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

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5
   6 from .common import InfoExtractor
   7 from ..compat import (
   8     compat_urlparse,
   9     compat_urllib_parse,
  10 )
  11 from ..utils import (
  12     unified_strdate,
  13 )
  14
  15
  16 class NHLBaseInfoExtractor(InfoExtractor):
  17     @staticmethod
  18     def _fix_json(json_string):
  19         return json_string.replace('\\\'', '\'')
  20
  21     def _extract_video(self, info):
  22         video_id = info['id']
  23         self.report_extraction(video_id)
  24
  25         initial_video_url = info['publishPoint']
  26         if info['formats'] == '1':
  27             data = compat_urllib_parse.urlencode({
  28                 'type': 'fvod',
  29                 'path': initial_video_url.replace('.mp4', '_sd.mp4'),
  30             })
  31             path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  32             path_doc = self._download_xml(
  33                 path_url, video_id, 'Downloading final video url')
  34             video_url = path_doc.find('path').text
  35         else:
  36             video_url = initial_video_url
  37
  38         join = compat_urlparse.urljoin
  39         return {
  40             'id': video_id,
  41             'title': info['name'],
  42             'url': video_url,
  43             'description': info['description'],
  44             'duration': int(info['duration']),
  45             'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
  46             'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
  47         }
  48
  49
  50 class NHLIE(NHLBaseInfoExtractor):
  51     IE_NAME = 'nhl.com'
  52     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console(?:\?(?:.*?[?&])?)id=(?P<id>[0-9a-z-]+)'
  53
  54     _TESTS = [{
  55         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  56         'md5': 'db704a4ea09e8d3988c85e36cc892d09',
  57         'info_dict': {
  58             'id': '453614',
  59             'ext': 'mp4',
  60             'title': 'Quick clip: Weise 4-3 goal vs Flames',
  61             'description': 'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  62             'duration': 18,
  63             'upload_date': '20131006',
  64         },
  65     }, {
  66         'url': 'http://video.nhl.com/videocenter/console?id=2014020024-628-h',
  67         'md5': 'd22e82bc592f52d37d24b03531ee9696',
  68         'info_dict': {
  69             'id': '2014020024-628-h',
  70             'ext': 'mp4',
  71             'title': 'Alex Galchenyuk Goal on Ray Emery (14:40/3rd)',
  72             'description': 'Home broadcast - Montreal Canadiens at Philadelphia Flyers - October 11, 2014',
  73             'duration': 0,
  74             'upload_date': '20141011',
  75         },
  76     }, {
  77         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
  78         'only_matching': True,
  79     }]
  80
  81     def _real_extract(self, url):
  82         mobj = re.match(self._VALID_URL, url)
  83         video_id = mobj.group('id')
  84         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
  85         data = self._download_json(
  86             json_url, video_id, transform_source=self._fix_json)
  87         return self._extract_video(data[0])
  88
  89
  90 class NHLVideocenterIE(NHLBaseInfoExtractor):
  91     IE_NAME = 'nhl.com:videocenter'
  92     IE_DESC = 'NHL videocenter category'
  93     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?[^(id=)]*catid=(?P<catid>[0-9]+)(?![&?]id=).*?)?$'
  94     _TEST = {
  95         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=999',
  96         'info_dict': {
  97             'id': '999',
  98             'title': 'Highlights',
  99         },
 100         'playlist_count': 12,
 101     }
 102
 103     def _real_extract(self, url):
 104         mobj = re.match(self._VALID_URL, url)
 105         team = mobj.group('team')
 106         webpage = self._download_webpage(url, team)
 107         cat_id = self._search_regex(
 108             [r'var defaultCatId = "(.+?)";',
 109              r'{statusIndex:0,index:0,.*?id:(.*?),'],
 110             webpage, 'category id')
 111         playlist_title = self._html_search_regex(
 112             r'tab0"[^>]*?>(.*?)</td>',
 113             webpage, 'playlist title', flags=re.DOTALL).lower().capitalize()
 114
 115         data = compat_urllib_parse.urlencode({
 116             'cid': cat_id,
 117             # This is the default value
 118             'count': 12,
 119             'ptrs': 3,
 120             'format': 'json',
 121         })
 122         path = '/videocenter/servlets/browse?' + data
 123         request_url = compat_urlparse.urljoin(url, path)
 124         response = self._download_webpage(request_url, playlist_title)
 125         response = self._fix_json(response)
 126         if not response.strip():
 127             self._downloader.report_warning('Got an empty reponse, trying '
 128                                             'adding the "newvideos" parameter')
 129             response = self._download_webpage(request_url + '&newvideos=true',
 130                                               playlist_title)
 131             response = self._fix_json(response)
 132         videos = json.loads(response)
 133
 134         return {
 135             '_type': 'playlist',
 136             'title': playlist_title,
 137             'id': cat_id,
 138             'entries': [self._extract_video(v) for v in videos],
 139         }