_ 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 from xml.sax.saxutils import escape
   7
   8 from .common import InfoExtractor
   9 from ..compat import (
  10     compat_str,
  11     compat_HTTPError,
  12 )
  13 from ..utils import (
  14     js_to_json,
  15     smuggle_url,
  16     try_get,
  17     xpath_text,
  18     xpath_element,
  19     xpath_with_ns,
  20     find_xpath_attr,
  21     orderedSet,
  22     parse_duration,
  23     parse_iso8601,
  24     parse_age_limit,
  25     strip_or_none,
  26     int_or_none,
  27     ExtractorError,
  28 )
  29
  30
  31 class CBCIE(InfoExtractor):
  32     IE_NAME = 'cbc.ca'
  33     _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P<id>[^/?#]+)'
  34     _TESTS = [{
  35         # with mediaId
  36         'url': 'http://www.cbc.ca/22minutes/videos/clips-season-23/don-cherry-play-offs',
  37         'md5': '97e24d09672fc4cf56256d6faa6c25bc',
  38         'info_dict': {
  39             'id': '2682904050',
  40             'ext': 'mp4',
  41             'title': 'Don Cherry – All-Stars',
  42             'description': 'Don Cherry has a bee in his bonnet about AHL player John Scott because that guy’s got heart.',
  43             'timestamp': 1454463000,
  44             'upload_date': '20160203',
  45             'uploader': 'CBCC-NEW',
  46         },
  47         'skip': 'Geo-restricted to Canada',
  48     }, {
  49         # with clipId, feed available via tpfeed.cbc.ca and feed.theplatform.com
  50         'url': 'http://www.cbc.ca/22minutes/videos/22-minutes-update/22-minutes-update-episode-4',
  51         'md5': '162adfa070274b144f4fdc3c3b8207db',
  52         'info_dict': {
  53             'id': '2414435309',
  54             'ext': 'mp4',
  55             'title': '22 Minutes Update: What Not To Wear Quebec',
  56             'description': "This week's latest Canadian top political story is What Not To Wear Quebec.",
  57             'upload_date': '20131025',
  58             'uploader': 'CBCC-NEW',
  59             'timestamp': 1382717907,
  60         },
  61     }, {
  62         # with clipId, feed only available via tpfeed.cbc.ca
  63         'url': 'http://www.cbc.ca/archives/entry/1978-robin-williams-freestyles-on-90-minutes-live',
  64         'md5': '0274a90b51a9b4971fe005c63f592f12',
  65         'info_dict': {
  66             'id': '2487345465',
  67             'ext': 'mp4',
  68             'title': 'Robin Williams freestyles on 90 Minutes Live',
  69             'description': 'Wacky American comedian Robin Williams shows off his infamous "freestyle" comedic talents while being interviewed on CBC\'s 90 Minutes Live.',
  70             'upload_date': '19780210',
  71             'uploader': 'CBCC-NEW',
  72             'timestamp': 255977160,
  73         },
  74     }, {
  75         # multiple iframes
  76         'url': 'http://www.cbc.ca/natureofthings/blog/birds-eye-view-from-vancouvers-burrard-street-bridge-how-we-got-the-shot',
  77         'playlist': [{
  78             'md5': '377572d0b49c4ce0c9ad77470e0b96b4',
  79             'info_dict': {
  80                 'id': '2680832926',
  81                 'ext': 'mp4',
  82                 'title': 'An Eagle\'s-Eye View Off Burrard Bridge',
  83                 'description': 'Hercules the eagle flies from Vancouver\'s Burrard Bridge down to a nearby park with a mini-camera strapped to his back.',
  84                 'upload_date': '20160201',
  85                 'timestamp': 1454342820,
  86                 'uploader': 'CBCC-NEW',
  87             },
  88         }, {
  89             'md5': '415a0e3f586113894174dfb31aa5bb1a',
  90             'info_dict': {
  91                 'id': '2658915080',
  92                 'ext': 'mp4',
  93                 'title': 'Fly like an eagle!',
  94                 'description': 'Eagle equipped with a mini camera flies from the world\'s tallest tower',
  95                 'upload_date': '20150315',
  96                 'timestamp': 1426443984,
  97                 'uploader': 'CBCC-NEW',
  98             },
  99         }],
 100         'skip': 'Geo-restricted to Canada',
 101     }, {
 102         # multiple CBC.APP.Caffeine.initInstance(...)
