yt_dlp/extractor/acfun.py

   1 from .common import InfoExtractor
   2 from ..utils import (
   3     float_or_none,
   4     format_field,
   5     int_or_none,
   6     parse_codecs,
   7     parse_qs,
   8     str_or_none,
   9     traverse_obj,
  10 )
  11
  12
  13 class AcFunVideoBaseIE(InfoExtractor):
  14     def _extract_metadata(self, video_id, video_info):
  15         playjson = self._parse_json(video_info['ksPlayJson'], video_id)
  16
  17         formats, subtitles = [], {}
  18         for video in traverse_obj(playjson, ('adaptationSet', 0, 'representation')):
  19             fmts, subs = self._extract_m3u8_formats_and_subtitles(video['url'], video_id, 'mp4', fatal=False)
  20             formats.extend(fmts)
  21             self._merge_subtitles(subs, target=subtitles)
  22             for f in fmts:
  23                 f.update({
  24                     'fps': float_or_none(video.get('frameRate')),
  25                     'width': int_or_none(video.get('width')),
  26                     'height': int_or_none(video.get('height')),
  27                     'tbr': float_or_none(video.get('avgBitrate')),
  28                     **parse_codecs(video.get('codecs', '')),
  29                 })
  30
  31         return {
  32             'id': video_id,
  33             'formats': formats,
  34             'subtitles': subtitles,
  35             'duration': float_or_none(video_info.get('durationMillis'), 1000),
  36             'timestamp': int_or_none(video_info.get('uploadTime'), 1000),
  37             'http_headers': {'Referer': 'https://www.acfun.cn/'},
  38         }
  39
  40
  41 class AcFunVideoIE(AcFunVideoBaseIE):
  42     _VALID_URL = r'https?://www\.acfun\.cn/v/ac(?P<id>[_\d]+)'
  43
  44     _TESTS = [{
  45         'url': 'https://www.acfun.cn/v/ac35457073',
  46         'info_dict': {
  47             'id': '35457073',
  48             'ext': 'mp4',
  49             'duration': 174.208,
  50             'timestamp': 1656403967,
  51             'title': '1 8 岁 现 状',
  52             'description': '“赶紧回去！班主任查班了！”',
  53             'uploader': '锤子game',
  54             'uploader_id': '51246077',
  55             'thumbnail': r're:^https?://.*\.(jpg|jpeg)',
  56             'upload_date': '20220628',
  57             'like_count': int,
  58             'view_count': int,
  59             'comment_count': int,
  60             'tags': list,
  61         },
  62     }, {
  63         # example for len(video_list) > 1
  64         'url': 'https://www.acfun.cn/v/ac35468952_2',
  65         'info_dict': {
  66             'id': '35468952_2',
  67             'ext': 'mp4',
  68             'title': '【动画剧集】Rocket & Groot Season 1（2022）/火箭浣熊与格鲁特第1季 P02 S01E02 十拿九穩',
  69             'duration': 90.459,
  70             'uploader': '比令',
  71             'uploader_id': '37259967',
  72             'upload_date': '20220629',
  73             'timestamp': 1656479962,
  74             'tags': list,
  75             'like_count': int,
  76             'view_count': int,
  77             'comment_count': int,
  78             'thumbnail': r're:^https?://.*\.(jpg|jpeg)',
  79             'description': 'md5:67583aaf3a0f933bd606bc8a2d3ebb17',
  80         },
  81     }]
  82
  83     def _real_extract(self, url):
  84         video_id = self._match_id(url)
  85
  86         webpage = self._download_webpage(url, video_id)
  87         json_all = self._search_json(r'window.videoInfo\s*=', webpage, 'videoInfo', video_id)
  88
  89         title = json_all.get('title')
  90         video_list = json_all.get('videoList') or []
  91         video_internal_id = traverse_obj(json_all, ('currentVideoInfo', 'id'))
  92         if video_internal_id and len(video_list) > 1:
  93             part_idx, part_video_info = next(
  94                 (idx + 1, v) for (idx, v) in enumerate(video_list)
  95                 if v['id'] == video_internal_id)
  96             title = f'{title} P{part_idx:02d} {part_video_info["title"]}'
  97
  98         return {
  99             **self._extract_metadata(video_id, json_all['currentVideoInfo']),
 100             'title': title,
 101             'thumbnail': json_all.get('coverUrl'),
 102             'description': json_all.get('description'),
 103             'uploader': traverse_obj(json_all, ('user', 'name')),
