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

   1 import re
   2 import json
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     compat_urlparse,
   7     compat_urllib_parse,
   8     determine_ext,
   9     unified_strdate,
  10 )
  11
  12
  13 class NHLBaseInfoExtractor(InfoExtractor):
  14     @staticmethod
  15     def _fix_json(json_string):
  16         return json_string.replace('\\\'', '\'')
  17
  18     def _extract_video(self, info):
  19         video_id = info['id']
  20         self.report_extraction(video_id)
  21
  22         initial_video_url = info['publishPoint']
  23         data = compat_urllib_parse.urlencode({
  24             'type': 'fvod',
  25             'path': initial_video_url.replace('.mp4', '_sd.mp4'),
  26         })
  27         path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  28         path_doc = self._download_xml(path_url, video_id,
  29             u'Downloading final video url')
  30         video_url = path_doc.find('path').text
  31
  32         join = compat_urlparse.urljoin
  33         return {
  34             'id': video_id,
  35             'title': info['name'],
  36             'url': video_url,
  37             'ext': determine_ext(video_url),
  38             'description': info['description'],
  39             'duration': int(info['duration']),
  40             'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
  41             'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
  42         }
  43
  44
  45 class NHLIE(NHLBaseInfoExtractor):
  46     IE_NAME = u'nhl.com'
  47     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console\?.*?(?<=[?&])id=(?P<id>\d+)'
  48
  49     _TEST = {
  50         u'url': u'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  51         u'file': u'453614.mp4',
  52         u'info_dict': {
  53             u'title': u'Quick clip: Weise 4-3 goal vs Flames',
  54             u'description': u'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  55             u'duration': 18,
  56             u'upload_date': u'20131006',
  57         },
  58     }
  59
  60     def _real_extract(self, url):
  61         mobj = re.match(self._VALID_URL, url)
  62         video_id = mobj.group('id')
  63         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
  64         info_json = self._download_webpage(json_url, video_id,
  65             u'Downloading info json')
  66         info_json = self._fix_json(info_json)
  67         info = json.loads(info_json)[0]
  68         return self._extract_video(info)
  69
  70
  71 class NHLVideocenterIE(NHLBaseInfoExtractor):
  72     IE_NAME = u'nhl.com:videocenter'
  73     IE_DESC = u'NHL videocenter category'
  74     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[^&]+))?'
  75
  76     @classmethod
  77     def suitable(cls, url):
  78         if NHLIE.suitable(url):
  79             return False
  80         return super(NHLVideocenterIE, cls).suitable(url)
  81
  82     def _real_extract(self, url):
  83         mobj = re.match(self._VALID_URL, url)
  84         team = mobj.group('team')
  85         webpage = self._download_webpage(url, team)
  86         cat_id = self._search_regex(
  87             [r'var defaultCatId = "(.+?)";',
  88              r'{statusIndex:0,index:0,.*?id:(.*?),'],
  89             webpage, u'category id')
  90         playlist_title = self._html_search_regex(
  91             r'tab0"[^>]*?>(.*?)</td>',
  92             webpage, u'playlist title', flags=re.DOTALL).lower().capitalize()
  93
  94         data = compat_urllib_parse.urlencode({
  95             'cid': cat_id,
  96             # This is the default value
  97             'count': 12,
  98             'ptrs': 3,
  99             'format': 'json',
 100         })
 101         path = '/videocenter/servlets/browse?' + data
 102         request_url = compat_urlparse.urljoin(url, path)
 103         response = self._download_webpage(request_url, playlist_title)
 104         response = self._fix_json(response)
 105         if not response.strip():
 106             self._downloader.report_warning(u'Got an empty reponse, trying '
 107                                             u'adding the "newvideos" parameter')
 108             response = self._download_webpage(request_url + '&newvideos=true',
 109                 playlist_title)
 110             response = self._fix_json(response)
 111         videos = json.loads(response)
 112
 113         return {
 114             '_type': 'playlist',
 115             'title': playlist_title,
 116             'id': cat_id,
 117             'entries': [self._extract_video(i) for i in videos],
 118         }