youtube_dl/extractor/njpwworld.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5
   6 from .common import InfoExtractor
   7 from ..compat import compat_urlparse
   8 from ..utils import (
   9     extract_attributes,
  10     get_element_by_class,
  11     urlencode_postdata,
  12 )
  13
  14
  15 class NJPWWorldIE(InfoExtractor):
  16     _VALID_URL = r'https?://njpwworld\.com/p/(?P<id>[a-z0-9_]+)'
  17     IE_DESC = '新日本プロレスワールド'
  18     _NETRC_MACHINE = 'njpwworld'
  19
  20     _TEST = {
  21         'url': 'http://njpwworld.com/p/s_series_00155_1_9/',
  22         'info_dict': {
  23             'id': 's_series_00155_1_9',
  24             'ext': 'mp4',
  25             'title': '第9試合　ランディ・サベージ　vs　リック・スタイナー',
  26             'tags': list,
  27         },
  28         'params': {
  29             'skip_download': True,  # AES-encrypted m3u8
  30         },
  31         'skip': 'Requires login',
  32     }
  33
  34     _LOGIN_URL = 'https://front.njpwworld.com/auth/login'
  35
  36     def _real_initialize(self):
  37         self._login()
  38
  39     def _login(self):
  40         username, password = self._get_login_info()
  41         # No authentication to be performed
  42         if not username:
  43             return True
  44
  45         # Setup session (will set necessary cookies)
  46         self._request_webpage(
  47             'https://njpwworld.com/', None, note='Setting up session')
  48
  49         webpage, urlh = self._download_webpage_handle(
  50             self._LOGIN_URL, None,
  51             note='Logging in', errnote='Unable to login',
  52             data=urlencode_postdata({'login_id': username, 'pw': password}),
  53             headers={'Referer': 'https://front.njpwworld.com/auth'})
  54         # /auth/login will return 302 for successful logins
  55         if urlh.geturl() == self._LOGIN_URL:
  56             self.report_warning('unable to login')
  57             return False
  58
  59         return True
  60
  61     def _real_extract(self, url):
  62         video_id = self._match_id(url)
  63
  64         webpage = self._download_webpage(url, video_id)
  65
  66         formats = []
  67         for mobj in re.finditer(r'<a[^>]+\bhref=(["\'])/player.+?[^>]*>', webpage):
  68             player = extract_attributes(mobj.group(0))
  69             player_path = player.get('href')
  70             if not player_path:
  71                 continue
  72             kind = self._search_regex(
  73                 r'(low|high)$', player.get('class') or '', 'kind',
  74                 default='low')
  75             player_url = compat_urlparse.urljoin(url, player_path)
  76             player_page = self._download_webpage(
  77                 player_url, video_id, note='Downloading player page')
  78             entries = self._parse_html5_media_entries(
  79                 player_url, player_page, video_id, m3u8_id='hls-%s' % kind,
  80                 m3u8_entry_protocol='m3u8_native')
  81             kind_formats = entries[0]['formats']
  82             for f in kind_formats:
  83                 f['quality'] = 2 if kind == 'high' else 1
  84             formats.extend(kind_formats)
  85
  86         self._sort_formats(formats)
  87
  88         post_content = get_element_by_class('post-content', webpage)
  89         tags = re.findall(
  90             r'<li[^>]+class="tag-[^"]+"><a[^>]*>([^<]+)</a></li>', post_content
  91         ) if post_content else None
  92
  93         return {
  94             'id': video_id,
  95             'title': self._og_search_title(webpage),
  96             'formats': formats,
  97             'tags': tags,
  98         }