Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/nhl.py

   1 from __future__ import unicode_literals
   2
   3 import re
   4 import json
   5 import os
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_urlparse,
  10     compat_urllib_parse,
  11     compat_urllib_parse_urlparse
  12 )
  13 from ..utils import (
  14     unified_strdate,
  15 )
  16
  17
  18 class NHLBaseInfoExtractor(InfoExtractor):
  19     @staticmethod
  20     def _fix_json(json_string):
  21         return json_string.replace('\\\'', '\'')
  22
  23     def _extract_video(self, info):
  24         video_id = info['id']
  25         self.report_extraction(video_id)
  26
  27         initial_video_url = info['publishPoint']
  28         if info['formats'] == '1':
  29             parsed_url = compat_urllib_parse_urlparse(initial_video_url)
  30             filename, ext = os.path.splitext(parsed_url.path)
  31             path = '%s_sd%s' % (filename, ext)
  32             data = compat_urllib_parse.urlencode({
  33                 'type': 'fvod',
  34                 'path': compat_urlparse.urlunparse(parsed_url[:2] + (path,) + parsed_url[3:])
  35             })
  36             path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  37             path_doc = self._download_xml(
  38                 path_url, video_id, 'Downloading final video url')
  39             video_url = path_doc.find('path').text
  40         else:
  41             video_url = initial_video_url
  42
  43         join = compat_urlparse.urljoin
  44         return {
  45             'id': video_id,
  46             'title': info['name'],
  47             'url': video_url,
  48             'description': info['description'],
  49             'duration': int(info['duration']),
  50             'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
  51             'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
  52         }
  53
  54
  55 class NHLIE(NHLBaseInfoExtractor):
  56     IE_NAME = 'nhl.com'
  57     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/console(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
  58
  59     _TESTS = [{
  60         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  61         'md5': 'db704a4ea09e8d3988c85e36cc892d09',
  62         'info_dict': {
  63             'id': '453614',
  64             'ext': 'mp4',
  65             'title': 'Quick clip: Weise 4-3 goal vs Flames',
  66             'description': 'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  67             'duration': 18,
  68             'upload_date': '20131006',
  69         },
  70     }, {
  71         'url': 'http://video.nhl.com/videocenter/console?id=2014020024-628-h',
  72         'md5': 'd22e82bc592f52d37d24b03531ee9696',
  73         'info_dict': {
  74             'id': '2014020024-628-h',
  75             'ext': 'mp4',
  76             'title': 'Alex Galchenyuk Goal on Ray Emery (14:40/3rd)',
  77             'description': 'Home broadcast - Montreal Canadiens at Philadelphia Flyers - October 11, 2014',
  78             'duration': 0,
  79             'upload_date': '20141011',
  80         },
  81     }, {
  82         'url': 'http://video.mapleleafs.nhl.com/videocenter/console?id=58665&catid=802',
  83         'md5': 'c78fc64ea01777e426cfc202b746c825',
  84         'info_dict': {
  85             'id': '58665',
  86             'ext': 'flv',
  87             'title': 'Classic Game In Six - April 22, 1979',
  88             'description': 'It was the last playoff game for the Leafs in the decade, and the last time the Leafs and Habs played in the playoffs. Great game, not a great ending.',
  89             'duration': 400,
  90             'upload_date': '20100129'
  91         },
  92     }, {
  93         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
  94         'only_matching': True,
  95     }]
  96
  97     def _real_extract(self, url):
  98         mobj = re.match(self._VALID_URL, url)
  99         video_id = mobj.group('id')
 100         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
 101         data = self._download_json(
 102             json_url, video_id, transform_source=self._fix_json)
 103         return self._extract_video(data[0])
 104
 105
 106 class NHLVideocenterIE(NHLBaseInfoExtractor):
 107     IE_NAME = 'nhl.com:videocenter'
 108     IE_DESC = 'NHL videocenter category'
 109     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?[^(id=)]*catid=(?P<catid>[0-9]+)(?![&?]id=).*?)?$'
 110     _TEST = {
 111         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=999',
 112         'info_dict': {
 113             'id': '999',
 114             'title': 'Highlights',
 115         },
 116         'playlist_count': 12,
 117     }
 118
 119     def _real_extract(self, url):
 120         mobj = re.match(self._VALID_URL, url)
 121         team = mobj.group('team')
 122         webpage = self._download_webpage(url, team)
 123         cat_id = self._search_regex(
 124             [r'var defaultCatId = "(.+?)";',
 125              r'{statusIndex:0,index:0,.*?id:(.*?),'],
 126             webpage, 'category id')
 127         playlist_title = self._html_search_regex(
 128             r'tab0"[^>]*?>(.*?)</td>',
 129             webpage, 'playlist title', flags=re.DOTALL).lower().capitalize()
 130
 131         data = compat_urllib_parse.urlencode({
 132             'cid': cat_id,
 133             # This is the default value
 134             'count': 12,
 135             'ptrs': 3,
 136             'format': 'json',
 137         })
 138         path = '/videocenter/servlets/browse?' + data
 139         request_url = compat_urlparse.urljoin(url, path)
 140         response = self._download_webpage(request_url, playlist_title)
 141         response = self._fix_json(response)
 142         if not response.strip():
 143             self._downloader.report_warning('Got an empty reponse, trying '
 144                                             'adding the "newvideos" parameter')
 145             response = self._download_webpage(request_url + '&newvideos=true',
 146                                               playlist_title)
 147             response = self._fix_json(response)
 148         videos = json.loads(response)
 149
 150         return {
 151             '_type': 'playlist',
 152             'title': playlist_title,
 153             'id': cat_id,
 154             'entries': [self._extract_video(v) for v in videos],
 155         }