_entries(page_data): class BilibiliSpaceAudioIE(BilibiliSpaceBaseIE): _VALID_URL = r'https?://space\.bilibili\.com/(?P\d+)/audio' _TESTS = [{ - 'url': 'https://space.bilibili.com/3985676/audio', + 'url': 'https://space.bilibili.com/313580179/audio', 'info_dict': { - 'id': '3985676', + 'id': '313580179', }, 'playlist_mincount': 1, }] @@ -522,13 +1107,35 @@ def get_entries(page_data): return self.playlist_result(paged_list, playlist_id) -class BilibiliSpacePlaylistIE(BilibiliSpaceBaseIE): - _VALID_URL = r'https?://space.bilibili\.com/(?P\d+)/channel/collectiondetail\?sid=(?P\d+)' +class BilibiliSpaceListBaseIE(BilibiliSpaceBaseIE): + def _get_entries(self, page_data, bvid_keys, ending_key='bvid'): + for bvid in traverse_obj(page_data, (*variadic(bvid_keys, (str, bytes, dict, set)), ..., ending_key, {str})): + yield self.url_result(f'https://www.bilibili.com/video/{bvid}', BiliBiliIE, bvid) + + def _get_uploader(self, uid, playlist_id): + webpage = self._download_webpage(f'https://space.bilibili.com/{uid}', playlist_id, fatal=False) + return self._search_regex(r'(?s)]*>([^<]+)的个人空间-', webpage, 'uploader', fatal=False) + + def _extract_playlist(self, fetch_page, get_metadata, get_entries): + metadata, page_list = super()._extract_playlist(fetch_page, get_metadata, get_entries) + metadata.pop('page_count', None) + metadata.pop('page_size', None) + return metadata, page_list + + +class BilibiliCollectionListIE(BilibiliSpaceListBaseIE): + _VALID_URL = r'https?://space\.bilibili\.com/(?P\d+)/channel/collectiondetail/?\?sid=(?P\d+)' _TESTS = [{ 'url': 'https://space.bilibili.com/2142762/channel/collectiondetail?sid=57445', 'info_dict': { 'id': '2142762_57445', - 'title': '《底特律 变人》' + 'title': '【完结】《底特律 变人》全结局流程解说', + 'description': '', + 'uploader': '老戴在此', + 'uploader_id': '2142762', + 'timestamp': int, + 'upload_date': str, + 'thumbnail': 'https://archive.biliimg.com/bfs/archive/e0e543ae35ad3df863ea7dea526bc32e70f4c091.jpg', }, 'playlist_mincount': 31, }] @@ -549,22 +1156,276 @@ def get_metadata(page_data): return { 'page_count': math.ceil(entry_count / page_size), 'page_size': page_size, - 'title': traverse_obj(page_data, ('meta', 'name')) + 'uploader': self._get_uploader(mid, playlist_id), + **traverse_obj(page_data, { + 'title': ('meta', 'name', {str}), + 'description': ('meta', 'description', {str}), + 'uploader_id': ('meta', 'mid', {str_or_none}), + 'timestamp': ('meta', 'ptime', {int_or_none}), + 'thumbnail': ('meta', 'cover', {url_or_none}), + }) + } + + def get_entries(page_data): + return self._get_entries(page_data, 'archives') + + metadata, paged_list = self._extract_playlist(fetch_page, get_metadata, get_entries) + return self.playlist_result(paged_list, playlist_id, **metadata) + + +class BilibiliSeriesListIE(BilibiliSpaceListBaseIE): + _VALID_URL = r'https?://space\.bilibili\.com/(?P\d+)/channel/seriesdetail/?\?