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 _real_extract_video(self, video_id):
  24         vid_parts = video_id.split(',')
  25         if len(vid_parts) == 3:
  26             video_id = '%s0%s%s-X-h' % (vid_parts[0][:4], vid_parts[1], vid_parts[2].rjust(4, '0'))
  27         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
  28         data = self._download_json(
  29             json_url, video_id, transform_source=self._fix_json)
  30         return self._extract_video(data[0])
  31
  32     def _extract_video(self, info):
  33         video_id = info['id']
  34         self.report_extraction(video_id)
  35
  36         initial_video_url = info['publishPoint']
  37         if info['formats'] == '1':
  38             parsed_url = compat_urllib_parse_urlparse(initial_video_url)
  39             filename, ext = os.path.splitext(parsed_url.path)
  40             path = '%s_sd%s' % (filename, ext)
  41             data = compat_urllib_parse.urlencode({
  42                 'type': 'fvod',
  43                 'path': compat_urlparse.urlunparse(parsed_url[:2] + (path,) + parsed_url[3:])
  44             })
  45             path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  46             path_doc = self._download_xml(
  47                 path_url, video_id, 'Downloading final video url')
  48             video_url = path_doc.find('path').text
  49         else:
  50             video_url = initial_video_url
  51
  52         join = compat_urlparse.urljoin
  53         ret = {
  54             'id': video_id,
  55             'title': info['name'],
  56             'url': video_url,
  57             'description': info['description'],
  58             'duration': int(info['duration']),
  59             'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
  60             'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
  61         }
  62         if video_url.startswith('rtmp:'):
  63             mobj = re.match(r'(?P<tc_url>rtmp://[^/]+/(?P<app>[a-z0-9/]+))/(?P<play_path>mp4:.*)', video_url)
  64             ret.update({
  65                 'tc_url': mobj.group('tc_url'),
  66                 'play_path': mobj.group('play_path'),
  67                 'app': mobj.group('app'),
  68                 'no_resume': True,
  69             })
  70         return ret
  71
  72
  73 class NHLIE(NHLBaseInfoExtractor):
  74     IE_NAME = 'nhl.com'
  75     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console|embed)?(?:\?(?:.*?[?&])?)(?:id|hlg|playlist)=(?P<id>[-0-9a-zA-Z,]+)'
  76
  77     _TESTS = [{
  78         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  79         'md5': 'db704a4ea09e8d3988c85e36cc892d09',
  80         'info_dict': {
  81             'id': '453614',
  82             'ext': 'mp4',
  83             'title': 'Quick clip: Weise 4-3 goal vs Flames',
  84             'description': 'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  85             'duration': 18,
  86             'upload_date': '20131006',
  87         },
  88     }, {
  89         'url': 'http://video.nhl.com/videocenter/console?id=2014020024-628-h',
  90         'md5': 'd22e82bc592f52d37d24b03531ee9696',
  91         'info_dict': {
  92             'id': '2014020024-628-h',
  93             'ext': 'mp4',
  94             'title': 'Alex Galchenyuk Goal on Ray Emery (14:40/3rd)',
  95             'description': 'Home broadcast - Montreal Canadiens at Philadelphia Flyers - October 11, 2014',
  96             'duration': 0,
  97             'upload_date': '20141011',
  98         },
  99     }, {
 100         'url': 'http://video.mapleleafs.nhl.com/videocenter/console?id=58665&catid=802',
 101         'md5': 'c78fc64ea01777e426cfc202b746c825',
 102         'info_dict': {
 103             'id': '58665',
 104             'ext': 'flv',
 105             'title': 'Classic Game In Six - April 22, 1979',
 106             '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.',
 107             'duration': 400,
 108             'upload_date': '20100129'
 109         },
 110     }, {
 111         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
 112         'only_matching': True,
 113     }, {
 114         'url': 'http://video.nhl.com/videocenter/?id=736722',
 115         'only_matching': True,
 116     }, {
 117         'url': 'http://video.nhl.com/videocenter/console?hlg=20142015,2,299&lang=en',
 118         'md5': '076fcb88c255154aacbf0a7accc3f340',
 119         'info_dict': {
 120             'id': '2014020299-X-h',
 121             'ext': 'mp4',
 122             'title': 'Penguins at Islanders / Game Highlights',
 123             'description': 'Home broadcast - Pittsburgh Penguins at New York Islanders - November 22, 2014',
 124             'duration': 268,
 125             'upload_date': '20141122',
