youtube_dl/extractor/nhl.py

   1 import re
   2 import json
   3 import xml.etree.ElementTree
   4
   5 from .common import InfoExtractor
   6 from ..utils import (
   7     compat_urlparse,
   8     compat_urllib_parse,
   9     determine_ext,
  10     unified_strdate,
  11 )
  12
  13
  14 class NHLBaseInfoExtractor(InfoExtractor):
  15     @staticmethod
  16     def _fix_json(json_string):
  17         return json_string.replace('\\\'', '\'')
  18
  19     def _extract_video(self, info):
  20         video_id = info['id']
  21         self.report_extraction(video_id)
  22
  23         initial_video_url = info['publishPoint']
  24         data = compat_urllib_parse.urlencode({
  25             'type': 'fvod',
  26             'path': initial_video_url.replace('.mp4', '_sd.mp4'),
  27         })
  28         path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  29         path_response = self._download_webpage(path_url, video_id,
  30             u'Downloading final video url')
  31         path_doc = xml.etree.ElementTree.fromstring(path_response)
  32         video_url = path_doc.find('path').text
  33
  34         join = compat_urlparse.urljoin
  35         return {
  36             'id': video_id,
  37             'title': info['name'],
  38             'url': video_url,
  39             'ext': determine_ext(video_url),
  40             'description': info['description'],
  41             'duration': int(info['duration']),
  42             'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
  43             'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
  44         }
  45
  46
  47 class NHLIE(NHLBaseInfoExtractor):
  48     IE_NAME = u'nhl.com'
  49     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console\?.*?(?<=[?&])id=(?P<id>\d+)'
  50
  51     _TEST = {
  52         u'url': u'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  53         u'file': u'453614.mp4',
  54         u'info_dict': {
  55             u'title': u'Quick clip: Weise 4-3 goal vs Flames',
  56             u'description': u'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  57             u'duration': 18,
  58             u'upload_date': u'20131006',
  59         },
  60     }
  61
  62     def _real_extract(self, url):
  63         mobj = re.match(self._VALID_URL, url)
  64         video_id = mobj.group('id')
  65         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
  66         info_json = self._download_webpage(json_url, video_id,
  67             u'Downloading info json')
  68         info_json = self._fix_json(info_json)
  69         info = json.loads(info_json)[0]
  70         return self._extract_video(info)
  71
  72
  73 class NHLVideocenterIE(NHLBaseInfoExtractor):
  74     IE_NAME = u'nhl.com:videocenter'
  75     IE_DESC = u'Download the first 12 videos from a videocenter category'
  76     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[^&]+))?'
  77
  78     @classmethod
  79     def suitable(cls, url):
  80         if NHLIE.suitable(url):
  81             return False
  82         return super(NHLVideocenterIE, cls).suitable(url)
  83
  84     def _real_extract(self, url):
  85         mobj = re.match(self._VALID_URL, url)
  86         team = mobj.group('team')
  87         webpage = self._download_webpage(url, team)
  88         cat_id = self._search_regex(
  89             [r'var defaultCatId = "(.+?)";',
  90              r'{statusIndex:0,index:0,.*?id:(.*?),'],
  91             webpage, u'category id')
  92         playlist_title = self._html_search_regex(
  93             r'tab0"[^>]*?>(.*?)</td>',
  94             webpage, u'playlist title', flags=re.DOTALL).lower().capitalize()
  95
  96         data = compat_urllib_parse.urlencode({
  97             'cid': cat_id,
  98             # This is the default value
  99             'count': 12,
 100             'ptrs': 3,
 101             'format': 'json',
 102         })
 103         path = '/videocenter/servlets/browse?' + data
 104         request_url = compat_urlparse.urljoin(url, path)
 105         response = self._download_webpage(request_url, playlist_title)
 106         response = self._fix_json(response)
 107         if not response.strip():
 108             self._downloader.report_warning(u'Got an empty reponse, trying '
 109                                             u'adding the "newvideos" parameter')
 110             response = self._download_webpage(request_url + '&newvideos=true',
 111                 playlist_title)
 112             response = self._fix_json(response)
 113         videos = json.loads(response)
 114
 115         return {
 116             '_type': 'playlist',
 117             'title': playlist_title,
 118             'id': cat_id,
 119             'entries': [self._extract_video(i) for i in videos],
 120         }