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