\bsid=(?P\d+)' + _TESTS = [{ + 'url': 'https://space.bilibili.com/1958703906/channel/seriesdetail?sid=547718&ctype=0', + 'info_dict': { + 'id': '1958703906_547718', + 'title': '直播回放', + 'description': '直播回放', + 'uploader': '靡烟miya', + 'uploader_id': '1958703906', + 'timestamp': 1637985853, + 'upload_date': '20211127', + 'modified_timestamp': int, + 'modified_date': str, + }, + 'playlist_mincount': 513, + }] + + def _real_extract(self, url): + mid, sid = self._match_valid_url(url).group('mid', 'sid') + playlist_id = f'{mid}_{sid}' + playlist_meta = traverse_obj(self._download_json( + f'https://api.bilibili.com/x/series/series?series_id={sid}', playlist_id, fatal=False + ), { + 'title': ('data', 'meta', 'name', {str}), + 'description': ('data', 'meta', 'description', {str}), + 'uploader_id': ('data', 'meta', 'mid', {str_or_none}), + 'timestamp': ('data', 'meta', 'ctime', {int_or_none}), + 'modified_timestamp': ('data', 'meta', 'mtime', {int_or_none}), + }) + + def fetch_page(page_idx): + return self._download_json( + 'https://api.bilibili.com/x/series/archives', + playlist_id, note=f'Downloading page {page_idx}', + query={'mid': mid, 'series_id': sid, 'pn': page_idx + 1, 'ps': 30})['data'] + + def get_metadata(page_data): + page_size = page_data['page']['size'] + entry_count = page_data['page']['total'] + return { + 'page_count': math.ceil(entry_count / page_size), + 'page_size': page_size, + 'uploader': self._get_uploader(mid, playlist_id), + **playlist_meta } def get_entries(page_data): - for entry in page_data.get('archives', []): - yield self.url_result(f'https://www.bilibili.com/video/{entry["bvid"]}', - BiliBiliIE, entry['bvid']) + return self._get_entries(page_data, 'archives') metadata, paged_list = self._extract_playlist(fetch_page, get_metadata, get_entries) - return self.playlist_result(paged_list, playlist_id, metadata['title']) + return self.playlist_result(paged_list, playlist_id, **metadata) + + +class BilibiliFavoritesListIE(BilibiliSpaceListBaseIE): + _VALID_URL = r'https?://(?:space\.bilibili\.com/\d+/favlist/?\?fid=|(?:www\.)?bilibili\.com/medialist/detail/ml)(?P\d+)' + _TESTS = [{ + 'url': 'https://space.bilibili.com/84912/favlist?fid=1103407912&ftype=create', + 'info_dict': { + 'id': '1103407912', + 'title': '【V2】(旧)', + 'description': '', + 'uploader': '晓月春日', + 'uploader_id': '84912', + 'timestamp': 1604905176, + 'upload_date': '20201109', + 'modified_timestamp': int, + 'modified_date': str, + 'thumbnail': r"re:http://i\d\.hdslb\.com/bfs/archive/14b83c62aa8871b79083df1e9ab4fbc699ad16fe\.jpg", + 'view_count': int, + 'like_count': int, + }, + 'playlist_mincount': 22, + }, { + 'url': 'https://www.bilibili.com/medialist/detail/ml1103407912', + 'only_matching': True, + }] + + def _real_extract(self, url): + fid = self._match_id(url) + + list_info = self._download_json( + f'https://api.bilibili.com/x/v3/fav/resource/list?media_id={fid}&pn=1&ps=20', + fid, note='Downloading favlist metadata') + if list_info['code'] == -403: + self.raise_login_required(msg='This is a private favorites list. You need to log in as its owner') + + entries = self._get_entries(self._download_json( + f'https://api.bilibili.com/x/v3/fav/resource/ids?media_id={fid}', + fid, note='Download favlist entries'), 'data') + + return self.playlist_result(entries, fid, **traverse_obj(list_info, ('data', 'info', { + 'title': ('title', {str}), + 'description': ('intro', {str}), + 'uploader': ('upper', 'name', {str}), + 'uploader_id': ('upper', 'mid', {str_or_none}), + 'timestamp': ('ctime', {int_or_none}), + 'modified_timestamp': ('mtime', {int_or_none}), + 'thumbnail': ('cover', {url_or_none}), + 'view_count': ('cnt_info', 'play', {int_or_none}), + 'like_count': ('cnt_info', 'thumb_up', {int_or_none}), + }))) + + +class BilibiliWatchlaterIE(BilibiliSpaceListBaseIE): + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/watchlater/?