X-Git-Url: https://git.rapsys.eu/youtubedl/blobdiff_plain/99b46abf79266c2bf5eabc291dc4f366e61b5dde..4231113a7be907539052ce59df44e23c54d21d38:/youtube_dl/extractor/cbc.py diff --git a/youtube_dl/extractor/cbc.py b/youtube_dl/extractor/cbc.py index a87e971..9faf402 100644 --- a/youtube_dl/extractor/cbc.py +++ b/youtube_dl/extractor/cbc.py @@ -9,10 +9,19 @@ from ..utils import ( js_to_json, smuggle_url, try_get, + xpath_text, + xpath_element, + xpath_with_ns, + find_xpath_attr, + parse_iso8601, + parse_age_limit, + int_or_none, + ExtractorError, ) class CBCIE(InfoExtractor): + IE_NAME = 'cbc.ca' _VALID_URL = r'https?://(?:www\.)?cbc\.ca/(?!player/)(?:[^/]+/)+(?P[^/?#]+)' _TESTS = [{ # with mediaId @@ -81,39 +90,54 @@ class CBCIE(InfoExtractor): }, }], 'skip': 'Geo-restricted to Canada', + }, { + # multiple CBC.APP.Caffeine.initInstance(...) + 'url': 'http://www.cbc.ca/news/canada/calgary/dog-indoor-exercise-winter-1.3928238', + 'info_dict': { + 'title': 'Keep Rover active during the deep freeze with doggie pushups and other fun indoor tasks', + 'id': 'dog-indoor-exercise-winter-1.3928238', + 'description': 'md5:c18552e41726ee95bd75210d1ca9194c', + }, + 'playlist_mincount': 6, }] @classmethod def suitable(cls, url): return False if CBCPlayerIE.suitable(url) else super(CBCIE, cls).suitable(url) + def _extract_player_init(self, player_init, display_id): + player_info = self._parse_json(player_init, display_id, js_to_json) + media_id = player_info.get('mediaId') + if not media_id: + clip_id = player_info['clipId'] + feed = self._download_json( + 'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id, + clip_id, fatal=False) + if feed: + media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str) + if not media_id: + media_id = self._download_json( + 'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id, + clip_id)['entries'][0]['id'].split('/')[-1] + return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) + def _real_extract(self, url): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - player_init = self._search_regex( - r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage, 'player init', - default=None) - if player_init: - player_info = self._parse_json(player_init, display_id, js_to_json) - media_id = player_info.get('mediaId') - if not media_id: - clip_id = player_info['clipId'] - feed = self._download_json( - 'http://tpfeed.cbc.ca/f/ExhSPC/vms_5akSXx4Ng_Zn?byCustomValue={:mpsReleases}{%s}' % clip_id, - clip_id, fatal=False) - if feed: - media_id = try_get(feed, lambda x: x['entries'][0]['guid'], compat_str) - if not media_id: - media_id = self._download_json( - 'http://feed.theplatform.com/f/h9dtGB/punlNGjMlc1F?fields=id&byContent=byReleases%3DbyId%253D' + clip_id, - clip_id)['entries'][0]['id'].split('/')[-1] - return self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) - else: - entries = [self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) for media_id in re.findall(r']+src="[^"]+?mediaId=(\d+)"', webpage)] - return self.playlist_result(entries) + entries = [ + self._extract_player_init(player_init, display_id) + for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)] + entries.extend([ + self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id) + for media_id in re.findall(r']+src="[^"]+?mediaId=(\d+)"', webpage)]) + return self.playlist_result( + entries, display_id, + self._og_search_title(webpage, fatal=False), + self._og_search_description(webpage)) class CBCPlayerIE(InfoExtractor): + IE_NAME = 'cbc.ca:player' _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P\d+)' _TESTS = [{ 'url': 'http://www.cbc.ca/player/play/2683190193', @@ -142,12 +166,11 @@ class CBCPlayerIE(InfoExtractor): 'uploader': 'CBCC-NEW', }, }, { - # available only when we add `formats=MPEG4,FLV,MP3` to theplatform url 'url': 'http://www.cbc.ca/player/play/2164402062', - 'md5': '17a61eb813539abea40618d6323a7f82', + 'md5': '33fcd8f6719b9dd60a5e73adcb83b9f6', 'info_dict': { 'id': '2164402062', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Cancer survivor four times over', 'description': 'Tim Mayer has beaten three different forms of cancer four times in five years.', 'timestamp': 1320410746, @@ -167,3 +190,172 @@ class CBCPlayerIE(InfoExtractor): }), 'id': video_id, } + + +class CBCWatchBaseIE(InfoExtractor): + _device_id = None + _device_token = None + _API_BASE_URL = 'https://api-cbc.cloud.clearleap.com/cloffice/client/' + _NS_MAP = { + 'media': 'http://search.yahoo.com/mrss/', + 'clearleap': 'http://www.clearleap.com/namespace/clearleap/1.0/', + } + _GEO_COUNTRIES = ['CA'] + + def _call_api(self, path, video_id): + url = path if path.startswith('http') else self._API_BASE_URL + path + result = self._download_xml(url, video_id, headers={ + 'X-Clearleap-DeviceId': self._device_id, + 'X-Clearleap-DeviceToken': self._device_token, + }) + error_message = xpath_text(result, 'userMessage') or xpath_text(result, 'systemMessage') + if