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