_ Git - youtube-dl/blob - youtube_dl/extractor/cbc.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import json
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_str
   9 from ..utils import (
  10     js_to_json,
  11     smuggle_url,
  12     try_get,
  13     xpath_text,
  14     xpath_element,
  15     xpath_with_ns,
  16     find_xpath_attr,
  17     parse_duration,
  18     parse_iso8601,
  19     parse_age_limit,
  20     int_or_none,
  21     ExtractorError,
  22 )
  23
  24
  25 class CBCIE(InfoExtractor):
  26     IE_NAME = 'cbc.ca'
  27     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)'
  28     _TESTS = [{
  29         # with mediaId
  30         'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs',
  31         'md5': '97e24d09672fc4cf56256d6faa6c25bc',
  32         'info_dict': {
  33             'id': '2682904050',
  34             'ext': 'mp4',
  35             'title': 'Don Cherry – All-Stars',
  36             'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.',
  37             'timestamp': 1454463000,
  38             'upload_date': '20160203',
  39             'uploader': 'CBCC-NEW',
  40         },
  41         'skip': 'Geo-restricted to Canada',
  42     }, {
  43         # with clipId, feed available via tpfeed.cbc.ca and feed.theplatform.com
  44         'url': 'http://www.cbc.ca/22minutes/videos/22-minutes-update/22-minutes-update-episode-4',
  45         'md5': '162adfa070274b144f4fdc3c3b8207db',
  46         'info_dict': {
  47             'id': '2414435309',
  48             'ext': 'mp4',
  49             'title': '22 Minutes Update: What Not To Wear Quebec',
  50             'description': "This week's latest Canadian top political story is What Not To Wear Quebec.",
  51             'upload_date': '20131025',
  52             'uploader': 'CBCC-NEW',
  53             'timestamp': 1382717907,
  54         },
  55     }, {
  56         # with clipId, feed only available via tpfeed.cbc.ca
  57         'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
  58         'md5': '0274a90b51a9b4971fe005c63f592f12',
  59         'info_dict': {
  60             'id': '2487345465',
  61             'ext': 'mp4',
  62             'title': 'Robin Williams freestyles on 90 Minutes Live',
  63             'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
  64             'upload_date': '19780210',
  65             'uploader': 'CBCC-NEW',
  66             'timestamp': 255977160,
  67         },
  68     }, {
  69         # multiple iframes
  70         'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
  71         'playlist': [{
  72             'md5': '377572d0b49c4ce0c9ad77470e0b96b4',
  73             'info_dict': {
  74                 'id': '2680832926',
  75                 'ext': 'mp4',
  76                 'title': 'An Eagle\'s-Eye View Off Burrard Bridge',
  77                 'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.',
  78                 'upload_date': '20160201',
  79                 'timestamp': 1454342820,
  80                 'uploader': 'CBCC-NEW',
  81             },
  82         }, {
  83             'md5': '415a0e3f586113894174dfb31aa5bb1a',
  84             'info_dict': {
  85                 'id': '2658915080',
  86                 'ext': 'mp4',
  87                 'title': 'Fly like an eagle!',
  88                 'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower',
  89                 'upload_date': '20150315',
  90                 'timestamp': 1426443984,
  91                 'uploader': 'CBCC-NEW',
  92             },
  93         }],
  94         'skip': 'Geo-restricted to Canada',
  95     }, {
  96         # multiple CBC.APP.Caffeine.initInstance(...)
