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 ..utils import (
   8     compat_urlparse,
   9     compat_urllib_parse,
  10     determine_ext,
  11     unified_strdate,
  12 )
  13
  14
  15 class NHLBaseInfoExtractor(InfoExtractor):
  16     @staticmethod
  17     def _fix_json(json_string):
  18         return json_string.replace('\\\'', '\'')
  19
  20     def _extract_video(self, info):
  21         video_id = info['id']
  22         self.report_extraction(video_id)
  23
  24         initial_video_url = info['publishPoint']
  25         data = compat_urllib_parse.urlencode({
  26             'type': 'fvod',
  27             'path': initial_video_url.replace('.mp4', '_sd.mp4'),
  28         })
  29         path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  30         path_doc = self._download_xml(
  31             path_url, video_id, 'Downloading final video url')
  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 = 'nhl.com'
  49     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console(?:\?(?:.*?[?&])?)id=(?P<id>[0-9]+)'
  50
  51     _TESTS = [{
  52         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  53         'info_dict': {
  54             'id': '453614',
  55             'ext': 'mp4',
  56             'title': 'Quick clip: Weise 4-3 goal vs Flames',
  57             'description': 'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  58             'duration': 18,
  59             'upload_date': '20131006',
  60         },
  61     }, {
  62         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
  63         'only_matching': True,
  64     }]
  65
  66     def _real_extract(self, url):
  67         mobj = re.match(self._VALID_URL, url)
  68         video_id = mobj.group('id')
  69         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
  70         data = self._download_json(
  71             json_url, video_id, transform_source=self._fix_json)
  72         return self._extract_video(data[0])
  73
  74
  75 class NHLVideocenterIE(NHLBaseInfoExtractor):
  76     IE_NAME = 'nhl.com:videocenter'
  77     IE_DESC = 'NHL videocenter category'
  78     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?.*?catid=(?P<catid>[0-9]+)(?![&?]id=).*?)?$'
  79     _TEST = {
  80         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=999',
  81         'info_dict': {
  82             'id': '999',
  83             'title': 'Highlights',
  84         },
  85         'playlist_count': 12,
  86     }
  87
  88     def _real_extract(self, url):
  89         mobj = re.match(self._VALID_URL, url)
  90         team = mobj.group('team')
  91         webpage = self._download_webpage(url, team)
  92         cat_id = self._search_regex(
  93             [r'var defaultCatId = "(.+?)";',
  94              r'{statusIndex:0,index:0,.*?id:(.*?),'],
  95             webpage, 'category id')
  96         playlist_title = self._html_search_regex(
  97             r'tab0"[^>]*?>(.*?)</td>',
  98             webpage, 'playlist title', flags=re.DOTALL).lower().capitalize()
  99
 100         data = compat_urllib_parse.urlencode({
 101             'cid': cat_id,
 102             # This is the default value
 103             'count': 12,
 104             'ptrs': 3,
 105             'format': 'json',
 106         })
 107         path = '/videocenter/servlets/browse?' + data
 108         request_url = compat_urlparse.urljoin(url, path)
 109         response = self._download_webpage(request_url, playlist_title)
 110         response = self._fix_json(response)
 111         if not response.strip():
 112             self._downloader.report_warning(u'Got an empty reponse, trying '
 113                                             'adding the "newvideos" parameter')
 114             response = self._download_webpage(request_url + '&newvideos=true',
 115                 playlist_title)
 116             response = self._fix_json(response)
 117         videos = json.loads(response)
 118
 119         return {
 120             '_type': 'playlist',
 121             'title': playlist_title,
 122             'id': cat_id,
 123             'entries': [self._extract_video(v) for v in videos],
 124         }