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