 103         'url': 'http://www.cbc.ca/news/canada/calgary/dog-indoor-exercise-winter-1.3928238',
 104         'info_dict': {
 105             'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks',
 106             'id': 'dog-indoor-exercise-winter-1.3928238',
 107             'description': 'md5:c18552e41726ee95bd75210d1ca9194c',
 108         },
 109         'playlist_mincount': 6,
 110     }]
 111
 112     @classmethod
 113     def suitable(cls, url):
 114         return False if CBCPlayerIE.suitable(url) else super(CBCIE, cls).suitable(url)
 115
 116     def _extract_player_init(self, player_init, display_id):
 117         player_info = self._parse_json(player_init, display_id, js_to_json)
 118         media_id = player_info.get('mediaId')
 119         if not media_id:
 120             clip_id = player_info['clipId']
 121             feed = self._download_json(
 122                 'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id,
 123                 clip_id, fatal=False)
 124             if feed:
 125                 media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str)
 126             if not media_id:
 127                 media_id = self._download_json(
 128                     'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id,
 129                     clip_id)['entries'][0]['id'].split('/')[-1]
 130         return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
 131
 132     def _real_extract(self, url):
 133         display_id = self._match_id(url)
 134         webpage = self._download_webpage(url, display_id)
 135         title = self._og_search_title(webpage, default=None) or self._html_search_meta(
 136             'twitter:title', webpage, 'title', default=None) or self._html_search_regex(
 137                 r'<title>([^<]+)</title>', webpage, 'title', fatal=False)
 138         entries = [
 139             self._extract_player_init(player_init, display_id)
 140             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
 141         media_ids = []
 142         for media_id_re in (
 143                 r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"',
 144                 r'<div[^>]+\bid=["\']player-(\d+)',
 145                 r'guid["\']\s*:\s*["\'](\d+)'):
 146             media_ids.extend(re.findall(media_id_re, webpage))
 147         entries.extend([
 148             self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
 149             for media_id in orderedSet(media_ids)])
 150         return self.playlist_result(
 151             entries, display_id, strip_or_none(title),
 152             self._og_search_description(webpage))
 153
 154
 155 class CBCPlayerIE(InfoExtractor):
 156     IE_NAME = 'cbc.ca:player'
 157     _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>\d+)'
 158     _TESTS = [{
 159         'url': 'http://www.cbc.ca/player/play/2683190193',
 160         'md5': '64d25f841ddf4ddb28a235338af32e2c',
 161         'info_dict': {
 162             'id': '2683190193',
 163             'ext': 'mp4',
 164             'title': 'Gerry Runs a Sweat Shop',
 165             'description': 'md5:b457e1c01e8ff408d9d801c1c2cd29b0',
 166             'timestamp': 1455071400,
 167             'upload_date': '20160210',
 168             'uploader': 'CBCC-NEW',
 169         },
 170         'skip': 'Geo-restricted to Canada',
 171     }, {
 172         # Redirected from http://www.cbc.ca/player/AudioMobile/All%20in%20a%20Weekend%20Montreal/ID/2657632011/
 173         'url': 'http://www.cbc.ca/player/play/2657631896',
 174         'md5': 'e5e708c34ae6fca156aafe17c43e8b75',
 175         'info_dict': {
 176             'id': '2657631896',
 177             'ext': 'mp3',
 178             'title': 'CBC Montreal is organizing its first ever community hackathon!',
 179             '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.',
 180             'timestamp': 1425704400,
 181             'upload_date': '20150307',
 182             'uploader': 'CBCC-NEW',
 183         },
 184     }, {
 185         'url': 'http://www.cbc.ca/player/play/2164402062',
 186         'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6',
 187         'info_dict': {
 188             'id': '2164402062',
 189             'ext': 'mp4',
 190             'title': 'Cancer survivor four times over',
 191             'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.',
 192             'timestamp': 1320410746,
 193             'upload_date': '20111104',