(?:[?#]|$)' + _TESTS = [{ + 'url': 'https://www.bilibili.com/watchlater/#/list', + 'info_dict': {'id': 'watchlater'}, + 'playlist_mincount': 0, + 'skip': 'login required', + }] + + def _real_extract(self, url): + list_id = getattr(self._get_cookies(url).get('DedeUserID'), 'value', 'watchlater') + watchlater_info = self._download_json( + 'https://api.bilibili.com/x/v2/history/toview/web?jsonp=jsonp', list_id) + if watchlater_info['code'] == -101: + self.raise_login_required(msg='You need to login to access your watchlater list') + entries = self._get_entries(watchlater_info, ('data', 'list')) + return self.playlist_result(entries, id=list_id, title='稍后再看') + + +class BilibiliPlaylistIE(BilibiliSpaceListBaseIE): + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/(?:medialist/play|list)/(?P\w+)' + _TESTS = [{ + 'url': 'https://www.bilibili.com/list/1958703906?sid=547718', + 'info_dict': { + 'id': '5_547718', + 'title': '直播回放', + 'uploader': '靡烟miya', + 'uploader_id': '1958703906', + 'timestamp': 1637985853, + 'upload_date': '20211127', + }, + 'playlist_mincount': 513, + }, { + 'url': 'https://www.bilibili.com/list/1958703906?sid=547718&oid=687146339&bvid=BV1DU4y1r7tz', + 'info_dict': { + 'id': 'BV1DU4y1r7tz', + 'ext': 'mp4', + 'title': '【直播回放】8.20晚9:30 3d发布喵 2022年8月20日21点场', + 'upload_date': '20220820', + 'description': '', + 'timestamp': 1661016330, + 'uploader_id': '1958703906', + 'uploader': '靡烟miya', + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + 'duration': 9552.903, + 'tags': list, + 'comment_count': int, + 'view_count': int, + 'like_count': int, + '_old_archive_ids': ['bilibili 687146339_part1'], + }, + 'params': {'noplaylist': True}, + }, { + 'url': 'https://www.bilibili.com/medialist/play/1958703906?business=space_series&business_id=547718&desc=1', + 'info_dict': { + 'id': '5_547718', + }, + 'playlist_mincount': 513, + 'skip': 'redirect url', + }, { + 'url': 'https://www.bilibili.com/list/ml1103407912', + 'info_dict': { + 'id': '3_1103407912', + 'title': '【V2】(旧)', + 'uploader': '晓月春日', + 'uploader_id': '84912', + 'timestamp': 1604905176, + 'upload_date': '20201109', + 'thumbnail': r"re:http://i\d\.hdslb\.com/bfs/archive/14b83c62aa8871b79083df1e9ab4fbc699ad16fe\.jpg", + }, + 'playlist_mincount': 22, + }, { + 'url': 'https://www.bilibili.com/medialist/play/ml1103407912', + 'info_dict': { + 'id': '3_1103407912', + }, + 'playlist_mincount': 22, + 'skip': 'redirect url', + }, { + 'url': 'https://www.bilibili.com/list/watchlater', + 'info_dict': {'id': 'watchlater'}, + 'playlist_mincount': 0, + 'skip': 'login required', + }, { + 'url': 'https://www.bilibili.com/medialist/play/watchlater', + 'info_dict': {'id': 'watchlater'}, + 'playlist_mincount': 0, + 'skip': 'login required', + }] + + def _extract_medialist(self, query, list_id): + for page_num in itertools.count(1): + page_data = self._download_json( + 'https://api.bilibili.com/x/v2/medialist/resource/list', + list_id, query=query, note=f'getting playlist {query["biz_id"]} page {page_num}' + )['data'] + yield from self._get_entries(page_data, 'media_list', ending_key='bv_id') + query['oid'] = traverse_obj(page_data, ('media_list', -1, 