_ Git - youtube-dl/blob - 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_str
   7 from ..utils import (
   8     int_or_none,
   9     strip_or_none,
  10     unified_timestamp,
  11     update_url_query,
  12 )
  13
  14
  15 class KakaoIE(InfoExtractor):
  16     _VALID_URL = r'https?://(?:play-)?tv\.kakao\.com/(?:channel/\d+|embed/player)/cliplink/(?P<id>\d+|[^?#&]+@my)'
  17     _API_BASE_TMPL = 'http://tv.kakao.com/api/v1/ft/cliplinks/%s/'
  18
  19     _TESTS = [{
  20         'url': 'http://tv.kakao.com/channel/2671005/cliplink/301965083',
  21         'md5': '702b2fbdeb51ad82f5c904e8c0766340',
  22         'info_dict': {
  23             'id': '301965083',
  24             'ext': 'mp4',
  25             '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』',
  26             'uploader_id': 2671005,
  27             'uploader': '그랑그랑이',
  28             'timestamp': 1488160199,
  29             'upload_date': '20170227',
  30         }
  31     }, {
  32         'url': 'http://tv.kakao.com/channel/2653210/cliplink/300103180',
  33         'md5': 'a8917742069a4dd442516b86e7d66529',
  34         'info_dict': {
  35             'id': '300103180',
  36             'ext': 'mp4',
  37             'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회',
  38             'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)',
  39             'uploader_id': 2653210,
  40             'uploader': '쇼! 음악중심',
  41             'timestamp': 1485684628,
  42             'upload_date': '20170129',
  43         }
  44     }]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48         display_id = video_id.rstrip('@my')
  49         api_base = self._API_BASE_TMPL % video_id
  50
  51         player_header = {
  52             'Referer': update_url_query(
  53                 'http://tv.kakao.com/embed/player/cliplink/%s' % video_id, {
  54                     'service': 'kakao_tv',
  55                     'autoplay': '1',
  56                     'profile': 'HIGH',
  57                     'wmode': 'transparent',
  58                 })
  59         }
  60
  61         query = {
  62             'player': 'monet_html5',
  63             'referer': url,
  64             'uuid': '',
  65             'service': 'kakao_tv',
  66             'section': '',
  67             'dteType': 'PC',
  68             'fields': ','.join([
  69                 '-*', 'tid', 'clipLink', 'displayTitle', 'clip', 'title',
  70                 'description', 'channelId', 'createTime', 'duration', 'playCount',
  71                 'likeCount', 'commentCount', 'tagList', 'channel', 'name',
  72                 'clipChapterThumbnailList', 'thumbnailUrl', 'timeInSec', 'isDefault'])
  73         }
  74
  75         impress = self._download_json(
  76             api_base + 'impress', display_id, 'Downloading video info',
  77             query=query, headers=player_header)
  78
  79         clip_link = impress['clipLink']
  80         clip = clip_link['clip']
  81
  82         title = clip.get('title') or clip_link.get('displayTitle')
  83
  84         tid = impress.get('tid', '')
  85
  86         query.update({
  87             'fields': '-*,outputList,profile,width,height,label,filesize',
  88             'tid': tid,
  89             'profile': 'HIGH',
  90         })
  91         raw = self._download_json(
  92             api_base + 'raw', display_id, 'Downloading video formats info',
  93             query=query, headers=player_header)
  94
  95         formats = []
  96         for fmt in raw.get('outputList', []):
  97             try:
  98                 profile_name = fmt['profile']
  99                 query.update({
 100                     'profile': profile_name,
 101                     'fields': '-*,url',
 102                 })
 103                 fmt_url_json = self._download_json(
 104                     api_base + 'raw/videolocation', display_id,
 105                     'Downloading video URL for profile %s' % profile_name,
 106                     query=query, headers=player_header, fatal=False)
 107
 108                 if fmt_url_json is None:
 109                     continue
 110
 111                 fmt_url = fmt_url_json['url']
 112                 formats.append({
 113                     'url': fmt_url,
 114                     'format_id': profile_name,
 115                     'width': int_or_none(fmt.get('width')),
 116                     'height': int_or_none(fmt.get('height')),
 117                     'format_note': fmt.get('label'),
 118                     'filesize': int_or_none(fmt.get('filesize'))
 119                 })
 120             except KeyError:
 121                 pass
 122         self._sort_formats(formats)
 123
 124         thumbs = []
 125         for thumb in clip.get('clipChapterThumbnailList', []):
 126             thumbs.append({
 127                 'url': thumb.get('thumbnailUrl'),
 128                 'id': compat_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': display_id,
 140             'title': title,
 141             'description': strip_or_none(clip.get('description')),
 142             'uploader': clip_link.get('channel', {}).get('name'),
 143             'uploader_id': 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         }