]> Raphaƫl G. Git Repositories - youtubedl/blobdiff - youtube_dl/extractor/cbc.py
Update upstream source from tag 'upstream/2020.03.24'
[youtubedl] / youtube_dl / extractor / cbc.py
index 87ad14e914132f099e30cc300254b40cb10f8f61..fd5ec6033b80513012cf2615fc56e80c7e82cadc 100644 (file)
@@ -1,10 +1,16 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import hashlib
+import json
 import re
+from xml.sax.saxutils import escape
 
 from .common import InfoExtractor
-from ..compat import compat_str
+from ..compat import (
+    compat_str,
+    compat_HTTPError,
+)
 from ..utils import (
     js_to_json,
     smuggle_url,
@@ -13,8 +19,11 @@ from ..utils import (
     xpath_element,
     xpath_with_ns,
     find_xpath_attr,
+    orderedSet,
+    parse_duration,
     parse_iso8601,
     parse_age_limit,
+    strip_or_none,
     int_or_none,
     ExtractorError,
 )
@@ -124,15 +133,23 @@ class CBCIE(InfoExtractor):
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
+        title = self._og_search_title(webpage, default=None) or self._html_search_meta(
+            'twitter:title', webpage, 'title', default=None) or self._html_search_regex(
+                r'<title>([^<]+)</title>', webpage, 'title', fatal=False)
         entries = [
             self._extract_player_init(player_init, display_id)
             for player_init in re.findall(r'CBC\.APP\.Caffeine\.initInstance\(({.+?})\);', webpage)]
+        media_ids = []
+        for media_id_re in (
+                r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"',
+                r'<div[^>]+\bid=["\']player-(\d+)',
+                r'guid["\']\s*:\s*["\'](\d+)'):
+            media_ids.extend(re.findall(media_id_re, webpage))
         entries.extend([
             self.url_result('cbcplayer:%s' % media_id, 'CBCPlayer', media_id)
-            for media_id in re.findall(r'<iframe[^>]+src="[^"]+?mediaId=(\d+)"', webpage)])
+            for media_id in orderedSet(media_ids)])
         return self.playlist_result(
-            entries, display_id,
-            self._og_search_title(webpage, fatal=False),
+            entries, display_id, strip_or_none(title),
             self._og_search_description(webpage))
 
 
@@ -200,33 +217,90 @@ class CBCWatchBaseIE(InfoExtractor):
         'media': 'http://search.yahoo.com/mrss/',
         'clearleap': 'http://www.clearleap.com/namespace/clearleap/1.0/',
     }
+    _GEO_COUNTRIES = ['CA']
+    _LOGIN_URL = 'https://api.loginradius.com/identity/v2/auth/login'
+    _TOKEN_URL = 'https://cloud-api.loginradius.com/sso/jwt/api/token'
+    _API_KEY = '3f4beddd-2061-49b0-ae80-6f1f2ed65b37'
+    _NETRC_MACHINE = 'cbcwatch'
+
+    def _signature(self, email, password):
+        data = json.dumps({
+            'email': email,
+            'password': password,
+        }).encode()
+        headers = {'content-type': 'application/json'}
+        query = {'apikey': self._API_KEY}
+        resp = self._download_json(self._LOGIN_URL, None, data=data, headers=headers, query=query)
+        access_token = resp['access_token']
+
+        # token
+        query = {
+            'access_token': access_token,
+            'apikey': self._API_KEY,
+            'jwtapp': 'jwt',
+        }
+        resp = self._download_json(self._TOKEN_URL, None, headers=headers, query=query)
+        return resp['signature']
 
     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,
-        })
+        for _ in range(2):
+            try:
+                result = self._download_xml(url, video_id, headers={
+                    'X-Clearleap-DeviceId': self._device_id,
+                    'X-Clearleap-DeviceToken': self._device_token,
+                })
+            except ExtractorError as e:
+                if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401:
+                    # Device token has expired, re-acquiring device token
+                    self._register_device()
+                    continue
+                raise
         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'<device><type>web</type></device>')
-                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,
-                    })
+        if self._valid_device_token():
+            return
+        device = self._downloader.cache.load(
+            'cbcwatch', self._cache_device_key()) or {}
+        self._device_id, self._device_token = device.get('id'), device.get('token')
+        if self._valid_device_token():
+            return
+        self._register_device()
+
+    def _valid_device_token(self):
+        return self._device_id and self._device_token
+
+    def _cache_device_key(self):
+        email, _ = self._get_login_info()
+        return '%s_device' % hashlib.sha256(email.encode()).hexdigest() if email else 'device'
+
+    def _register_device(self):
+        result = self._download_xml(
+            self._API_BASE_URL + 'device/register',
+            None, 'Acquiring device token',
+            data=b'<device><type>web</type></device>')
+        self._device_id = xpath_text(result, 'deviceId', fatal=True)
+        email, password = self._get_login_info()
+        if email and password:
+            signature = self._signature(email, password)
+            data = '<login><token>{0}</token><device><deviceId>{1}</deviceId><type>web</type></device></login>'.format(
+                escape(signature), escape(self._device_id)).encode()
+            url = self._API_BASE_URL + 'device/login'
+            result = self._download_xml(
+                url, None, data=data,
+                headers={'content-type': 'application/xml'})
+            self._device_token = xpath_text(result, 'token', fatal=True)
+        else:
+            self._device_token = xpath_text(result, 'deviceToken', fatal=True)
+        self._downloader.cache.store(
+            'cbcwatch', self._cache_device_key(), {
+                'id': self._device_id,
+                'token': self._device_token,
+            })
 
