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_urlencode,
  11     compat_urllib_parse_urlparse,
  12     compat_str,
  13 )
  14 from ..utils import (
  15     unified_strdate,
  16     determine_ext,
  17     int_or_none,
  18     parse_iso8601,
  19     parse_duration,
  20 )
  21
  22
  23 class NHLBaseInfoExtractor(InfoExtractor):
  24     @staticmethod
  25     def _fix_json(json_string):
  26         return json_string.replace('\\\'', '\'')
  27
  28     def _real_extract_video(self, video_id):
  29         vid_parts = video_id.split(',')
  30         if len(vid_parts) == 3:
  31             video_id = '%s0%s%s-X-h' % (vid_parts[0][:4], vid_parts[1], vid_parts[2].rjust(4, '0'))
  32         json_url = 'http://video.nhl.com/videocenter/servlets/playlist?ids=%s&format=json' % video_id
  33         data = self._download_json(
  34             json_url, video_id, transform_source=self._fix_json)
  35         return self._extract_video(data[0])
  36
  37     def _extract_video(self, info):
  38         video_id = info['id']
  39         self.report_extraction(video_id)
  40
  41         initial_video_url = info['publishPoint']
  42         if info['formats'] == '1':
  43             parsed_url = compat_urllib_parse_urlparse(initial_video_url)
  44             filename, ext = os.path.splitext(parsed_url.path)
  45             path = '%s_sd%s' % (filename, ext)
  46             data = compat_urllib_parse_urlencode({
  47                 'type': 'fvod',
  48                 'path': compat_urlparse.urlunparse(parsed_url[:2] + (path,) + parsed_url[3:])
  49             })
  50             path_url = 'http://video.nhl.com/videocenter/servlets/encryptvideopath?' + data
  51             path_doc = self._download_xml(
  52                 path_url, video_id, 'Downloading final video url')
  53             video_url = path_doc.find('path').text
  54         else:
  55             video_url = initial_video_url
  56
  57         join = compat_urlparse.urljoin
  58         ret = {
  59             'id': video_id,
  60             'title': info['name'],
  61             'url': video_url,
  62             'description': info['description'],
  63             'duration': int(info['duration']),
  64             'thumbnail': join(join(video_url, '/u/'), info['bigImage']),
  65             'upload_date': unified_strdate(info['releaseDate'].split('.')[0]),
  66         }
  67         if video_url.startswith('rtmp:'):
  68             mobj = re.match(r'(?P<tc_url>rtmp://[^/]+/(?P<app>[a-z0-9/]+))/(?P<play_path>mp4:.*)', video_url)
  69             ret.update({
  70                 'tc_url': mobj.group('tc_url'),
  71                 'play_path': mobj.group('play_path'),
  72                 'app': mobj.group('app'),
  73                 'no_resume': True,
  74             })
  75         return ret
  76
  77
  78 class NHLVideocenterIE(NHLBaseInfoExtractor):
  79     IE_NAME = 'nhl.com:videocenter'
  80     _VALID_URL = r'https?://video(?P<team>\.[^.]*)?\.nhl\.com/videocenter/(?:console|embed)?(?:\?(?:.*?[?&])?)(?:id|hlg|playlist)=(?P<id>[-0-9a-zA-Z,]+)'
  81
  82     _TESTS = [{
  83         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=6?id=453614',
  84         'md5': 'db704a4ea09e8d3988c85e36cc892d09',
  85         'info_dict': {
  86             'id': '453614',
  87             'ext': 'mp4',
  88             'title': 'Quick clip: Weise 4-3 goal vs Flames',
  89             'description': 'Dale Weise scores his first of the season to put the Canucks up 4-3.',
  90             'duration': 18,
  91             'upload_date': '20131006',
  92         },
  93     }, {
  94         'url': 'http://video.nhl.com/videocenter/console?id=2014020024-628-h',
  95         'md5': 'd22e82bc592f52d37d24b03531ee9696',
  96         'info_dict': {
  97             'id': '2014020024-628-h',
  98             'ext': 'mp4',
  99             'title': 'Alex Galchenyuk Goal on Ray Emery (14:40/3rd)',
 100             'description': 'Home broadcast - Montreal Canadiens at Philadelphia Flyers - October 11, 2014',
 101             'duration': 0,
 102             'upload_date': '20141011',
 103         },
 104     }, {
 105         'url': 'http://video.mapleleafs.nhl.com/videocenter/console?id=58665&catid=802',
 106         'md5': 'c78fc64ea01777e426cfc202b746c825',
 107         'info_dict': {
 108             'id': '58665',
 109             'ext': 'flv',
 110             'title': 'Classic Game In Six - April 22, 1979',
 111             '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.',
 112             'duration': 400,
 113             'upload_date': '20100129'
 114         },
