yt_dlp/extractor/kakao.py

   1 from .common import InfoExtractor
   2 from ..networking.exceptions import HTTPError
   3 from ..utils import (
   4     ExtractorError,
   5     int_or_none,
   6     strip_or_none,
   7     str_or_none,
   8     traverse_obj,
   9     unified_timestamp,
  10 )
  11
  12
  13 class KakaoIE(InfoExtractor):
  14     _VALID_URL = r'https?://(?:play-)?tv\.kakao\.com/(?:channel/\d+|embed/player)/cliplink/(?P<id>\d+|[^?#&]+@my)'
  15     _API_BASE_TMPL = 'http://tv.kakao.com/api/v1/ft/playmeta/cliplink/%s/'
  16     _CDN_API = 'https://tv.kakao.com/katz/v1/ft/cliplink/%s/readyNplay?'
  17
  18     _TESTS = [{
  19         'url': 'http://tv.kakao.com/channel/2671005/cliplink/301965083',
  20         'md5': '702b2fbdeb51ad82f5c904e8c0766340',
  21         'info_dict': {
  22             'id': '301965083',
  23             'ext': 'mp4',
  24             'title': '乃木坂46 バナナマン 「3期生紹介コーナーが始動！顔高低差GPも！」 『乃木坂工事中』',
  25             'description': '',
  26             'uploader_id': '2671005',
  27             'uploader': '그랑그랑이',
  28             'timestamp': 1488160199,
  29             'upload_date': '20170227',
  30             'like_count': int,
  31             'thumbnail': r're:http://.+/thumb\.png',
  32             'tags': ['乃木坂'],
  33             'view_count': int,
  34             'duration': 1503,
  35             'comment_count': int,
  36         }
  37     }, {
  38         'url': 'http://tv.kakao.com/channel/2653210/cliplink/300103180',
  39         'md5': 'a8917742069a4dd442516b86e7d66529',
  40         'info_dict': {
  41             'id': '300103180',
  42             'ext': 'mp4',
  43             'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회',
  44             'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)',
  45             'uploader_id': '2653210',
  46             'uploader': '쇼! 음악중심',
  47             'timestamp': 1485684628,
  48             'upload_date': '20170129',
  49             'like_count': int,
  50             'thumbnail': r're:http://.+/thumb\.png',
  51             'tags': 'count:28',
  52             'view_count': int,
  53             'duration': 184,
  54             'comment_count': int,
  55         }
  56     }, {
  57         # geo restricted
  58         'url': 'https://tv.kakao.com/channel/3643855/cliplink/412069491',
  59         'only_matching': True,
  60     }]
  61
  62     def _real_extract(self, url):
  63         video_id = self._match_id(url)
  64         api_base = self._API_BASE_TMPL % video_id
  65         cdn_api_base = self._CDN_API % video_id
  66
  67         query = {
  68             'player': 'monet_html5',
  69             'referer': url,
  70             'uuid': '',
  71             'service': 'kakao_tv',
  72             'section': '',
  73             'dteType': 'PC',
  74             'fields': ','.join([
  75                 '-*', 'tid', 'clipLink', 'displayTitle', 'clip', 'title',
  76                 'description', 'channelId', 'createTime', 'duration', 'playCount',
  77                 'likeCount', 'commentCount', 'tagList', 'channel', 'name',
  78                 'clipChapterThumbnailList', 'thumbnailUrl', 'timeInSec', 'isDefault',
  79                 'videoOutputList', 'width', 'height', 'kbps', 'profile', 'label'])
  80         }
  81
  82         api_json = self._download_json(
  83             api_base, video_id, 'Downloading video info')
  84
  85         clip_link = api_json['clipLink']
  86         clip = clip_link['clip']
  87
  88         title = clip.get('title') or clip_link.get('displayTitle')
  89
  90         formats = []
  91         for fmt in clip.get('videoOutputList') or []:
  92             profile_name = fmt.get('profile')
  93             if not profile_name or profile_name == 'AUDIO':
  94                 continue
  95             query.update({
  96                 'profile': profile_name,
  97                 'fields': '-*,code,message,url',
  98             })
  99             try:
 100                 fmt_url_json = self._download_json(
 101                     cdn_api_base, video_id, query=query,
 102                     note='Downloading video URL for profile %s' % profile_name)
 103             except ExtractorError as e:
 104                 if isinstance(e.cause, HTTPError) and e.cause.status == 403:
 105                     resp = self._parse_json(e.cause.response.read().decode(), video_id)
 106                     if resp.get('code') == 'GeoBlocked':
 107                         self.raise_geo_restricted()
 108                 raise
 109
 110             fmt_url = traverse_obj(fmt_url_json, ('videoLocation', 'url'))
 111             if not fmt_url:
 112                 continue
 113
 114             formats.append({
 115                 'url': fmt_url,
 116                 'format_id': profile_name,
 117                 'width': int_or_none(fmt.get('width')),
 118                 'height': int_or_none(fmt.get('height')),
 119                 'format_note': fmt.get('label'),
 120                 'filesize': int_or_none(fmt.get('filesize')),
 121                 'tbr': int_or_none(fmt.get('kbps')),
 122             })
 123
 124         thumbs = []
 125         for thumb in clip.get('clipChapterThumbnailList') or []:
 126             thumbs.append({
 127                 'url': thumb.get('thumbnailUrl'),
 128                 'id': str(thumb.get('timeInSec')),
 129                 'preference': -1 if thumb.get('isDefault') else 0
 130             })
 131         top_thumbnail = clip.get('thumbnailUrl')
 132         if top_thumbnail:
 133             thumbs.append({
 134                 'url': top_thumbnail,
 135                 'preference': 10,
 136             })
 137
 138         return {
 139             'id': video_id,
 140             'title': title,
 141             'description': strip_or_none(clip.get('description')),
 142             'uploader': traverse_obj(clip_link, ('channel', 'name')),
 143             'uploader_id': str_or_none(clip_link.get('channelId')),
 144             'thumbnails': thumbs,
 145             'timestamp': unified_timestamp(clip_link.get('createTime')),
 146             'duration': int_or_none(clip.get('duration')),
 147             'view_count': int_or_none(clip.get('playCount')),
 148             'like_count': int_or_none(clip.get('likeCount')),
 149             'comment_count': int_or_none(clip.get('commentCount')),
 150             'formats': formats,
 151             'tags': clip.get('tagList'),
 152         }