     def _parse_rss_feed(self, rss):
         channel = xpath_element(rss, 'channel', fatal=True)
@@ -287,6 +361,11 @@ class CBCWatchBaseIE(InfoExtractor):
 class CBCWatchVideoIE(CBCWatchBaseIE):
     IE_NAME = 'cbc.ca:watch:video'
     _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})'
+    _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)
@@ -321,11 +400,12 @@ class CBCWatchVideoIE(CBCWatchBaseIE):
 
 class CBCWatchIE(CBCWatchBaseIE):
     IE_NAME = 'cbc.ca:watch'
-    _VALID_URL = r'https?://watch\.cbc\.ca/(?:[^/]+/)+(?P<id>[0-9a-f-]+)'
+    _VALID_URL = r'https?://(?:gem|watch)\.cbc\.ca/(?:[^/]+/)+(?P<id>[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': '38e815a-009e3ab12e4',
+            'id': '9673749a-5e77-484c-8b62-a1092a6b5168',
             'ext': 'mp4',
             'title': 'Customer (Dis)Service',
             'description': 'md5:8bdd6913a0fe03d4b2a17ebe169c7c87',
@@ -337,8 +417,8 @@ class CBCWatchIE(CBCWatchBaseIE):
             'skip_download': True,
             'format': 'bestvideo',
         },
-        'skip': 'Geo-restricted to Canada',
     }, {
+        # 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',
@@ -346,10 +426,72 @@ class CBCWatchIE(CBCWatchBaseIE):
             'description': 'Arthur, the sweetest 8-year-old aardvark, and his pals solve all kinds of problems with humour, kindness and teamwork.',
         },
         'playlist_mincount': 30,
-        'skip': 'Geo-restricted to Canada',
+    }, {
+        'url': 'https://gem.cbc.ca/media/this-hour-has-22-minutes/season-26/episode-20/38e815a-0108c6c6a42',
+        'only_matching': True,
     }]
 
     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)
+
+
+class CBCOlympicsIE(InfoExtractor):
+    IE_NAME = 'cbc.ca:olympics'
+    _VALID_URL = r'https?://olympics\.cbc\.ca/video/[^/]+/(?P<id>[^/?#]+)'
+    _TESTS = [{
+        'url': 'https://olympics.cbc.ca/video/whats-on-tv/olympic-morning-featuring-the-opening-ceremony/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+        video_id = self._hidden_inputs(webpage)['videoId']
+        video_doc = self._download_xml(
+            'https://olympics.cbc.ca/videodata/%s.xml' % video_id, video_id)
+        title = xpath_text(video_doc, 'title', fatal=True)
+        is_live = xpath_text(video_doc, 'kind') == 'Live'
+        if is_live:
+            title = self._live_title(title)
+
+        formats = []
+        for video_source in video_doc.findall('videoSources/videoSource'):
+            uri = xpath_text(video_source, 'uri')
+            if not uri:
+                continue
+            tokenize = self._download_json(
+                'https://olympics.cbc.ca/api/api-akamai/tokenize',
+                video_id, data=json.dumps({
+                    'VideoSource': uri,
+                }).encode(), headers={
+                    'Content-Type': 'application/json',
+                    'Referer': url,
+                    # d3.VideoPlayer._init in https://olympics.cbc.ca/components/script/base.js
+                    'Cookie': '_dvp=TK:C0ObxjerU',  # AKAMAI CDN cookie
+                }, fatal=False)
+            if not tokenize:
+                continue
+            content_url = tokenize['ContentUrl']
+            video_source_format = video_source.get('format')
+            if video_source_format == 'IIS':
+                formats.extend(self._extract_ism_formats(
+                    content_url, video_id, ism_id=video_source_format, fatal=False))
+            else:
+                formats.extend(self._extract_m3u8_formats(
+                    content_url, video_id, 'mp4',
+                    'm3u8' if is_live else 'm3u8_native',
+                    m3u8_id=video_source_format, fatal=False))
+        self._sort_formats(formats)
+
+        return {
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'description': xpath_text(video_doc, 'description'),
+            'thumbnail': xpath_text(video_doc, 'thumbnailUrl'),
+            'duration': parse_duration(xpath_text(video_doc, 'duration')),
+            'formats': formats,
+            'is_live': is_live,
+        }