 126         }
 127     }, {
 128         'url': 'http://video.oilers.nhl.com/videocenter/console?id=691469&catid=4',
 129         'info_dict': {
 130             'id': '691469',
 131             'ext': 'mp4',
 132             'title': 'RAW | Craig MacTavish Full Press Conference',
 133             'description': 'Oilers GM Craig MacTavish addresses the media at Rexall Place on Friday.',
 134             'upload_date': '20141205',
 135         },
 136         'params': {
 137             'skip_download': True,  # Requires rtmpdump
 138         }
 139     }, {
 140         'url': 'http://video.nhl.com/videocenter/embed?playlist=836127',
 141         'only_matching': True,
 142     }]
 143
 144     def _real_extract(self, url):
 145         video_id = self._match_id(url)
 146         return self._real_extract_video(video_id)
 147
 148
 149 class NHLNewsIE(NHLBaseInfoExtractor):
 150     IE_NAME = 'nhl.com:news'
 151     IE_DESC = 'NHL news'
 152     _VALID_URL = r'https?://(?:.+?\.)?nhl\.com/(?:ice|club)/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
 153
 154     _TESTS = [{
 155         'url': 'http://www.nhl.com/ice/news.htm?id=750727',
 156         'md5': '4b3d1262e177687a3009937bd9ec0be8',
 157         'info_dict': {
 158             'id': '736722',
 159             'ext': 'mp4',
 160             'title': 'Cal Clutterbuck has been fined $2,000',
 161             'description': 'md5:45fe547d30edab88b23e0dd0ab1ed9e6',
 162             'duration': 37,
 163             'upload_date': '20150128',
 164         },
 165     }, {
 166         # iframe embed
 167         'url': 'http://sabres.nhl.com/club/news.htm?id=780189',
 168         'md5': '9f663d1c006c90ac9fb82777d4294e12',
 169         'info_dict': {
 170             'id': '836127',
 171             'ext': 'mp4',
 172             'title': 'Morning Skate: OTT vs. BUF (9/23/15)',
 173             'description': "Brian Duff chats with Tyler Ennis prior to Buffalo's first preseason home game.",
 174             'duration': 93,
 175             'upload_date': '20150923',
 176         },
 177     }]
 178
 179     def _real_extract(self, url):
 180         news_id = self._match_id(url)
 181         webpage = self._download_webpage(url, news_id)
 182         video_id = self._search_regex(
 183             [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'",
 184              r'<iframe[^>]+src=["\']https?://video.*?\.nhl\.com/videocenter/embed\?.*\bplaylist=(\d+)'],
 185             webpage, 'video id')
 186         return self._real_extract_video(video_id)
 187
 188
 189 class NHLVideocenterIE(NHLBaseInfoExtractor):
 190     IE_NAME = 'nhl.com:videocenter'
 191     IE_DESC = 'NHL videocenter category'
 192     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?[^(id=)]*catid=(?P<catid>[0-9]+)(?![&?]id=).*?)?$'
 193     _TEST = {
 194         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=999',
 195         'info_dict': {
 196             'id': '999',
 197             'title': 'Highlights',
 198         },
 199         'playlist_count': 12,
 200     }
 201
 202     def _real_extract(self, url):
 203         mobj = re.match(self._VALID_URL, url)
 204         team = mobj.group('team')
 205         webpage = self._download_webpage(url, team)
 206         cat_id = self._search_regex(
 207             [r'var defaultCatId = "(.+?)";',
 208              r'{statusIndex:0,index:0,.*?id:(.*?),'],
 209             webpage, 'category id')
 210         playlist_title = self._html_search_regex(
 211             r'tab0"[^>]*?>(.*?)</td>',
 212             webpage, 'playlist title', flags=re.DOTALL).lower().capitalize()
 213
 214         data = compat_urllib_parse.urlencode({
 215             'cid': cat_id,
 216             # This is the default value
 217             'count': 12,
 218             'ptrs': 3,
 219             'format': 'json',
 220         })
 221         path = '/videocenter/servlets/browse?' + data
 222         request_url = compat_urlparse.urljoin(url, path)
 223         response = self._download_webpage(request_url, playlist_title)
 224         response = self._fix_json(response)
 225         if not response.strip():
 226             self._downloader.report_warning('Got an empty response, trying '
 227                                             'adding the "newvideos" parameter')
 228             response = self._download_webpage(request_url + '&newvideos=true',
 229                                               playlist_title)
 230             response = self._fix_json(response)
 231         videos = json.loads(response)
 232
 233         return {
 234             '_type': 'playlist',
 235             'title': playlist_title,
 236             'id': cat_id,
 237             'entries': [self._extract_video(v) for v in videos],
 238         }