'id')) + if not page_data.get('has_more', False): + break + + def _real_extract(self, url): + list_id = self._match_id(url) + + bvid = traverse_obj(parse_qs(url), ('bvid', 0)) + if not self._yes_playlist(list_id, bvid): + return self.url_result(f'https://www.bilibili.com/video/{bvid}', BiliBiliIE) + + webpage = self._download_webpage(url, list_id) + initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', list_id) + if traverse_obj(initial_state, ('error', 'code', {int_or_none})) != 200: + error_code = traverse_obj(initial_state, ('error', 'trueCode', {int_or_none})) + error_message = traverse_obj(initial_state, ('error', 'message', {str_or_none})) + if error_code == -400 and list_id == 'watchlater': + self.raise_login_required('You need to login to access your watchlater playlist') + elif error_code == -403: + self.raise_login_required('This is a private playlist. You need to login as its owner') + elif error_code == 11010: + raise ExtractorError('Playlist is no longer available', expected=True) + raise ExtractorError(f'Could not access playlist: {error_code} {error_message}') + + query = { + 'ps': 20, + 'with_current': False, + **traverse_obj(initial_state, { + 'type': ('playlist', 'type', {int_or_none}), + 'biz_id': ('playlist', 'id', {int_or_none}), + 'tid': ('tid', {int_or_none}), + 'sort_field': ('sortFiled', {int_or_none}), + 'desc': ('desc', {bool_or_none}, {str_or_none}, {str.lower}), + }) + } + metadata = { + 'id': f'{query["type"]}_{query["biz_id"]}', + **traverse_obj(initial_state, ('mediaListInfo', { + 'title': ('title', {str}), + 'uploader': ('upper', 'name', {str}), + 'uploader_id': ('upper', 'mid', {str_or_none}), + 'timestamp': ('ctime', {int_or_none}), + 'thumbnail': ('cover', {url_or_none}), + })), + } + return self.playlist_result(self._extract_medialist(query, list_id), **metadata) class BilibiliCategoryIE(InfoExtractor): IE_NAME = 'Bilibili category extractor' _MAX_RESULTS = 1000000 - _VALID_URL = r'https?://www\.bilibili\.com/v/[a-zA-Z]+\/[a-zA-Z]+' + _VALID_URL = r'https?://(?:www\.)?bilibili\.com/v/[a-zA-Z]+\/[a-zA-Z]+' _TESTS = [{ 'url': 'https://www.bilibili.com/v/kichiku/mad', 'info_dict': { @@ -633,8 +1494,37 @@ class BiliBiliSearchIE(SearchInfoExtractor): IE_DESC = 'Bilibili video search' _MAX_RESULTS = 100000 _SEARCH_KEY = 'bilisearch' + _TESTS = [{ + 'url': 'bilisearch3:靡烟 出道一年,我怎么还在等你单推的女人睡觉后开播啊', + 'playlist_count': 3, + 'info_dict': { + 'id': '靡烟 出道一年,我怎么还在等你单推的女人睡觉后开播啊', + 'title': '靡烟 出道一年,我怎么还在等你单推的女人睡觉后开播啊', + }, + 'playlist': [{ + 'info_dict': { + 'id': 'BV1n44y1Q7sc', + 'ext': 'mp4', + 'title': '“出道一年,我怎么还在等你单推的女人睡觉后开播啊?”【一分钟了解靡烟miya】', + 'timestamp': 1669889987, + 'upload_date': '20221201', + 'description': 'md5:43343c0973defff527b5a4b403b4abf9', + 'tags': list, + 'uploader': '靡烟miya', + 'duration': 123.156, + 'uploader_id': '1958703906', + 'comment_count': int, + 'view_count': int, + 'like_count': int, + 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$', + '_old_archive_ids': ['bilibili 988222410_part1'], + }, + }], + }] def _search_results(self, query): + if not self._get_cookies('https://api.bilibili.com').get('buvid3'): + self._set_cookie('.bilibili.com', 'buvid3', f'{uuid.uuid4()}infoc') for page_num in itertools.count(1): videos = self._download_json( 