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