 194             'uploader': 'CBCC-NEW',
 195         },
 196     }]
 197
 198     def _real_extract(self, url):
 199         video_id = self._match_id(url)
 200         return {
 201             '_type': 'url_transparent',
 202             'ie_key': 'ThePlatform',
 203             'url': smuggle_url(
 204                 'http://link.theplatform.com/s/ExhSPC/media/guid/2655402169/%s?mbr=true&formats=MPEG4,FLV,MP3' % video_id, {
 205                     'force_smil_url': True
 206                 }),
 207             'id': video_id,
 208         }
 209
 210
 211 class CBCWatchBaseIE(InfoExtractor):
 212     _device_id = None
 213     _device_token = None
 214     _API_BASE_URL = 'https://api-cbc.cloud.clearleap.com/cloffice/client/'
 215     _NS_MAP = {
 216         'media': 'http://search.yahoo.com/mrss/',
 217         'clearleap': 'http://www.clearleap.com/namespace/clearleap/1.0/',
 218     }
 219     _GEO_COUNTRIES = ['CA']
 220     _LOGIN_URL = 'https://api.loginradius.com/identity/v2/auth/login'
 221     _TOKEN_URL = 'https://cloud-api.loginradius.com/sso/jwt/api/token'
 222     _API_KEY = '3f4beddd-2061-49b0-ae80-6f1f2ed65b37'
 223     _NETRC_MACHINE = 'cbcwatch'
 224
 225     def _signature(self, email, password):
 226         data = json.dumps({
 227             'email': email,
 228             'password': password,
 229         }).encode()
 230         headers = {'content-type': 'application/json'}
 231         query = {'apikey': self._API_KEY}
 232         resp = self._download_json(self._LOGIN_URL, None, data=data, headers=headers, query=query)
 233         access_token = resp['access_token']
 234
 235         # token
 236         query = {
 237             'access_token': access_token,
 238             'apikey': self._API_KEY,
 239             'jwtapp': 'jwt',
 240         }
 241         resp = self._download_json(self._TOKEN_URL, None, headers=headers, query=query)
 242         return resp['signature']
 243
 244     def _call_api(self, path, video_id):
 245         url = path if path.startswith('http') else self._API_BASE_URL + path
 246         for _ in range(2):
 247             try:
 248                 result = self._download_xml(url, video_id, headers={
 249                     'X-Clearleap-DeviceId': self._device_id,
 250                     'X-Clearleap-DeviceToken': self._device_token,
 251                 })
 252             except ExtractorError as e:
 253                 if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
 254                     # Device token has expired, re-acquiring device token
 255                     self._register_device()
 256                     continue
 257                 raise
 258         error_message = xpath_text(result, 'userMessage') or xpath_text(result, 'systemMessage')
 259         if error_message:
 260             raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message))
 261         return result
 262
 263     def _real_initialize(self):
 264         if self._valid_device_token():
 265             return
 266         device = self._downloader.cache.load('cbcwatch', 'device') or {}
 267         self._device_id, self._device_token = device.get('id'), device.get('token')
 268         if self._valid_device_token():
 269             return
 270         self._register_device()
 271
 272     def _valid_device_token(self):
 273         return self._device_id and self._device_token
 274
 275     def _register_device(self):
 276         result = self._download_xml(
 277             self._API_BASE_URL + 'device/register',
 278             None, 'Acquiring device token',
 279             data=b'<device><type>web</type></device>')
 280         self._device_id = xpath_text(result, 'deviceId', fatal=True)
 281         anon_device_token = xpath_text(result, 'deviceToken', fatal=True)
 282         email, password = self._get_login_info()
 283         if email and password:
 284             signature = self._signature(email, password)
 285             data = '<login><token>{0}</token><device><deviceId>{1}</deviceId><type>web</type></device></login>'.format(escape(signature), escape(self._device_id)).encode()
 286             url = self._API_BASE_URL + 'device/login'
 287             result = self._download_xml(url, None, data=data, headers={'content-type': 'application/xml'})
 288             self._device_token = xpath_text(result, 'token', fatal=True)
 289         else:
 290             self._device_token = anon_device_token
