youtube_dl/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     str_or_none,
  11     strip_or_none,
  12     try_get,
  13     unified_timestamp,
  14     update_url_query,
  15 )
  16
  17
  18 class KakaoIE(InfoExtractor):
  19     _VALID_URL = r'https?://(?:play-)?tv\.kakao\.com/(?:channel/\d+|embed/player)/cliplink/(?P<id>\d+|[^?#&]+@my)'
  20     _API_BASE_TMPL = 'http://tv.kakao.com/api/v1/ft/cliplinks/%s/'
  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             'uploader_id': '2671005',
  30             'uploader': '그랑그랑이',
  31             'timestamp': 1488160199,
  32             'upload_date': '20170227',
  33         }
  34     }, {
  35         'url': 'http://tv.kakao.com/channel/2653210/cliplink/300103180',
  36         'md5': 'a8917742069a4dd442516b86e7d66529',
  37         'info_dict': {
  38             'id': '300103180',
  39             'ext': 'mp4',
  40             'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회',
  41             'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)',
  42             'uploader_id': '2653210',
  43             'uploader': '쇼! 음악중심',
  44             'timestamp': 1485684628,
  45             'upload_date': '20170129',
  46         }
  47     }, {
  48         # geo restricted
  49         'url': 'https://tv.kakao.com/channel/3643855/cliplink/412069491',
  50         'only_matching': True,
  51     }]
  52
  53     def _real_extract(self, url):
  54         video_id = self._match_id(url)
  55         display_id = video_id.rstrip('@my')
  56         api_base = self._API_BASE_TMPL % video_id
  57
  58         player_header = {
  59             'Referer': update_url_query(
  60                 'http://tv.kakao.com/embed/player/cliplink/%s' % video_id, {
  61                     'service': 'kakao_tv',
  62                     'autoplay': '1',
  63                     'profile': 'HIGH',
  64                     'wmode': 'transparent',
  65                 })
  66         }
  67
  68         query = {
  69             'player': 'monet_html5',
  70             'referer': url,
  71             'uuid': '',
  72             'service': 'kakao_tv',
  73             'section': '',
  74             'dteType': 'PC',
  75             'fields': ','.join([
  76                 '-*', 'tid', 'clipLink', 'displayTitle', 'clip', 'title',
  77                 'description', 'channelId', 'createTime', 'duration', 'playCount',
  78                 'likeCount', 'commentCount', 'tagList', 'channel', 'name', 'thumbnailUrl',
  79                 'videoOutputList', 'width', 'height', 'kbps', 'profile', 'label'])
  80         }
  81
  82         impress = self._download_json(
  83             api_base + 'impress', display_id, 'Downloading video info',
  84             query=query, headers=player_header)
  85
  86         clip_link = impress['clipLink']
  87         clip = clip_link['clip']
  88
  89         title = clip.get('title') or clip_link.get('displayTitle')
  90
  91         query.update({
  92             'fields': '-*,code,message,url',
  93             'tid': impress.get('tid') or '',
  94         })
  95
  96         formats = []
  97         for fmt in (clip.get('videoOutputList') or []):
  98             try:
  99                 profile_name = fmt['profile']
 100                 if profile_name == 'AUDIO':
 101                     continue
 102                 query['profile'] = profile_name
 103                 try:
 104                     fmt_url_json = self._download_json(
 105                         api_base + 'raw/videolocation', display_id,
 106                         'Downloading video URL for profile %s' % profile_name,
 107                         query=query, headers=player_header)
 108                 except ExtractorError as e:
 109                     if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
 110                         resp = self._parse_json(e.cause.read().decode(), video_id)
 111                         if resp.get('code') == 'GeoBlocked':
 112                             self.raise_geo_restricted()
 113                     continue
 114
 115                 fmt_url = fmt_url_json['url']
 116                 formats.append({
 117                     'url': fmt_url,
 118                     'format_id': profile_name,
 119                     'width': int_or_none(fmt.get('width')),
 120                     'height': int_or_none(fmt.get('height')),
 121                     'format_note': fmt.get('label'),
 122                     'filesize': int_or_none(fmt.get('filesize')),
 123                     'tbr': int_or_none(fmt.get('kbps')),
 124                 })
 125             except KeyError:
 126                 pass
 127         self._sort_formats(formats)
 128
 129         return {
 130             'id': display_id,
 131             'title': title,
 132             'description': strip_or_none(clip.get('description')),
 133             'uploader': try_get(clip_link, lambda x: x['channel']['name']),
 134             'uploader_id': str_or_none(clip_link.get('channelId')),
 135             'thumbnail': clip.get('thumbnailUrl'),
 136             'timestamp': unified_timestamp(clip_link.get('createTime')),
 137             'duration': int_or_none(clip.get('duration')),
 138             'view_count': int_or_none(clip.get('playCount')),
 139             'like_count': int_or_none(clip.get('likeCount')),
 140             'comment_count': int_or_none(clip.get('commentCount')),
 141             'formats': formats,
 142             'tags': clip.get('tagList'),
 143         }