'https://api.bilibili.com/x/web-interface/search/type', query, @@ -791,6 +1681,7 @@ def _real_extract(self, url): class BiliIntlBaseIE(InfoExtractor): _API_URL = 'https://api.bilibili.tv/intl/gateway' _NETRC_MACHINE = 'biliintl' + _HEADERS = {'Referer': 'https://www.bilibili.com/'} def _call_api(self, endpoint, *args, **kwargs): json = self._download_json(self._API_URL + endpoint, *args, **kwargs) @@ -828,19 +1719,34 @@ def _get_subtitles(self, *, ep_id=None, aid=None): 'aid': aid, })) or {} subtitles = {} - for sub in sub_json.get('subtitles') or []: - sub_url = sub.get('url') - if not sub_url: - continue - sub_data = self._download_json( - sub_url, ep_id or aid, errnote='Unable to download subtitles', fatal=False, - note='Downloading subtitles%s' % f' for {sub["lang"]}' if sub.get('lang') else '') - if not sub_data: - continue - subtitles.setdefault(sub.get('lang_key', 'en'), []).append({ - 'ext': 'srt', - 'data': self.json2srt(sub_data) - }) + fetched_urls = set() + for sub in traverse_obj(sub_json, (('subtitles', 'video_subtitle'), ..., {dict})): + for url in traverse_obj(sub, ((None, 'ass', 'srt'), 'url', {url_or_none})): + if url in fetched_urls: + continue + fetched_urls.add(url) + sub_ext = determine_ext(url) + sub_lang = sub.get('lang_key') or 'en' + + if sub_ext == 'ass': + subtitles.setdefault(sub_lang, []).append({ + 'ext': 'ass', + 'url': url, + }) + elif sub_ext == 'json': + sub_data = self._download_json( + url, ep_id or aid, fatal=False, + note=f'Downloading subtitles{format_field(sub, "lang", " for %s")} ({sub_lang})', + errnote='Unable to download subtitles') + + if sub_data: + subtitles.setdefault(sub_lang, []).append({ + 'ext': 'srt', + 'data': self.json2srt(sub_data), + }) + else: + self.report_warning('Unexpected subtitle extension', ep_id or aid) + return subtitles def _get_formats(self, *, ep_id=None, aid=None): @@ -886,28 +1792,23 @@ def _get_formats(self, *, ep_id=None, aid=None): def _parse_video_metadata(self, video_data): return { 'title': video_data.get('title_display') or video_data.get('title'), + 'description': video_data.get('desc'), 'thumbnail': video_data.get('cover'), + 'timestamp': unified_timestamp(video_data.get('formatted_pub_date')), 'episode_number': int_or_none(self._search_regex( r'^E(\d+)(?:$| - )', video_data.get('title_display') or '', 'episode number', default=None)), } def _perform_login(self, username, password): - try: - from Cryptodome.PublicKey import RSA - from Cryptodome.Cipher import PKCS1_v1_5 - except ImportError: - try: - from Crypto.PublicKey import RSA - from Crypto.Cipher import PKCS1_v1_5 - except ImportError: - raise ExtractorError('pycryptodomex not found. Please install', expected=True) + if not Cryptodome.RSA: + raise ExtractorError('pycryptodomex not found. Please install', expected=True) key_data = self._download_json( 'https://passport.bilibili.tv/x/intl/passport-login/web/key?lang=en-US', None, note='Downloading login key', errnote='Unable to download login key')['data'] - public_key = RSA.importKey(key_data['key']) - password_hash = PKCS1_v1_5.new(public_key).encrypt((key_data['hash'] + password).encode('utf-8')) + public_key = Cryptodome.RSA.importKey(key_data['key']) + password_hash = Cryptodome.PKCS1_v1_5.new(public_key).encrypt((key_data['hash'] + password).encode('utf-8')) login_post = self._download_json( 'https://passport.bilibili.tv/x/intl/passport-login/web/login/password?lang=en-US', None, data=urlencode_postdata({ 'username': username, @@ -934,6 +1835,23 @@ class BiliIntlIE(BiliIntlBaseIE): 'title': 'E2 - The First Night', 'thumbnail': r're:^https://pic\.bstarstatic\.com/ogv/.