[scrippsnetworks:watch] Fix extraction (closes #14389)
authorSergey M․ <dstftw@gmail.com>
Sat, 14 Oct 2017 18:46:05 +0000 (01:46 +0700)
committerSergey M․ <dstftw@gmail.com>
Sat, 14 Oct 2017 18:57:43 +0000 (01:57 +0700)
youtube_dl/extractor/scrippsnetworks.py

index 597d6f543a362791a0af0d111156bdac79ac4270..30bb31d69195c1ae6047f17e54b15e928527d526 100644 (file)
 # coding: utf-8
 from __future__ import unicode_literals
 
-from .adobepass import AdobePassIE
+import datetime
+import json
+import hashlib
+import hmac
+import re
+
+from .common import InfoExtractor
+from .anvato import AnvatoIE
 from ..utils import (
-    int_or_none,
-    smuggle_url,
-    update_url_query,
+    urlencode_postdata,
+    xpath_text,
 )
 
 
-class ScrippsNetworksWatchIE(AdobePassIE):
+class ScrippsNetworksWatchIE(InfoExtractor):
     IE_NAME = 'scrippsnetworks:watch'
-    _VALID_URL = r'https?://watch\.(?:hgtv|foodnetwork|travelchannel|diynetwork|cookingchanneltv)\.com/player\.[A-Z0-9]+\.html#(?P<id>\d+)'
-    _TEST = {
-        'url': 'http://watch.hgtv.com/player.HNT.html#0256538',
+    _VALID_URL = r'''(?x)
+                    https?://
+                        watch\.
+                        (?P<site>hgtv|foodnetwork|travelchannel|diynetwork|cookingchanneltv)\.com/
+                        (?:
+                            player\.[A-Z0-9]+\.html\#|
+                            show/(?:[^/]+/){2}
+                        )
+                        (?P<id>\d+)
+                    '''
+    _TESTS = [{
+        'url': 'http://watch.hgtv.com/show/HGTVE/Best-Ever-Treehouses/2241515/Best-Ever-Treehouses/',
         'md5': '26545fd676d939954c6808274bdb905a',
         'info_dict': {
-            'id': '0256538',
+            'id': '4173834',
             'ext': 'mp4',
-            'title': 'Seeking a Wow House',
-            'description': 'Buyers retiring in Palm Springs, California, want a modern house with major wow factor. They\'re also looking for a pool and a large, open floorplan with tall windows looking out at the views.',
-            'uploader': 'SCNI',
-            'upload_date': '20170207',
-            'timestamp': 1486450493,
+            'title': 'Best Ever Treehouses',
+            'description': "We're searching for the most over the top treehouses.",
+            'uploader': 'ANV',
+            'upload_date': '20170922',
+            'timestamp': 1506056400,
+        },
+        'params': {
+            'skip_download': True,
         },
-        'skip': 'requires TV provider authentication',
+        'add_ie': [AnvatoIE.ie_key()],
+    }, {
+        'url': 'http://watch.diynetwork.com/show/DSAL/Salvage-Dawgs/2656646/Covington-Church/',
+        'only_matching': True,
+    }, {
+        'url': 'http://watch.diynetwork.com/player.HNT.html#2656646',
+        'only_matching': True,
+    }]
+
+    _SNI_TABLE = {
+        'hgtv': 'hgtv',
+        'diynetwork': 'diy',
+        'foodnetwork': 'food',
+        'cookingchanneltv': 'cook',
+        'travelchannel': 'trav',
+        'geniuskitchen': 'geniuskitchen',
     }
+    _SNI_HOST = 'web.api.video.snidigital.com'
+
+    _AWS_REGION = 'us-east-1'
+    _AWS_IDENTITY_ID_JSON = json.dumps({
+        'IdentityId': '%s:7655847c-0ae7-4d9b-80d6-56c062927eb3' % _AWS_REGION
+    })
+    _AWS_USER_AGENT = 'aws-sdk-js/2.80.0 callback'
+    _AWS_API_KEY = 'E7wSQmq0qK6xPrF13WmzKiHo4BQ7tip4pQcSXVl1'
+    _AWS_SERVICE = 'execute-api'
+    _AWS_REQUEST = 'aws4_request'
+    _AWS_SIGNED_HEADERS = ';'.join([
+        'host', 'x-amz-date', 'x-amz-security-token', 'x-api-key'])
+    _AWS_CANONICAL_REQUEST_TEMPLATE = '''GET
+%(uri)s
+
+host:%(host)s
+x-amz-date:%(date)s
+x-amz-security-token:%(token)s
+x-api-key:%(key)s
+
+%(signed_headers)s
+%(payload_hash)s'''
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
-        channel = self._parse_json(self._search_regex(
-            r'"channels"\s*:\s*(\[.+\])',
-            webpage, 'channels'), video_id)[0]
-        video_data = next(v for v in channel['videos'] if v.get('nlvid') == video_id)
-        title = video_data['title']
-        release_url = video_data['releaseUrl']
-        if video_data.get('restricted'):
-            requestor_id = self._search_regex(
-                r'requestorId\s*=\s*"([^"]+)";', webpage, 'requestor id')
-            resource = self._get_mvpd_resource(
