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