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     urlencode_postdata,
  14 )
  15 from ..compat import compat_urllib_parse_urlencode
  16
  17
  18 class VLiveIE(InfoExtractor):
  19     IE_NAME = 'vlive'
  20     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
  21     _TESTS = [{
  22         'url': 'http://www.vlive.tv/video/1326',
  23         'md5': 'cc7314812855ce56de70a06a27314983',
  24         'info_dict': {
  25             'id': '1326',
  26             'ext': 'mp4',
  27             'title': "[V LIVE] Girl's Day's Broadcast",
  28             'creator': "Girl's Day",
  29             'view_count': int,
  30         },
  31     }, {
  32         'url': 'http://www.vlive.tv/video/16937',
  33         'info_dict': {
  34             'id': '16937',
  35             'ext': 'mp4',
  36             'title': '[V LIVE] 첸백시 걍방',
  37             'creator': 'EXO',
  38             'view_count': int,
  39             'subtitles': 'mincount:12',
  40         },
  41         'params': {
  42             'skip_download': True,
  43         },
  44     }]
  45
  46     def _real_extract(self, url):
  47         video_id = self._match_id(url)
  48
  49         webpage = self._download_webpage(
  50             'http://www.vlive.tv/video/%s' % video_id, video_id)
  51
  52         VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)'
  53         VIDEO_PARAMS_FIELD = 'video params'
  54
  55         params = self._parse_json(self._search_regex(
  56             VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD, default=''), video_id,
  57             transform_source=lambda s: '[' + s + ']', fatal=False)
  58
  59         if not params or len(params) < 7:
  60             params = self._search_regex(
  61                 VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD)
  62             params = [p.strip(r'"') for p in re.split(r'\s*,\s*', params)]
  63
  64         status, long_video_id, key = params[2], params[5], params[6]
  65         status = remove_start(status, 'PRODUCT_')
  66
  67         if status == 'LIVE_ON_AIR' or status == 'BIG_EVENT_ON_AIR':
  68             return self._live(video_id, webpage)
  69         elif status == 'VOD_ON_AIR' or status == 'BIG_EVENT_INTRO':
  70             if long_video_id and key:
  71                 return self._replay(video_id, webpage, long_video_id, key)
  72             else:
  73                 status = 'COMING_SOON'
  74
  75         if status == 'LIVE_END':
  76             raise ExtractorError('Uploading for replay. Please wait...',
  77                                  expected=True)
  78         elif status == 'COMING_SOON':
  79             raise ExtractorError('Coming soon!', expected=True)
  80         elif status == 'CANCELED':
  81             raise ExtractorError('We are sorry, '
  82                                  'but the live broadcast has been canceled.',
  83                                  expected=True)
  84         else:
  85             raise ExtractorError('Unknown status %s' % status)
  86
  87     def _get_common_fields(self, webpage):
  88         title = self._og_search_title(webpage)
  89         creator = self._html_search_regex(
  90             r'<div[^>]+class="info_area"[^>]*>\s*<a\s+[^>]*>([^<]+)',
  91             webpage, 'creator', fatal=False)
  92         thumbnail = self._og_search_thumbnail(webpage)
  93         return {
  94             'title': title,
  95             'creator': creator,
  96             'thumbnail': thumbnail,
  97         }
  98
  99     def _live(self, video_id, webpage):
 100         init_page = self._download_webpage(
 101             'http://www.vlive.tv/video/init/view',
 102             video_id, note='Downloading live webpage',
 103             data=urlencode_postdata({'videoSeq': video_id}),
 104             headers={
 105                 'Referer': 'http://www.vlive.tv/video/%s' % video_id,
 106                 'Content-Type': 'application/x-www-form-urlencoded'
 107             })
 108
 109         live_params = self._search_regex(
 110             r'"liveStreamInfo"\s*:\s*(".*"),',
 111             init_page, 'live stream info')
 112         live_params = self._parse_json(live_params, video_id)
 113         live_params = self._parse_json(live_params, video_id)
 114
 115         formats = []
 116         for vid in live_params.get('resolutions', []):
 117             formats.extend(self._extract_m3u8_formats(
 118                 vid['cdnUrl'], video_id, 'mp4',
 119                 m3u8_id=vid.get('name'),
 120                 fatal=False, live=True))
 121         self._sort_formats(formats)
 122
 123         info = self._get_common_fields(webpage)
 124         info.update({
 125             'title': self._live_title(info['title']),
 126             'id': video_id,
 127             'formats': formats,
 128             'is_live': True,
 129         })
 130         return info
 131
 132     def _replay(self, video_id, webpage, long_video_id, key):
 133         playinfo = self._download_json(
 134             'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s'
 135             % compat_urllib_parse_urlencode({
 136                 'videoId': long_video_id,
 137                 'key': key,
 138                 'ptc': 'http',
 139                 'doct': 'json',  # document type (xml or json)
 140                 'cpt': 'vtt',  # captions type (vtt or ttml)
 141             }), video_id)
 142
 143         formats = [{
 144             'url': vid['source'],
 145             'format_id': vid.get('encodingOption', {}).get('name'),
 146             'abr': float_or_none(vid.get('bitrate', {}).get('audio')),
 147             'vbr': float_or_none(vid.get('bitrate', {}).get('video')),
 148             'width': int_or_none(vid.get('encodingOption', {}).get('width')),
 149             'height': int_or_none(vid.get('encodingOption', {}).get('height')),
 150             'filesize': int_or_none(vid.get('size')),
 151         } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')]
 152         self._sort_formats(formats)
 153
 154         view_count = int_or_none(playinfo.get('meta', {}).get('count'))
 155
 156         subtitles = {}
 157         for caption in playinfo.get('captions', {}).get('list', []):
 158             lang = dict_get(caption, ('locale', 'language', 'country', 'label'))
 159             if lang and caption.get('source'):
 160                 subtitles[lang] = [{
 161                     'ext': 'vtt',
 162                     'url': caption['source']}]
 163
 164         info = self._get_common_fields(webpage)
 165         info.update({
 166             'id': video_id,
 167             'formats': formats,
 168             'view_count': view_count,
 169             'subtitles': subtitles,
 170         })
 171         return info