 115     }, {
 116         'url': 'http://video.flames.nhl.com/videocenter/console?id=630616',
 117         'only_matching': True,
 118     }, {
 119         'url': 'http://video.nhl.com/videocenter/?id=736722',
 120         'only_matching': True,
 121     }, {
 122         'url': 'http://video.nhl.com/videocenter/console?hlg=20142015,2,299&lang=en',
 123         'md5': '076fcb88c255154aacbf0a7accc3f340',
 124         'info_dict': {
 125             'id': '2014020299-X-h',
 126             'ext': 'mp4',
 127             'title': 'Penguins at Islanders / Game Highlights',
 128             'description': 'Home broadcast - Pittsburgh Penguins at New York Islanders - November 22, 2014',
 129             'duration': 268,
 130             'upload_date': '20141122',
 131         }
 132     }, {
 133         'url': 'http://video.oilers.nhl.com/videocenter/console?id=691469&catid=4',
 134         'info_dict': {
 135             'id': '691469',
 136             'ext': 'mp4',
 137             'title': 'RAW | Craig MacTavish Full Press Conference',
 138             'description': 'Oilers GM Craig MacTavish addresses the media at Rexall Place on Friday.',
 139             'upload_date': '20141205',
 140         },
 141         'params': {
 142             'skip_download': True,  # Requires rtmpdump
 143         }
 144     }, {
 145         'url': 'http://video.nhl.com/videocenter/embed?playlist=836127',
 146         'only_matching': True,
 147     }]
 148
 149     def _real_extract(self, url):
 150         video_id = self._match_id(url)
 151         return self._real_extract_video(video_id)
 152
 153
 154 class NHLNewsIE(NHLBaseInfoExtractor):
 155     IE_NAME = 'nhl.com:news'
 156     IE_DESC = 'NHL news'
 157     _VALID_URL = r'https?://(?:.+?\.)?nhl\.com/(?:ice|club)/news\.html?(?:\?(?:.*?[?&])?)id=(?P<id>[-0-9a-zA-Z]+)'
 158
 159     _TESTS = [{
 160         'url': 'http://www.nhl.com/ice/news.htm?id=750727',
 161         'md5': '4b3d1262e177687a3009937bd9ec0be8',
 162         'info_dict': {
 163             'id': '736722',
 164             'ext': 'mp4',
 165             'title': 'Cal Clutterbuck has been fined $2,000',
 166             'description': 'md5:45fe547d30edab88b23e0dd0ab1ed9e6',
 167             'duration': 37,
 168             'upload_date': '20150128',
 169         },
 170     }, {
 171         # iframe embed
 172         'url': 'http://sabres.nhl.com/club/news.htm?id=780189',
 173         'md5': '9f663d1c006c90ac9fb82777d4294e12',
 174         'info_dict': {
 175             'id': '836127',
 176             'ext': 'mp4',
 177             'title': 'Morning Skate: OTT vs. BUF (9/23/15)',
 178             'description': "Brian Duff chats with Tyler Ennis prior to Buffalo's first preseason home game.",
 179             'duration': 93,
 180             'upload_date': '20150923',
 181         },
 182     }]
 183
 184     def _real_extract(self, url):
 185         news_id = self._match_id(url)
 186         webpage = self._download_webpage(url, news_id)
 187         video_id = self._search_regex(
 188             [r'pVid(\d+)', r"nlid\s*:\s*'(\d+)'",
 189              r'<iframe[^>]+src=["\']https?://video.*?\.nhl\.com/videocenter/embed\?.*\bplaylist=(\d+)'],
 190             webpage, 'video id')
 191         return self._real_extract_video(video_id)
 192
 193
 194 class NHLVideocenterCategoryIE(NHLBaseInfoExtractor):
 195     IE_NAME = 'nhl.com:videocenter:category'
 196     IE_DESC = 'NHL videocenter category'
 197     _VALID_URL = r'https?://video\.(?P<team>[^.]*)\.nhl\.com/videocenter/(console\?[^(id=)]*catid=(?P<catid>[0-9]+)(?![&?]id=).*?)?$'
 198     _TEST = {
 199         'url': 'http://video.canucks.nhl.com/videocenter/console?catid=999',
 200         'info_dict': {
 201             'id': '999',
 202             'title': 'Highlights',
 203         },
 204         'playlist_count': 12,
 205     }
 206
 207     def _real_extract(self, url):
 208         mobj = re.match(self._VALID_URL, url)
 209         team = mobj.group('team')
 210         webpage = self._download_webpage(url, team)
 211         cat_id = self._search_regex(
 212             [r'var defaultCatId = "(.+?)";',
 213              r'{statusIndex:0,index:0,.*?id:(.*?),'],
 214             webpage, 'category id')
 215         playlist_title = self._html_search_regex(
 216             r'tab0"[^>]*?>(.*?)</td>',
 217             webpage, 'playlist title', flags=re.DOTALL).lower().capitalize()
 218
 219         data = compat_urllib_parse_urlencode({
 220             'cid': cat_id,
 221             # This is the default value
 222             'count': 12,
 223             'ptrs': 3,