+\.png$', 'episode_number': 2, + 'upload_date': '20201009', + 'episode': 'Episode 2', + 'timestamp': 1602259500, + 'description': 'md5:297b5a17155eb645e14a14b385ab547e', + 'chapters': [{ + 'start_time': 0, + 'end_time': 76.242, + 'title': '' + }, { + 'start_time': 76.242, + 'end_time': 161.161, + 'title': 'Intro' + }, { + 'start_time': 1325.742, + 'end_time': 1403.903, + 'title': 'Outro' + }], } }, { # Non-Bstation page @@ -944,6 +1862,23 @@ class BiliIntlIE(BiliIntlBaseIE): 'title': 'E3 - Who?', 'thumbnail': r're:^https://pic\.bstarstatic\.com/ogv/.+\.png$', 'episode_number': 3, + 'description': 'md5:e1a775e71a35c43f141484715470ad09', + 'episode': 'Episode 3', + 'upload_date': '20211219', + 'timestamp': 1639928700, + 'chapters': [{ + 'start_time': 0, + 'end_time': 88.0, + 'title': '' + }, { + 'start_time': 88.0, + 'end_time': 156.0, + 'title': 'Intro' + }, { + 'start_time': 1173.0, + 'end_time': 1259.535, + 'title': 'Outro' + }], } }, { # Subtitle with empty content @@ -956,6 +1891,67 @@ class BiliIntlIE(BiliIntlBaseIE): 'episode_number': 140, }, 'skip': 'According to the copyright owner\'s request, you may only watch the video after you log in.' + }, { + # episode comment extraction + 'url': 'https://www.bilibili.tv/en/play/34580/340317', + 'info_dict': { + 'id': '340317', + 'ext': 'mp4', + 'timestamp': 1604057820, + 'upload_date': '20201030', + 'episode_number': 5, + 'title': 'E5 - My Own Steel', + 'description': 'md5:2b17ab10aebb33e3c2a54da9e8e487e2', + 'thumbnail': r're:https?://pic\.bstarstatic\.com/ogv/.+\.png$', + 'episode': 'Episode 5', + 'comment_count': int, + 'chapters': [{ + 'start_time': 0, + 'end_time': 61.0, + 'title': '' + }, { + 'start_time': 61.0, + 'end_time': 134.0, + 'title': 'Intro' + }, { + 'start_time': 1290.0, + 'end_time': 1379.0, + 'title': 'Outro' + }], + }, + 'params': { + 'getcomments': True + } + }, { + # user generated content comment extraction + 'url': 'https://www.bilibili.tv/en/video/2045730385', + 'info_dict': { + 'id': '2045730385', + 'ext': 'mp4', + 'description': 'md5:693b6f3967fb4e7e7764ea817857c33a', + 'timestamp': 1667891924, + 'upload_date': '20221108', + 'title': 'That Time I Got Reincarnated as a Slime: Scarlet Bond - Official Trailer 3| AnimeStan', + 'comment_count': int, + 'thumbnail': r're:https://pic\.bstarstatic\.(?:com|net)/ugc/f6c363659efd2eabe5683fbb906b1582\.jpg', + }, + 'params': { + 'getcomments': True + } + }, { + # episode id without intro and outro + 'url': 'https://www.bilibili.tv/en/play/1048837/11246489', + 'info_dict': { + 'id': '11246489', + 'ext': 'mp4', + 'title': 'E1 - Operation \'Strix\' ', + 'description': 'md5:b4434eb1a9a97ad2bccb779514b89f17', + 'timestamp': 1649516400, + 'thumbnail': 'https://pic.bstarstatic.com/ogv/62cb1de23ada17fb70fbe7bdd6ff29c29da02a64.png', + 'episode': 'Episode 1', + 'episode_number': 1, + 'upload_date': '20220409', + }, }, { 'url': 'https://www.biliintl.com/en/play/34613/341736', 'only_matching': True, @@ -973,6 +1969,7 @@ class BiliIntlIE(BiliIntlBaseIE): 'only_matching': True, }] + @staticmethod def _make_url(video_id, series_id=None): if series_id: return f'https://www.bilibili.tv/en/play/{series_id}/{video_id}' @@ -989,7 +1986,7 @@ def _extract_video_metadata(self, url, video_id, season_id): self._search_json(r'window\.