_ 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     int_or_none
  14 )
  15 from ..compat import compat_urllib_parse
  16
  17
  18 class VLiveIE(InfoExtractor):
  19     IE_NAME = 'vlive'
  20     # vlive.tv/video/ links redirect to www.vlive.tv/video/
  21     _VALID_URL = r'https?://(?:(www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
  22     _TEST = {
  23         'url': 'http://www.vlive.tv/video/1326',
  24         'md5': 'cc7314812855ce56de70a06a27314983',
  25         'info_dict': {
  26             'id': '1326',
  27             'ext': 'mp4',
  28             'title': '[V] Girl\'s Day\'s Broadcast',
  29             'creator': 'Girl\'s Day',
  30         },
  31     }
  32
  33     def _real_extract(self, url):
  34         video_id = self._match_id(url)
  35
  36         webpage = self._download_webpage(
  37             'http://www.vlive.tv/video/%s' % video_id,
  38             video_id, note='Download video page')
  39
  40         long_video_id = self._search_regex(
  41             r'vlive\.tv\.video\.ajax\.request\.handler\.init\("[0-9]+",\s?"[^"]*",\s?"([^"]+)",\s?"[^"]+",\s?"[^"]*",\s?"[^"]*"\)', webpage, 'long_video_id')
  42
  43         key = self._search_regex(
  44             r'vlive\.tv\.video\.ajax\.request\.handler\.init\("[0-9]+",\s?"[^"]*",\s?"[^"]+",\s?"([^"]+)",\s?"[^"]*",\s?"[^"]*"\)', webpage, 'key')
  45
  46         title = self._og_search_title(webpage)
  47         thumbnail = self._og_search_thumbnail(webpage)
  48         creator = self._html_search_regex(
  49             r'<div class="info_area">\s*<strong[^>]+class="name">([^<>]+)</strong>', webpage, 'creator',fatal=False)
  50
  51         # doct = document type (xml or json), cpt = caption type (vtt or ttml)
  52         url = "http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?videoId=%s&key=%s&ptc=http&doct=json&cpt=vtt" % (long_video_id, key)
  53
  54         playinfo = self._download_json(url, video_id, 'Downloading video json')
  55
  56         formats = []
  57         for vid in playinfo.get('videos', {}).get('list', []):
  58             formats.append({
  59                 'url': vid['source'],
  60                 'ext': 'mp4',
  61                 'abr': vid.get('bitrate', {}).get('audio'),
  62                 'vbr': vid.get('bitrate', {}).get('video'),
  63                 'format_id': vid.get('encodingOption', {}).get('name'),
  64                 'height': int_or_none(vid.get('encodingOption', {}).get('height')),
  65                 'width': int_or_none(vid.get('encodingOption', {}).get('width')),
  66             })
  67         self._sort_formats(formats)
  68
  69         subtitles = {}
  70         for caption in playinfo.get('captions', {}).get('list', []):
  71             subtitles[caption['language']] = [
  72                 {'ext': determine_ext(caption['source'], default_ext='vtt'),
  73                  'url': caption['source']}]
  74
  75         return {
  76             'id': video_id,
  77             'title': title,
  78             'creator': creator,
  79             'thumbnail': thumbnail,
  80             'formats': formats,
  81             'subtitles': subtitles,
  82         }