_ Git - youtube-dl/blob - youtube_dl/extractor/vlive.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import hmac
   5 from hashlib import sha1
   6 from base64 import b64encode
   7 from time import time
   8
   9 from .common import InfoExtractor
  10 from ..utils import (
  11     ExtractorError,
  12     determine_ext
  13 )
  14 from ..compat import compat_urllib_parse
  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     _TEST = {
  21         'url': 'http://m.vlive.tv/video/1326',
  22         'md5': 'cc7314812855ce56de70a06a27314983',
  23         'info_dict': {
  24             'id': '1326',
  25             'ext': 'mp4',
  26             'title': '[V] Girl\'s Day\'s Broadcast',
  27             'creator': 'Girl\'s Day',
  28             'upload_date': '20150817',
  29         },
  30     }
  31     _SECRET = 'rFkwZet6pqk1vQt6SxxUkAHX7YL3lmqzUMrU4IDusTo4jEBdtOhNfT4BYYAdArwH'
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         webpage = self._download_webpage(
  37             'http://m.vlive.tv/video/%s' % video_id,
  38             video_id, note='Download video page')
  39
  40         title = self._og_search_title(webpage)
  41         thumbnail = self._og_search_thumbnail(webpage)
  42         creator = self._html_search_regex(
  43             r'<span class="name">([^<>]+)</span>', webpage, 'creator')
  44         upload_date = self._html_search_regex(
  45             r'<span class="time">(\d{4}\.\d{2}\.\d{2})</span>', webpage,
  46             'upload date', default=None, fatal=False)
  47         if upload_date:
  48             upload_date = upload_date.replace('.', '')
  49
  50         url = 'http://global.apis.naver.com/globalV/globalV/vod/%s/playinfo?' % video_id
  51         msgpad = {'msgpad': '%.0f' % (time() * 1000)}
  52         md = {
  53             'md': b64encode(
  54                 hmac.new(self._SECRET.encode('ascii'),
  55                          (url[:255] + msgpad['msgpad']).encode('ascii'), sha1).digest())
  56         }
  57         url += '&' + compat_urllib_parse.urlencode(msgpad) + '&' + compat_urllib_parse.urlencode(md)
  58
  59         playinfo = self._download_json(url, video_id, 'Downloading video json')
  60
  61         if playinfo.get('message', '') != 'success':
  62             raise ExtractorError(playinfo['message'])
  63
  64         if not playinfo.get('result'):
  65             raise ExtractorError('No videos found.')
  66
  67         formats = []
  68         for vid in playinfo['result'].get('videos', {}).get('list', []):
  69             formats.append({
  70                 'url': vid['source'],
  71                 'ext': 'mp4',
  72                 'abr': vid.get('bitrate', {}).get('audio'),
  73                 'vbr': vid.get('bitrate', {}).get('video'),
  74                 'format_id': vid['encodingOption']['name'],
  75                 'height': vid.get('height'),
  76                 'width': vid.get('width'),
  77             })
  78         self._sort_formats(formats)
  79
  80         subtitles = {}
  81         for caption in playinfo['result'].get('captions', {}).get('list', []):
  82             subtitles[caption['language']] = [
  83                 {'ext': determine_ext(caption['source'], default_ext='vtt'),
  84                  'url': caption['source']}]
  85
  86         return {
  87             'id': video_id,
  88             'title': title,
  89             'creator': creator,
  90             'thumbnail': thumbnail,
  91             'formats': formats,
  92             'upload_date': upload_date,
  93             'subtitles': subtitles,
  94         }