 104             'uploader_id': traverse_obj(json_all, ('user', 'href')),
 105             'tags': traverse_obj(json_all, ('tagList', ..., 'name')),
 106             'view_count': int_or_none(json_all.get('viewCount')),
 107             'like_count': int_or_none(json_all.get('likeCountShow')),
 108             'comment_count': int_or_none(json_all.get('commentCountShow')),
 109         }
 110
 111
 112 class AcFunBangumiIE(AcFunVideoBaseIE):
 113     _VALID_URL = r'https?://www\.acfun\.cn/bangumi/(?P<id>aa[_\d]+)'
 114
 115     _TESTS = [{
 116         'url': 'https://www.acfun.cn/bangumi/aa6002917_36188_1745457?ac=2',
 117         'info_dict': {
 118             'id': 'aa6002917_36188_1745457__2',
 119             'ext': 'mp4',
 120             'title': '【7月】租借女友 水原千鹤角色曲『DATE』特别PV',
 121             'upload_date': '20200916',
 122             'timestamp': 1600243813,
 123             'duration': 92.091,
 124         },
 125     }, {
 126         'url': 'https://www.acfun.cn/bangumi/aa5023171_36188_1750645',
 127         'info_dict': {
 128             'id': 'aa5023171_36188_1750645',
 129             'ext': 'mp4',
 130             'title': '红孩儿之趴趴蛙寻石记 第5话 ',
 131             'duration': 760.0,
 132             'season': '红孩儿之趴趴蛙寻石记',
 133             'season_id': '5023171',
 134             'season_number': 1,  # series has only 1 season
 135             'episode': 'Episode 5',
 136             'episode_number': 5,
 137             'upload_date': '20181223',
 138             'timestamp': 1545552185,
 139             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)',
 140             'comment_count': int,
 141         },
 142     }, {
 143         'url': 'https://www.acfun.cn/bangumi/aa6065485_36188_1885061',
 144         'info_dict': {
 145             'id': 'aa6065485_36188_1885061',
 146             'ext': 'mp4',
 147             'title': '叽歪老表（第二季） 第5话 坚不可摧',
 148             'season': '叽歪老表（第二季）',
 149             'season_number': 2,
 150             'season_id': '6065485',
 151             'episode': '坚不可摧',
 152             'episode_number': 5,
 153             'upload_date': '20220324',
 154             'timestamp': 1648082786,
 155             'duration': 105.002,
 156             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)',
 157             'comment_count': int,
 158         },
 159     }]
 160
 161     def _real_extract(self, url):
 162         video_id = self._match_id(url)
 163         ac_idx = parse_qs(url).get('ac', [None])[-1]
 164         video_id = f'{video_id}{format_field(ac_idx, None, "__%s")}'
 165
 166         webpage = self._download_webpage(url, video_id)
 167         json_bangumi_data = self._search_json(r'window.bangumiData\s*=', webpage, 'bangumiData', video_id)
 168
 169         if ac_idx:
 170             video_info = json_bangumi_data['hlVideoInfo']
 171             return {
 172                 **self._extract_metadata(video_id, video_info),
 173                 'title': video_info.get('title'),
 174             }
 175
 176         video_info = json_bangumi_data['currentVideoInfo']
 177
 178         season_id = json_bangumi_data.get('bangumiId')
 179         season_number = season_id and next((
 180             idx for idx, v in enumerate(json_bangumi_data.get('relatedBangumis') or [], 1)
 181             if v.get('id') == season_id), 1)
 182
 183         json_bangumi_list = self._search_json(
 184             r'window\.bangumiList\s*=', webpage, 'bangumiList', video_id, fatal=False)
 185         video_internal_id = int_or_none(traverse_obj(json_bangumi_data, ('currentVideoInfo', 'id')))
 186         episode_number = video_internal_id and next((
 187             idx for idx, v in enumerate(json_bangumi_list.get('items') or [], 1)
 188             if v.get('videoId') == video_internal_id), None)
 189
 190         return {
 191             **self._extract_metadata(video_id, video_info),
 192             'title': json_bangumi_data.get('showTitle'),
 193             'thumbnail': json_bangumi_data.get('image'),
 194             'season': json_bangumi_data.get('bangumiTitle'),
 195             'season_id': str_or_none(season_id),
 196             'season_number': season_number,
 197             'episode': json_bangumi_data.get('title'),
 198             'episode_number': episode_number,
 199             'comment_count': int_or_none(json_bangumi_data.get('commentCount')),
 200         }