  97         'url': 'http://www.cbc.ca/news/canada/calgary/dog-indoor-exercise-winter-1.3928238',
  98         'info_dict': {
  99             'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks',
 100             'id': 'dog-indoor-exercise-winter-1.3928238',
 101             'description': 'md5:c18552e41726ee95bd75210d1ca9194c',
 102         },
 103         'playlist_mincount': 6,
 104     }]
 105
 106     @classmethod
 107     def suitable(cls, url):
 108         return False if CBCPlayerIE.suitable(url) else super(CBCIE, cls).suitable(url)
 109
 110     def _extract_player_init(self, player_init, display_id):
 111         player_info = self._parse_json(player_init, display_id, js_to_json)
 112         media_id = player_info.get('mediaId')
 113         if not media_id:
 114             clip_id = player_info['clipId']
 115             feed = self._download_json(
 116                 'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id,
 117                 clip_id, fatal=False)
 118             if feed:
 119                 media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str)
 120             if not media_id:
 121                 media_id = self._download_json(
 122                     'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
 123                     clip_id)['entries'][0]['id'].split('/')[-1]
 124         return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
 125
 126     def _real_extract(self, url):
 127         display_id = self._match_id(url)
 128         webpage = self._download_webpage(url, display_id)
 129         entries = [
 130             self._extract_player_init(player_init, display_id)
 131             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
 132         entries.extend([
 133             self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
 134             for media_id in re.findall(r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"', webpage)])
 135         return self.playlist_result(
 136             entries, display_id,
 137             self._og_search_title(webpage, fatal=False),
 138             self._og_search_description(webpage))
 139
 140
 141 class CBCPlayerIE(InfoExtractor):
 142     IE_NAME = 'cbc.ca:player'
 143     _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>\d+)'
 144     _TESTS = [{
 145         'url': 'http://www.cbc.ca/player/play/2683190193',
 146         'md5': '64d25f841ddf4ddb28a235338af32e2c',
 147         'info_dict': {
 148             'id': '2683190193',
 149             'ext': 'mp4',
 150             'title': 'Gerry Runs a Sweat Shop',
 151             'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0',
 152             'timestamp': 1455071400,
 153             'upload_date': '20160210',
 154             'uploader': 'CBCC-NEW',
 155         },
 156         'skip': 'Geo-restricted to Canada',
 157     }, {
 158         # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
 159         'url': 'http://www.cbc.ca/player/play/2657631896',
 160         'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
 161         'info_dict': {
 162             'id': '2657631896',
 163             'ext': 'mp3',
 164             'title': 'CBC Montreal is organizing its first ever community hackathon!',
 165             'description': 'The modern technology we tend to depend on so heavily, is never without it\'s share of hiccups and headaches. Next weekend - CBC Montreal will be getting members of the public for its first Hackathon.',
 166             'timestamp': 1425704400,
 167             'upload_date': '20150307',
 168             'uploader': 'CBCC-NEW',
 169         },
 170     }, {
 171         'url': 'http://www.cbc.ca/player/play/2164402062',
 172         'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6',
 173         'info_dict': {
 174             'id': '2164402062',
 175             'ext': 'mp4',
 176             'title': 'Cancer survivor four times over',
 177             'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
 178             'timestamp': 1320410746,
 179             'upload_date': '20111104',
 180             'uploader': 'CBCC-NEW',
 181         },
 182     }]
 183
 184     def _real_extract(self, url):
 185         video_id = self._match_id(url)
 186         return {
 187             '_type': 'url_transparent',
 188             'ie_key': 'ThePlatform',
 189             'url': smuggle_url(
 190                 'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/%s?mbr=true&formats=MPEG4,FLV,MP3' % video_id, {
 191                     'force_smil_url': True
 192                 }),
 193             'id': video_id,
 194         }
 195
 196
 197 class CBCWatchBaseIE(InfoExtractor):
 198     _device_id = None
 199     _device_token = None
 200     _API_BASE_URL = 'https://api-cbc.cloud.clearleap.com/cloffice/client/'
 201     _NS_MAP = {
 202         'media': 'http://search.yahoo.com/mrss/',
 203         'clearleap': 'http://www.clearleap.com/namespace/clearleap/1.0/',
 204     }
 205     _GEO_COUNTRIES = ['CA']
 206
 207     def _call_api(self, path, video_id):