error_message: + raise ExtractorError('%s said: %s' % (self.IE_NAME, error_message)) + return result + + def _real_initialize(self): + if not self._device_id or not self._device_token: + device = self._downloader.cache.load('cbcwatch', 'device') or {} + self._device_id, self._device_token = device.get('id'), device.get('token') + if not self._device_id or not self._device_token: + result = self._download_xml( + self._API_BASE_URL + 'device/register', + None, data=b'web') + self._device_id = xpath_text(result, 'deviceId', fatal=True) + self._device_token = xpath_text(result, 'deviceToken', fatal=True) + self._downloader.cache.store( + 'cbcwatch', 'device', { + 'id': self._device_id, + 'token': self._device_token, + }) + + def _parse_rss_feed(self, rss): + channel = xpath_element(rss, 'channel', fatal=True) + + def _add_ns(path): + return xpath_with_ns(path, self._NS_MAP) + + entries = [] + for item in channel.findall('item'): + guid = xpath_text(item, 'guid', fatal=True) + title = xpath_text(item, 'title', fatal=True) + + media_group = xpath_element(item, _add_ns('media:group'), fatal=True) + content = xpath_element(media_group, _add_ns('media:content'), fatal=True) + content_url = content.attrib['url'] + + thumbnails = [] + for thumbnail in media_group.findall(_add_ns('media:thumbnail')): + thumbnail_url = thumbnail.get('url') + if not thumbnail_url: + continue + thumbnails.append({ + 'id': thumbnail.get('profile'), + 'url': thumbnail_url, + 'width': int_or_none(thumbnail.get('width')), + 'height': int_or_none(thumbnail.get('height')), + }) + + timestamp = None + release_date = find_xpath_attr( + item, _add_ns('media:credit'), 'role', 'releaseDate') + if release_date is not None: + timestamp = parse_iso8601(release_date.text) + + entries.append({ + '_type': 'url_transparent', + 'url': content_url, + 'id': guid, + 'title': title, + 'description': xpath_text(item, 'description'), + 'timestamp': timestamp, + 'duration': int_or_none(content.get('duration')), + 'age_limit': parse_age_limit(xpath_text(item, _add_ns('media:rating'))), + 'episode': xpath_text(item, _add_ns('clearleap:episode')), + 'episode_number': int_or_none(xpath_text(item, _add_ns('clearleap:episodeInSeason'))), + 'series': xpath_text(item, _add_ns('clearleap:series')), + 'season_number': int_or_none(xpath_text(item, _add_ns('clearleap:season'))), + 'thumbnails': thumbnails, + 'ie_key': 'CBCWatchVideo', + }) + + return self.playlist_result( + entries, xpath_text(channel, 'guid'), + xpath_text(channel, 'title'), + xpath_text(channel, 'description')) + + +class CBCWatchVideoIE(CBCWatchBaseIE): + IE_NAME = 'cbc.ca:watch:video' + _VALID_URL = r'https?://api-cbc\.cloud\.clearleap\.com/cloffice/client/web/play/?\?.*?\bcontentId=(?P[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _TEST = { + # geo-restricted to Canada, bypassable + 'url': 'https://api-cbc.cloud.clearleap.com/cloffice/client/web/play/?contentId=3c84472a-1eea-4dee-9267-2655d5055dcf&categoryId=ebc258f5-ee40-4cca-b66b-ba6bd55b7235', + 'only_matching': True, + } + + def _real_extract(self, url): + video_id = self._match_id(url) + result = self._call_api(url, video_id) + + m3u8_url = xpath_text(result, 'url', fatal=True) + formats = self._extract_m3u8_formats(re.sub(r'/([^/]+)/[^/?]+\.m3u8', r'/\1/\1.m3u8', m3u8_url), video_id, 'mp4', fatal=False) + if len(formats) < 2: + formats = self._extract_m3u8_formats(m3u8_url, video_id, 'mp4') + for f in formats: + format_id = f.get('format_id') + if format_id.startswith('AAC'): + f['acodec'] = 'aac' + elif format_id.startswith('AC3'): + f['acodec'] = 'ac-3' + self._sort_formats(formats) + + info = { + 'id': video_id, + 'title': video_id, + 'formats': formats, + } + + rss = xpath_element(result, 'rss') + if rss: + info.update(self._parse_rss_feed(rss)['entries'][0]) + del info['url'] + del info['_type'] + del info['ie_key'] + return info + + +class CBCWatchIE(CBCWatchBaseIE): + IE_NAME = 'cbc.ca:watch' + _VALID_URL = r'https?://watch\.cbc\.ca/(?:[^/]+/)+(?P[0-9a-f-]+)' + _TESTS = [{ + # geo-restricted to Canada, bypassable + 'url': 'http://watch.cbc.ca/doc-zone/season-6/customer-disservice/38e815a-009e3ab12e4', + 'info_dict': { + 'id': '9673749a-5e77-484c-8b62-a1092a6b5168', + 'ext': 'mp4', + 'title': 'Customer (Dis)Service', + 'description': 'md5:8bdd6913a0fe03d4b2a17ebe169c7c87', + 'upload_date': '20160219', + 'timestamp': 1455840000, + }, + 'params': { + # m3u8 download + 'skip_download': True, + 'format': 'bestvideo', + }, + }, { + # geo-restricted to Canada, bypassable + 'url': 'http://watch.cbc.ca/arthur/all/1ed4b385-cd84-49cf-95f0-80f004680057', + 'info_dict': { + 'id': '1ed4b385-cd84-49cf-95f0-80f004680057', + 'title': 'Arthur', + 'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.', + }, + 'playlist_mincount': 30, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + rss = self._call_api('web/browse/' + video_id, video_id) + return self._parse_rss_feed(rss)