 291         self._downloader.cache.store(
 292             'cbcwatch', 'device', {
 293                 'id': self._device_id,
 294                 'token': self._device_token,
 295             })
 296
 297     def _parse_rss_feed(self, rss):
 298         channel = xpath_element(rss, 'channel', fatal=True)
 299
 300         def _add_ns(path):
 301             return xpath_with_ns(path, self._NS_MAP)
 302
 303         entries = []
 304         for item in channel.findall('item'):
 305             guid = xpath_text(item, 'guid', fatal=True)
 306             title = xpath_text(item, 'title', fatal=True)
 307
 308             media_group = xpath_element(item, _add_ns('media:group'), fatal=True)
 309             content = xpath_element(media_group, _add_ns('media:content'), fatal=True)
 310             content_url = content.attrib['url']
 311
 312             thumbnails = []
 313             for thumbnail in media_group.findall(_add_ns('media:thumbnail')):
 314                 thumbnail_url = thumbnail.get('url')
 315                 if not thumbnail_url:
 316                     continue
 317                 thumbnails.append({
 318                     'id': thumbnail.get('profile'),
 319                     'url': thumbnail_url,
 320                     'width': int_or_none(thumbnail.get('width')),
 321                     'height': int_or_none(thumbnail.get('height')),
 322                 })
 323
 324             timestamp = None
 325             release_date = find_xpath_attr(
 326                 item, _add_ns('media:credit'), 'role', 'releaseDate')
 327             if release_date is not None:
 328                 timestamp = parse_iso8601(release_date.text)
 329
 330             entries.append({
 331                 '_type': 'url_transparent',
 332                 'url': content_url,
 333                 'id': guid,
 334                 'title': title,
 335                 'description': xpath_text(item, 'description'),
 336                 'timestamp': timestamp,
 337                 'duration': int_or_none(content.get('duration')),
 338                 'age_limit': parse_age_limit(xpath_text(item, _add_ns('media:rating'))),
 339                 'episode': xpath_text(item, _add_ns('clearleap:episode')),
 340                 'episode_number': int_or_none(xpath_text(item, _add_ns('clearleap:episodeInSeason'))),
 341                 'series': xpath_text(item, _add_ns('clearleap:series')),
 342                 'season_number': int_or_none(xpath_text(item, _add_ns('clearleap:season'))),
 343                 'thumbnails': thumbnails,
 344                 'ie_key': 'CBCWatchVideo',
 345             })
 346
 347         return self.playlist_result(
 348             entries, xpath_text(channel, 'guid'),
 349             xpath_text(channel, 'title'),
 350             xpath_text(channel, 'description'))
 351
 352
 353 class CBCWatchVideoIE(CBCWatchBaseIE):
 354     IE_NAME = 'cbc.ca:watch:video'
 355     _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})'
 356     _TEST = {
 357         # geo-restricted to Canada, bypassable
 358         'url': 'https://api-cbc.cloud.clearleap.com/cloffice/client/web/play/?contentId=3c84472a-1eea-4dee-9267-2655d5055dcf&categoryId=ebc258f5-ee40-4cca-b66b-ba6bd55b7235',
 359         'only_matching': True,
 360     }
 361
 362     def _real_extract(self, url):
 363         video_id = self._match_id(url)
 364         result = self._call_api(url, video_id)
 365
 366         m3u8_url = xpath_text(result, 'url', fatal=True)
 367         formats = self._extract_m3u8_formats(re.sub(r'/([^/]+)/[^/?]+\.m3u8', r'/\1/\1.m3u8', m3u8_url), video_id, 'mp4', fatal=False)
 368         if len(formats) < 2:
 369             formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4')
 370         for f in formats:
 371             format_id = f.get('format_id')
 372             if format_id.startswith('AAC'):
 373                 f['acodec'] = 'aac'
 374             elif format_id.startswith('AC3'):
 375                 f['acodec'] = 'ac-3'
 376         self._sort_formats(formats)
 377
 378         info = {
 379             'id': video_id,
 380             'title': video_id,
 381             'formats': formats,
 382         }
 383
 384         rss = xpath_element(result, 'rss')
 385         if rss:
 386             info.update(self._parse_rss_feed(rss)['entries'][0])
 387             del info['url']
 388             del info['_type']
 389             del info['ie_key']
 390         return info