 208         url = path if path.startswith('http') else self._API_BASE_URL + path
 209         result = self._download_xml(url, video_id, headers={
 210             'X-Clearleap-DeviceId': self._device_id,
 211             'X-Clearleap-DeviceToken': self._device_token,
 212         })
 213         error_message = xpath_text(result, 'userMessage') or xpath_text(result, 'systemMessage')
 214         if error_message:
 215             raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message))
 216         return result
 217
 218     def _real_initialize(self):
 219         if not self._device_id or not self._device_token:
 220             device = self._downloader.cache.load('cbcwatch', 'device') or {}
 221             self._device_id, self._device_token = device.get('id'), device.get('token')
 222             if not self._device_id or not self._device_token:
 223                 result = self._download_xml(
 224                     self._API_BASE_URL + 'device/register',
 225                     None, data=b'<device><type>web</type></device>')
 226                 self._device_id = xpath_text(result, 'deviceId', fatal=True)
 227                 self._device_token = xpath_text(result, 'deviceToken', fatal=True)
 228                 self._downloader.cache.store(
 229                     'cbcwatch', 'device', {
 230                         'id': self._device_id,
 231                         'token': self._device_token,
 232                     })
 233
 234     def _parse_rss_feed(self, rss):
 235         channel = xpath_element(rss, 'channel', fatal=True)
 236
 237         def _add_ns(path):
 238             return xpath_with_ns(path, self._NS_MAP)
 239
 240         entries = []
 241         for item in channel.findall('item'):
 242             guid = xpath_text(item, 'guid', fatal=True)
 243             title = xpath_text(item, 'title', fatal=True)
 244
 245             media_group = xpath_element(item, _add_ns('media:group'), fatal=True)
 246             content = xpath_element(media_group, _add_ns('media:content'), fatal=True)
 247             content_url = content.attrib['url']
 248
 249             thumbnails = []
 250             for thumbnail in media_group.findall(_add_ns('media:thumbnail')):
 251                 thumbnail_url = thumbnail.get('url')
 252                 if not thumbnail_url:
 253                     continue
 254                 thumbnails.append({
 255                     'id': thumbnail.get('profile'),
 256                     'url': thumbnail_url,
 257                     'width': int_or_none(thumbnail.get('width')),
 258                     'height': int_or_none(thumbnail.get('height')),
 259                 })
 260
 261             timestamp = None
 262             release_date = find_xpath_attr(
 263                 item, _add_ns('media:credit'), 'role', 'releaseDate')
 264             if release_date is not None:
 265                 timestamp = parse_iso8601(release_date.text)
 266
 267             entries.append({
 268                 '_type': 'url_transparent',
 269                 'url': content_url,
 270                 'id': guid,
 271                 'title': title,
 272                 'description': xpath_text(item, 'description'),
 273                 'timestamp': timestamp,
 274                 'duration': int_or_none(content.get('duration')),
 275                 'age_limit': parse_age_limit(xpath_text(item, _add_ns('media:rating'))),
 276                 'episode': xpath_text(item, _add_ns('clearleap:episode')),
 277                 'episode_number': int_or_none(xpath_text(item, _add_ns('clearleap:episodeInSeason'))),
 278                 'series': xpath_text(item, _add_ns('clearleap:series')),
 279                 'season_number': int_or_none(xpath_text(item, _add_ns('clearleap:season'))),
 280                 'thumbnails': thumbnails,
 281                 'ie_key': 'CBCWatchVideo',
 282             })
 283
 284         return self.playlist_result(
 285             entries, xpath_text(channel, 'guid'),
 286             xpath_text(channel, 'title'),
 287             xpath_text(channel, 'description'))
 288
 289
 290 class CBCWatchVideoIE(CBCWatchBaseIE):
 291     IE_NAME = 'cbc.ca:watch:video'
 292     _VALID_URL = r'https?://api-cbc\.cloud\.clearleap\.com/cloffice/client/web/play/?\?.*?\bcontentId=(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'
 293     _TEST = {
 294         # geo-restricted to Canada, bypassable
 295         'url': 'https://api-cbc.cloud.clearleap.com/cloffice/client/web/play/?contentId=3c84472a-1eea-4dee-9267-2655d5055dcf&categoryId=ebc258f5-ee40-4cca-b66b-ba6bd55b7235',
 296         'only_matching': True,
 297     }
 298
 299     def _real_extract(self, url):
 300         video_id = self._match_id(url)
 301         result = self._call_api(url, video_id)
 302
 303         m3u8_url = xpath_text(result, 'url', fatal=True)
 304         formats = self._extract_m3u8_formats(re.sub(r'/([^/]+)/[^/?]+\.m3u8', r'/\1/\1.m3u8', m3u8_url), video_id, 'mp4', fatal=False)
 305         if len(formats) < 2:
 306             formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
 307         for f in formats:
 308             format_id = f.get('format_id')
 309             if format_id.startswith('AAC'):
 310                 f['acodec'] = 'aac'
 311             elif format_id.startswith('AC3'):
 312                 f['acodec'] = 'ac-3'
 313         self._sort_formats(formats)
 314
 315         info = {
 316             'id': video_id,
 317             'title': video_id,
 318             'formats': formats,
 319         }
 320
 321         rss = xpath_element(result, 'rss')
 322         if rss:
 323             info.update(self._parse_rss_feed(rss)['entries'][0])
 324             del info['url']
 325             del info['_type']
 326             del info['ie_key']
 327         return info
 328
 329
 330 class CBCWatchIE(CBCWatchBaseIE):
 331     IE_NAME = 'cbc.ca:watch'
 332     _VALID_URL = r'https?://watch\.cbc\.ca/(?:[^/]+/)+(?P<id>[0-9a-f-]+)'
 333     _TESTS = [{
 334         # geo-restricted to Canada, bypassable
 335         'url': 'http://watch.cbc.ca/doc-zone/season-6/customer-disservice/38e815a-009e3ab12e4',
 336         'info_dict': {
 337             'id': '9673749a-5e77-484c-8b62-a1092a6b5168',
 338             'ext': 'mp4',
 339             'title': 'Customer (Dis)Service',
 340             'description': 'md5:8bdd6913a0fe03d4b2a17ebe169c7c87',
 341             'upload_date': '20160219',
 342             'timestamp': 1455840000,
 343         },
 344         'params': {
 345             # m3u8 download
 346             'skip_download': True,
 347             'format': 'bestvideo',
 348         },
 349     }, {
 350         # geo-restricted to Canada, bypassable
 351         'url': 'http://watch.cbc.ca/arthur/all/1ed4b385-cd84-49cf-95f0-80f004680057',
 352         'info_dict': {
 353             'id': '1ed4b385-cd84-49cf-95f0-80f004680057',
 354             'title': 'Arthur',
 355             'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.',
 356         },
 357         'playlist_mincount': 30,
 358     }]
 359
 360     def _real_extract(self, url):
 361         video_id = self._match_id(url)
 362         rss = self._call_api('web/browse/' + video_id, video_id)
 363         return self._parse_rss_feed(rss)
 364
 365
 366 class CBCOlympicsIE(InfoExtractor):
 367     IE_NAME = 'cbc.ca:olympics'
 368     _VALID_URL = r'https?://olympics\.cbc\.ca/video/[^/]+/(?P<id>[^/?#]+)'
 369     _TESTS = [{
 370         'url': 'https://olympics.cbc.ca/video/whats-on-tv/olympic-morning-featuring-the-opening-ceremony/',
 371         'only_matching': True,
 372     }]
 373
 374     def _real_extract(self, url):
 375         display_id = self._match_id(url)
 376         webpage = self._download_webpage(url, display_id)
 377         video_id = self._hidden_inputs(webpage)['videoId']
 378         video_doc = self._download_xml(
 379             'https://olympics.cbc.ca/videodata/%s.xml' % video_id, video_id)
 380         title = xpath_text(video_doc, 'title', fatal=True)
 381         is_live = xpath_text(video_doc, 'kind') == 'Live'
 382         if is_live:
 383             title = self._live_title(title)
 384
 385         formats = []
 386         for video_source in video_doc.findall('videoSources/videoSource'):
 387             uri = xpath_text(video_source, 'uri')
 388             if not uri:
 389                 continue
 390             tokenize = self._download_json(
 391                 'https://olympics.cbc.ca/api/api-akamai/tokenize',
 392                 video_id, data=json.dumps({
 393                     'VideoSource': uri,
 394                 }).encode(), headers={
 395                     'Content-Type': 'application/json',
 396                     'Referer': url,
 397                     # d3.VideoPlayer._init in https://olympics.cbc.ca/components/script/base.js
 398                     'Cookie': '_dvp=TK:C0ObxjerU',  # AKAMAI CDN cookie
 399                 }, fatal=False)
 400             if not tokenize:
 401                 continue
 402             content_url = tokenize['ContentUrl']
 403             video_source_format = video_source.get('format')
 404             if video_source_format == 'IIS':
 405                 formats.extend(self._extract_ism_formats(
 406                     content_url, video_id, ism_id=video_source_format, fatal=False))
 407             else:
 408                 formats.extend(self._extract_m3u8_formats(
 409                     content_url, video_id, 'mp4',
 410                     'm3u8' if is_live else 'm3u8_native',
 411                     m3u8_id=video_source_format, fatal=False))
 412         self._sort_formats(formats)
 413
 414         return {
 415             'id': video_id,
 416             'display_id': display_id,
 417             'title': title,
 418             'description': xpath_text(video_doc, 'description'),
 419             'thumbnail': xpath_text(video_doc, 'thumbnailUrl'),
 420             'duration': parse_duration(xpath_text(video_doc, 'duration')),
 421             'formats': formats,
 422             'is_live': is_live,
 423         }