__INITIAL_(?:DATA|STATE)__\s*=', webpage, 'preload state', video_id, default={}) or self._search_nuxt_data(webpage, video_id, '__initialState', fatal=False, traverse=None)) video_data = traverse_obj( - initial_data, ('OgvVideo', 'epDetail'), ('UgcVideo', 'videoData'), ('ugc', 'archive'), expected_type=dict) + initial_data, ('OgvVideo', 'epDetail'), ('UgcVideo', 'videoData'), ('ugc', 'archive'), expected_type=dict) or {} if season_id and not video_data: # Non-Bstation layout, read through episode list @@ -998,37 +1995,134 @@ def _extract_video_metadata(self, url, video_id, season_id): 'sections', ..., 'episodes', lambda _, v: str(v['episode_id']) == video_id ), expected_type=dict, get_all=False) - return self._parse_video_metadata(video_data) + # XXX: webpage metadata may not accurate, it just used to not crash when video_data not found + return merge_dicts( + self._parse_video_metadata(video_data), { + 'title': get_element_by_class( + 'bstar-meta__title', webpage) or self._html_search_meta('og:title', webpage), + 'description': get_element_by_class( + 'bstar-meta__desc', webpage) or self._html_search_meta('og:description', webpage), + }, self._search_json_ld(webpage, video_id, default={})) + + def _get_comments_reply(self, root_id, next_id=0, display_id=None): + comment_api_raw_data = self._download_json( + 'https://api.bilibili.tv/reply/web/detail', display_id, + note=f'Downloading reply comment of {root_id} - {next_id}', + query={ + 'platform': 'web', + 'ps': 20, # comment's reply per page (default: 3) + 'root': root_id, + 'next': next_id, + }) + + for replies in traverse_obj(comment_api_raw_data, ('data', 'replies', ...)): + yield { + 'author': traverse_obj(replies, ('member', 'name')), + 'author_id': traverse_obj(replies, ('member', 'mid')), + 'author_thumbnail': traverse_obj(replies, ('member', 'face')), + 'text': traverse_obj(replies, ('content', 'message')), + 'id': replies.get('rpid'), + 'like_count': int_or_none(replies.get('like_count')), + 'parent': replies.get('parent'), + 'timestamp': unified_timestamp(replies.get('ctime_text')) + } + + if not traverse_obj(comment_api_raw_data, ('data', 'cursor', 'is_end')): + yield from self._get_comments_reply( + root_id, comment_api_raw_data['data']['cursor']['next'], display_id) + + def _get_comments(self, video_id, ep_id): + for i in itertools.count(0): + comment_api_raw_data = self._download_json( + 'https://api.bilibili.tv/reply/web/root', video_id, + note=f'Downloading comment page {i + 1}', + query={ + 'platform': 'web', + 'pn': i, # page number + 'ps': 20, # comment per page (default: 20) + 'oid': video_id, + 'type': 3 if ep_id else 1, # 1: user generated content, 3: series content + 'sort_type': 1, # 1: best, 2: recent + }) + + for replies in traverse_obj(comment_api_raw_data, ('data', 'replies', ...)): + yield { + 'author': traverse_obj(replies, ('member', 'name')), + 'author_id': traverse_obj(replies, ('member', 'mid')), + 'author_thumbnail': traverse_obj(replies, ('member', 'face')), + 'text': traverse_obj(replies, ('content', 