Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vlive.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..utils import (
   8     dict_get,
   9     ExtractorError,
  10     float_or_none,
  11     int_or_none,
  12     remove_start,
  13 )
  14 from ..compat import compat_urllib_parse_urlencode
  15
  16
  17 class VLiveIE(InfoExtractor):
  18     IE_NAME = 'vlive'
  19     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
  20     _TESTS = [{
  21         'url': 'http://www.vlive.tv/video/1326',
  22         'md5': 'cc7314812855ce56de70a06a27314983',
  23         'info_dict': {
  24             'id': '1326',
  25             'ext': 'mp4',
  26             'title': "[V LIVE] Girl's Day's Broadcast",
  27             'creator': "Girl's Day",
  28             'view_count': int,
  29         },
  30     }, {
  31         'url': 'http://www.vlive.tv/video/16937',
  32         'info_dict': {
  33             'id': '16937',
  34             'ext': 'mp4',
  35             'title': '[V LIVE] 첸백시 걍방',
  36             'creator': 'EXO',
  37             'view_count': int,
  38             'subtitles': 'mincount:12',
  39         },
  40         'params': {
  41             'skip_download': True,
  42         },
  43     }]
  44
  45     def _real_extract(self, url):
  46         video_id = self._match_id(url)
  47
  48         webpage = self._download_webpage(
  49             'http://www.vlive.tv/video/%s' % video_id, video_id)
  50
  51         video_params = self._search_regex(
  52             r'\bvlive\.video\.init\(([^)]+)\)',
  53             webpage, 'video params')
  54         status, _, _, live_params, long_video_id, key = re.split(
  55             r'"\s*,\s*"', video_params)[2:8]
  56         status = remove_start(status, 'PRODUCT_')
  57
  58         if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR':
  59             live_params = self._parse_json('"%s"' % live_params, video_id)
  60             live_params = self._parse_json(live_params, video_id)
  61             return self._live(video_id, webpage, live_params)
  62         elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO':
  63             if long_video_id and key:
  64                 return self._replay(video_id, webpage, long_video_id, key)
  65             else:
  66                 status = 'COMING_SOON'
  67
  68         if status == 'LIVE_END':
  69             raise ExtractorError('Uploading for replay. Please wait...',
  70                                  expected=True)
  71         elif status == 'COMING_SOON':
  72             raise ExtractorError('Coming soon!', expected=True)
  73         elif status == 'CANCELED':
  74             raise ExtractorError('We are sorry, '
  75                                  'but the live broadcast has been canceled.',
  76                                  expected=True)
  77         else:
  78             raise ExtractorError('Unknown status %s' % status)
  79
  80     def _get_common_fields(self, webpage):
  81         title = self._og_search_title(webpage)
  82         creator = self._html_search_regex(
  83             r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)',
  84             webpage, 'creator', fatal=False)
  85         thumbnail = self._og_search_thumbnail(webpage)
  86         return {
  87             'title': title,
  88             'creator': creator,
  89             'thumbnail': thumbnail,
  90         }
  91
  92     def _live(self, video_id, webpage, live_params):
  93         formats = []
  94         for vid in live_params.get('resolutions', []):
  95             formats.extend(self._extract_m3u8_formats(
  96                 vid['cdnUrl'], video_id, 'mp4',
  97                 m3u8_id=vid.get('name'),
  98                 fatal=False, live=True))
  99         self._sort_formats(formats)
 100
 101         return dict(self._get_common_fields(webpage),
 102                     id=video_id,
 103                     formats=formats,
 104                     is_live=True)
 105
 106     def _replay(self, video_id, webpage, long_video_id, key):
 107         playinfo = self._download_json(
 108             'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s'
 109             % compat_urllib_parse_urlencode({
 110                 'videoId': long_video_id,
 111                 'key': key,
 112                 'ptc': 'http',
 113                 'doct': 'json',  # document type (xml or json)
 114                 'cpt': 'vtt',  # captions type (vtt or ttml)
 115             }), video_id)
 116
 117         formats = [{
 118             'url': vid['source'],
 119             'format_id': vid.get('encodingOption', {}).get('name'),
 120             'abr': float_or_none(vid.get('bitrate', {}).get('audio')),
 121             'vbr': float_or_none(vid.get('bitrate', {}).get('video')),
 122             'width': int_or_none(vid.get('encodingOption', {}).get('width')),
 123             'height': int_or_none(vid.get('encodingOption', {}).get('height')),
 124             'filesize': int_or_none(vid.get('size')),
 125         } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')]
 126         self._sort_formats(formats)
 127
 128         view_count = int_or_none(playinfo.get('meta', {}).get('count'))
 129
 130         subtitles = {}
 131         for caption in playinfo.get('captions', {}).get('list', []):
 132             lang = dict_get(caption, ('locale', 'language', 'country', 'label'))
 133             if lang and caption.get('source'):
 134                 subtitles[lang] = [{
 135                     'ext': 'vtt',
 136                     'url': caption['source']}]
 137
 138         return dict(self._get_common_fields(webpage),
 139                     id=video_id,
 140                     formats=formats,
 141                     view_count=view_count,
 142                     subtitles=subtitles)