yt_dlp/extractor/kakao.py

   1 # coding: utf-8
   2
   3 from __future__ import unicode_literals
   4
   5 from .common import InfoExtractor
   6 from ..compat import compat_HTTPError
   7 from ..utils import (
   8     ExtractorError,
   9     int_or_none,
  10     strip_or_none,
  11     str_or_none,
  12     traverse_obj,
  13     unified_timestamp,
  14 )
  15
  16
  17 class KakaoIE(InfoExtractor):
  18     _VALID_URL = r'https?://(?:play-)?tv\.kakao\.com/(?:channel/\d+|embed/player)/cliplink/(?P<id>\d+|[^?#&]+@my)'
  19     _API_BASE_TMPL = 'http://tv.kakao.com/api/v1/ft/playmeta/cliplink/%s/'
  20     _CDN_API = 'https://tv.kakao.com/katz/v1/ft/cliplink/%s/readyNplay?'
  21
  22     _TESTS = [{
  23         'url': 'http://tv.kakao.com/channel/2671005/cliplink/301965083',
  24         'md5': '702b2fbdeb51ad82f5c904e8c0766340',
  25         'info_dict': {
  26             'id': '301965083',
  27             'ext': 'mp4',
  28             '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』',
  29             'description': '',
  30             'uploader_id': '2671005',
  31             'uploader': '그랑그랑이',
  32             'timestamp': 1488160199,
  33             'upload_date': '20170227',
  34             'like_count': int,
  35             'thumbnail': r're:http://.+/thumb\.png',
  36             'tags': ['乃木坂'],
  37             'view_count': int,
  38             'duration': 1503,
  39             'comment_count': int,
  40         }
  41     }, {
  42         'url': 'http://tv.kakao.com/channel/2653210/cliplink/300103180',
  43         'md5': 'a8917742069a4dd442516b86e7d66529',
  44         'info_dict': {
  45             'id': '300103180',
  46             'ext': 'mp4',
  47             'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회',
  48             'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)',
  49             'uploader_id': '2653210',
  50             'uploader': '쇼! 음악중심',
  51             'timestamp': 1485684628,
  52             'upload_date': '20170129',
  53             'like_count': int,
  54             'thumbnail': r're:http://.+/thumb\.png',
  55             'tags': 'count:28',
  56             'view_count': int,
  57             'duration': 184,
  58             'comment_count': int,
  59         }
  60     }, {
  61         # geo restricted
  62         'url': 'https://tv.kakao.com/channel/3643855/cliplink/412069491',
  63         'only_matching': True,
  64     }]
  65
  66     def _real_extract(self, url):
  67         video_id = self._match_id(url)
  68         api_base = self._API_BASE_TMPL % video_id
  69         cdn_api_base = self._CDN_API % video_id
  70
  71         query = {
  72             'player': 'monet_html5',
  73             'referer': url,
  74             'uuid': '',
  75             'service': 'kakao_tv',
  76             'section': '',
  77             'dteType': 'PC',
  78             'fields': ','.join([
  79                 '-*', 'tid', 'clipLink', 'displayTitle', 'clip', 'title',
  80                 'description', 'channelId', 'createTime', 'duration', 'playCount',
  81                 'likeCount', 'commentCount', 'tagList', 'channel', 'name',
  82                 'clipChapterThumbnailList', 'thumbnailUrl', 'timeInSec', 'isDefault',
  83                 'videoOutputList', 'width', 'height', 'kbps', 'profile', 'label'])
  84         }
  85
  86         api_json = self._download_json(
  87             api_base, video_id, 'Downloading video info')
  88
  89         clip_link = api_json['clipLink']
  90         clip = clip_link['clip']
  91
  92         title = clip.get('title') or clip_link.get('displayTitle')
  93
  94         formats = []
  95         for fmt in clip.get('videoOutputList') or []:
  96             profile_name = fmt.get('profile')
  97             if not profile_name or profile_name == 'AUDIO':
  98                 continue
  99             query.update({
 100                 'profile': profile_name,
 101                 'fields': '-*,code,message,url',
 102             })
 103             try:
 104                 fmt_url_json = self._download_json(
 105                     cdn_api_base, video_id, query=query,
 106                     note='Downloading video URL for profile %s' % profile_name)
 107             except ExtractorError as e:
 108                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 109                     resp = self._parse_json(e.cause.read().decode(), video_id)
 110                     if resp.get('code') == 'GeoBlocked':
 111                         self.raise_geo_restricted()
 112
 113             fmt_url = traverse_obj(fmt_url_json, ('videoLocation', 'url'))
 114             if not fmt_url:
 115                 continue
 116
 117             formats.append({
 118                 'url': fmt_url,
 119                 'format_id': profile_name,
 120                 'width': int_or_none(fmt.get('width')),
 121                 'height': int_or_none(fmt.get('height')),
 122                 'format_note': fmt.get('label'),
 123                 'filesize': int_or_none(fmt.get('filesize')),
 124                 'tbr': int_or_none(fmt.get('kbps')),
 125             })
 126         self._sort_formats(formats)
 127
 128         thumbs = []
 129         for thumb in clip.get('clipChapterThumbnailList') or []:
 130             thumbs.append({
 131                 'url': thumb.get('thumbnailUrl'),
 132                 'id': str(thumb.get('timeInSec')),
 133                 'preference': -1 if thumb.get('isDefault') else 0
 134             })
 135         top_thumbnail = clip.get('thumbnailUrl')
 136         if top_thumbnail:
 137             thumbs.append({
 138                 'url': top_thumbnail,
 139                 'preference': 10,
 140             })
 141
 142         return {
 143             'id': video_id,
 144             'title': title,
 145             'description': strip_or_none(clip.get('description')),
 146             'uploader': traverse_obj(clip_link, ('channel', 'name')),
 147             'uploader_id': str_or_none(clip_link.get('channelId')),
 148             'thumbnails': thumbs,
 149             'timestamp': unified_timestamp(clip_link.get('createTime')),
 150             'duration': int_or_none(clip.get('duration')),
 151             'view_count': int_or_none(clip.get('playCount')),
 152             'like_count': int_or_none(clip.get('likeCount')),
 153             'comment_count': int_or_none(clip.get('commentCount')),
 154             'formats': formats,
 155             'tags': clip.get('tagList'),
 156         }