 391
 392
 393 class CBCWatchIE(CBCWatchBaseIE):
 394     IE_NAME = 'cbc.ca:watch'
 395     _VALID_URL = r'https?://(?:gem|watch)\.cbc\.ca/(?:[^/]+/)+(?P<id>[0-9a-f-]+)'
 396     _TESTS = [{
 397         # geo-restricted to Canada, bypassable
 398         'url': 'http://watch.cbc.ca/doc-zone/season-6/customer-disservice/38e815a-009e3ab12e4',
 399         'info_dict': {
 400             'id': '9673749a-5e77-484c-8b62-a1092a6b5168',
 401             'ext': 'mp4',
 402             'title': 'Customer (Dis)Service',
 403             'description': 'md5:8bdd6913a0fe03d4b2a17ebe169c7c87',
 404             'upload_date': '20160219',
 405             'timestamp': 1455840000,
 406         },
 407         'params': {
 408             # m3u8 download
 409             'skip_download': True,
 410             'format': 'bestvideo',
 411         },
 412     }, {
 413         # geo-restricted to Canada, bypassable
 414         'url': 'http://watch.cbc.ca/arthur/all/1ed4b385-cd84-49cf-95f0-80f004680057',
 415         'info_dict': {
 416             'id': '1ed4b385-cd84-49cf-95f0-80f004680057',
 417             'title': 'Arthur',
 418             'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.',
 419         },
 420         'playlist_mincount': 30,
 421     }, {
 422         'url': 'https://gem.cbc.ca/media/this-hour-has-22-minutes/season-26/episode-20/38e815a-0108c6c6a42',
 423         'only_matching': True,
 424     }]
 425
 426     def _real_extract(self, url):
 427         video_id = self._match_id(url)
 428         rss = self._call_api('web/browse/' + video_id, video_id)
 429         return self._parse_rss_feed(rss)
 430
 431
 432 class CBCOlympicsIE(InfoExtractor):
 433     IE_NAME = 'cbc.ca:olympics'
 434     _VALID_URL = r'https?://olympics\.cbc\.ca/video/[^/]+/(?P<id>[^/?#]+)'
 435     _TESTS = [{
 436         'url': 'https://olympics.cbc.ca/video/whats-on-tv/olympic-morning-featuring-the-opening-ceremony/',
 437         'only_matching': True,
 438     }]
 439
 440     def _real_extract(self, url):
 441         display_id = self._match_id(url)
 442         webpage = self._download_webpage(url, display_id)
 443         video_id = self._hidden_inputs(webpage)['videoId']
 444         video_doc = self._download_xml(
 445             'https://olympics.cbc.ca/videodata/%s.xml' % video_id, video_id)
 446         title = xpath_text(video_doc, 'title', fatal=True)
 447         is_live = xpath_text(video_doc, 'kind') == 'Live'
 448         if is_live:
 449             title = self._live_title(title)
 450
 451         formats = []
 452         for video_source in video_doc.findall('videoSources/videoSource'):
 453             uri = xpath_text(video_source, 'uri')
 454             if not uri:
 455                 continue
 456             tokenize = self._download_json(
 457                 'https://olympics.cbc.ca/api/api-akamai/tokenize',
 458                 video_id, data=json.dumps({
 459                     'VideoSource': uri,
 460                 }).encode(), headers={
 461                     'Content-Type': 'application/json',
 462                     'Referer': url,
 463                     # d3.VideoPlayer._init in https://olympics.cbc.ca/components/script/base.js
 464                     'Cookie': '_dvp=TK:C0ObxjerU',  # AKAMAI CDN cookie
 465                 }, fatal=False)
 466             if not tokenize:
 467                 continue
 468             content_url = tokenize['ContentUrl']
 469             video_source_format = video_source.get('format')
 470             if video_source_format == 'IIS':
 471                 formats.extend(self._extract_ism_formats(
 472                     content_url, video_id, ism_id=video_source_format, fatal=False))
 473             else:
 474                 formats.extend(self._extract_m3u8_formats(
 475                     content_url, video_id, 'mp4',
 476                     'm3u8' if is_live else 'm3u8_native',
 477                     m3u8_id=video_source_format, fatal=False))
 478         self._sort_formats(formats)
 479
 480         return {
 481             'id': video_id,
 482             'display_id': display_id,
 483             'title': title,
 484             'description': xpath_text(video_doc, 'description'),
 485             'thumbnail': xpath_text(video_doc, 'thumbnailUrl'),
 486             'duration': parse_duration(xpath_text(video_doc, 'duration')),
 487             'formats': formats,
 488             'is_live': is_live,
 489         }