'message')), + 'id': replies.get('rpid'), + 'like_count': int_or_none(replies.get('like_count')), + 'timestamp': unified_timestamp(replies.get('ctime_text')), + 'author_is_uploader': bool(traverse_obj(replies, ('member', 'type'))), + } + if replies.get('count'): + yield from self._get_comments_reply(replies.get('rpid'), display_id=video_id) + + if traverse_obj(comment_api_raw_data, ('data', 'cursor', 'is_end')): + break def _real_extract(self, url): season_id, ep_id, aid = self._match_valid_url(url).group('season_id', 'ep_id', 'aid') video_id = ep_id or aid + chapters = None + + if ep_id: + intro_ending_json = self._call_api( + f'/web/v2/ogv/play/episode?episode_id={ep_id}&platform=web', + video_id, fatal=False) or {} + if intro_ending_json.get('skip'): + # FIXME: start time and end time seems a bit off a few second even it corrext based on ogv.*.js + # ref: https://p.bstarstatic.com/fe-static/bstar-web-new/assets/ogv.2b147442.js + chapters = [{ + 'start_time': float_or_none(traverse_obj(intro_ending_json, ('skip', 'opening_start_time')), 1000), + 'end_time': float_or_none(traverse_obj(intro_ending_json, ('skip', 'opening_end_time')), 1000), + 'title': 'Intro' + }, { + 'start_time': float_or_none(traverse_obj(intro_ending_json, ('skip', 'ending_start_time')), 1000), + 'end_time': float_or_none(traverse_obj(intro_ending_json, ('skip', 'ending_end_time')), 1000), + 'title': 'Outro' + }] return { 'id': video_id, **self._extract_video_metadata(url, video_id, season_id), 'formats': self._get_formats(ep_id=ep_id, aid=aid), 'subtitles': self.extract_subtitles(ep_id=ep_id, aid=aid), + 'chapters': chapters, + '__post_extractor': self.extract_comments(video_id, ep_id), + 'http_headers': self._HEADERS, } class BiliIntlSeriesIE(BiliIntlBaseIE): - IE_NAME = 'biliintl:series' - _VALID_URL = r'https?://(?:www\.)?bili(?:bili\.tv|intl\.com)/(?:[a-zA-Z]{2}/)?play/(?P\d+)/?(?:[?#]|$)' + IE_NAME = 'biliIntl:series' + _VALID_URL = r'https?://(?:www\.)?bili(?:bili\.tv|intl\.com)/(?:[a-zA-Z]{2}/)?(?:play|media)/(?P\d+)/?(?:[?#]|$)' _TESTS = [{ 'url': 'https://www.bilibili.tv/en/play/34613', 'playlist_mincount': 15, 'info_dict': { 'id': '34613', - 'title': 'Fly Me to the Moon', - 'description': 'md5:a861ee1c4dc0acfad85f557cc42ac627', - 'categories': ['Romance', 'Comedy', 'Slice of life'], + 'title': 'TONIKAWA: Over the Moon For You', + 'description': 'md5:297b5a17155eb645e14a14b385ab547e', + 'categories': ['Slice of life', 'Comedy', 'Romance'], 'thumbnail': r're:^https://pic\.bstarstatic\.com/ogv/.+\.png$', 'view_count': int, }, 'params': { 'skip_download': True, }, + }, { + 'url': 'https://www.bilibili.tv/en/media/1048837', + 'info_dict': { + 'id': '1048837', + 'title': 'SPY×FAMILY', + 'description': 'md5:b4434eb1a9a97ad2bccb779514b89f17', + 'categories': ['Adventure', 'Action', 'Comedy'], + 'thumbnail': r're:^https://pic\.bstarstatic\.com/ogv/.+\.jpg$', + 'view_count': int, + }, + 'playlist_mincount': 25, }, { 'url': 'https://www.biliintl.com/en/play/34613', 'only_matching': True, @@ -1056,7 +2150,7 @@ def _real_extract(self, url): class BiliLiveIE(InfoExtractor): - _VALID_URL = r'https?://live.bilibili.com/(?:blanc/)?(?P\d+)' + _VALID_URL = r'https?://live\.bilibili\.com/(?:blanc/)?(?P\d+)' _TESTS = [{ 'url': 'https://live.bilibili.com/196',