 224             'format': 'json',
 225         })
 226         path = '/videocenter/servlets/browse?' + data
 227         request_url = compat_urlparse.urljoin(url, path)
 228         response = self._download_webpage(request_url, playlist_title)
 229         response = self._fix_json(response)
 230         if not response.strip():
 231             self._downloader.report_warning('Got an empty response, trying '
 232                                             'adding the "newvideos" parameter')
 233             response = self._download_webpage(request_url + '&newvideos=true',
 234                                               playlist_title)
 235             response = self._fix_json(response)
 236         videos = json.loads(response)
 237
 238         return {
 239             '_type': 'playlist',
 240             'title': playlist_title,
 241             'id': cat_id,
 242             'entries': [self._extract_video(v) for v in videos],
 243         }
 244
 245
 246 class NHLIE(InfoExtractor):
 247     IE_NAME = 'nhl.com'
 248     _VALID_URL = r'https?://(?:www\.)?nhl\.com/([^/]+/)*c-(?P<id>\d+)'
 249     _TESTS = [{
 250         # type=video
 251         'url': 'https://www.nhl.com/video/anisimov-cleans-up-mess/t-277752844/c-43663503',
 252         'md5': '0f7b9a8f986fb4b4eeeece9a56416eaf',
 253         'info_dict': {
 254             'id': '43663503',
 255             'ext': 'mp4',
 256             'title': 'Anisimov cleans up mess',
 257             'description': 'md5:a02354acdfe900e940ce40706939ca63',
 258             'timestamp': 1461288600,
 259             'upload_date': '20160422',
 260         },
 261     }, {
 262         # type=article
 263         'url': 'https://www.nhl.com/news/dennis-wideman-suspended/c-278258934',
 264         'md5': '1f39f4ea74c1394dea110699a25b366c',
 265         'info_dict': {
 266             'id': '40784403',
 267             'ext': 'mp4',
 268             'title': 'Wideman suspended by NHL',
 269             'description': 'Flames defenseman Dennis Wideman was banned 20 games for violation of Rule 40 (Physical Abuse of Officials)',
 270             'upload_date': '20160204',
 271             'timestamp': 1454544904,
 272         },
 273     }]
 274
 275     def _real_extract(self, url):
 276         tmp_id = self._match_id(url)
 277         video_data = self._download_json(
 278             'https://nhl.bamcontent.com/nhl/id/v1/%s/details/web-v1.json' % tmp_id,
 279             tmp_id)
 280         if video_data.get('type') == 'article':
 281             video_data = video_data['media']
 282
 283         video_id = compat_str(video_data['id'])
 284         title = video_data['title']
 285
 286         formats = []
 287         for playback in video_data.get('playbacks', []):
 288             playback_url = playback.get('url')
 289             if not playback_url:
 290                 continue
 291             ext = determine_ext(playback_url)
 292             if ext == 'm3u8':
 293                 formats.extend(self._extract_m3u8_formats(
 294                     playback_url, video_id, 'mp4', 'm3u8_native',
 295                     m3u8_id=playback.get('name', 'hls'), fatal=False))
 296             else:
 297                 height = int_or_none(playback.get('height'))
 298                 formats.append({
 299                     'format_id': playback.get('name', 'http' + ('-%dp' % height if height else '')),
 300                     'url': playback_url,
 301                     'width': int_or_none(playback.get('width')),
 302                     'height': height,
 303                 })
 304         self._sort_formats(formats, ('preference', 'width', 'height', 'tbr', 'format_id'))
 305
 306         thumbnails = []
 307         for thumbnail_id, thumbnail_data in video_data.get('image', {}).get('cuts', {}).items():
 308             thumbnail_url = thumbnail_data.get('src')
 309             if not thumbnail_url:
 310                 continue
 311             thumbnails.append({
 312                 'id': thumbnail_id,
 313                 'url': thumbnail_url,
 314                 'width': int_or_none(thumbnail_data.get('width')),
 315                 'height': int_or_none(thumbnail_data.get('height')),
 316             })
 317
 318         return {
 319             'id': video_id,
 320             'title': title,
 321             'description': video_data.get('description'),
 322             'timestamp': parse_iso8601(video_data.get('date')),
 323             'duration': parse_duration(video_data.get('duration')),
 324             'thumbnails': thumbnails,
 325             'formats': formats,
 326         }