-                requestor_id, title, video_id,
-                video_data.get('ratings', [{}])[0].get('rating'))
-            auth = self._extract_mvpd_auth(
-                url, video_id, requestor_id, resource)
-            release_url = update_url_query(release_url, {'auth': auth})
-
-        return {
-            '_type': 'url_transparent',
-            'id': video_id,
-            'title': title,
-            'url': smuggle_url(release_url, {'force_smil_url': True}),
-            'description': video_data.get('description'),
-            'thumbnail': video_data.get('thumbnailUrl'),
-            'series': video_data.get('showTitle'),
-            'season_number': int_or_none(video_data.get('season')),
-            'episode_number': int_or_none(video_data.get('episodeNumber')),
-            'ie_key': 'ThePlatform',
+        mobj = re.match(self._VALID_URL, url)
+        site_id, video_id = mobj.group('site', 'id')
+
+        def aws_hash(s):
+            return hashlib.sha256(s.encode('utf-8')).hexdigest()
+
+        token = self._download_json(
+            'https://cognito-identity.us-east-1.amazonaws.com/', video_id,
+            data=self._AWS_IDENTITY_ID_JSON.encode('utf-8'),
+            headers={
+                'Accept': '*/*',
+                'Content-Type': 'application/x-amz-json-1.1',
+                'Referer': url,
+                'X-Amz-Content-Sha256': aws_hash(self._AWS_IDENTITY_ID_JSON),
+                'X-Amz-Target': 'AWSCognitoIdentityService.GetOpenIdToken',
+                'X-Amz-User-Agent': self._AWS_USER_AGENT,
+            })['Token']
+
+        sts = self._download_xml(
+            'https://sts.amazonaws.com/', video_id, data=urlencode_postdata({
+                'Action': 'AssumeRoleWithWebIdentity',
+                'RoleArn': 'arn:aws:iam::710330595350:role/Cognito_WebAPIUnauth_Role',
+                'RoleSessionName': 'web-identity',
+                'Version': '2011-06-15',
+                'WebIdentityToken': token,
+            }), headers={
+                'Referer': url,
+                'X-Amz-User-Agent': self._AWS_USER_AGENT,
+                'Content-Type': 'application/x-www-form-urlencoded; charset=utf-8',
+            })
+
+        def get(key):
+            return xpath_text(
+                sts, './/{https://sts.amazonaws.com/doc/2011-06-15/}%s' % key,
+                fatal=True)
+
+        access_key_id = get('AccessKeyId')
+        secret_access_key = get('SecretAccessKey')
+        session_token = get('SessionToken')
+
+        # Task 1: http://docs.aws.amazon.com/general/latest/gr/sigv4-create-canonical-request.html
+        uri = '/1/web/brands/%s/episodes/scrid/%s' % (self._SNI_TABLE[site_id], video_id)
+        datetime_now = datetime.datetime.utcnow().strftime('%Y%m%dT%H%M%SZ')
+        date = datetime_now[:8]
+        canonical_string = self._AWS_CANONICAL_REQUEST_TEMPLATE % {
+            'uri': uri,
+            'host': self._SNI_HOST,
+            'date': datetime_now,
+            'token': session_token,
+            'key': self._AWS_API_KEY,
+            'signed_headers': self._AWS_SIGNED_HEADERS,
+            'payload_hash': aws_hash(''),
         }
+
+        # Task 2: http://docs.aws.amazon.com/general/latest/gr/sigv4-create-string-to-sign.html
+        credential_string = '/'.join([date, self._AWS_REGION, self._AWS_SERVICE, self._AWS_REQUEST])
+        string_to_sign = '\n'.join([
+            'AWS4-HMAC-SHA256', datetime_now, credential_string,
+            aws_hash(canonical_string)])
+
+        # Task 3: http://docs.aws.amazon.com/general/latest/gr/sigv4-calculate-signature.html
+        def aws_hmac(key, msg):
+            return hmac.new(key, msg.encode('utf-8'), hashlib.sha256)
+
+        def aws_hmac_digest(key, msg):
+            return aws_hmac(key, msg).digest()
+
+        def aws_hmac_hexdigest(key, msg):
+            return aws_hmac(key, msg).hexdigest()
+
+        k_secret = 'AWS4' + secret_access_key
+        k_date = aws_hmac_digest(k_secret.encode('utf-8'), date)
+        k_region = aws_hmac_digest(k_date, self._AWS_REGION)
+        k_service = aws_hmac_digest(k_region, self._AWS_SERVICE)
+        k_signing = aws_hmac_digest(k_service, self._AWS_REQUEST)
+
+        signature = aws_hmac_hexdigest(k_signing, string_to_sign)
+
+        auth_header = ', '.join([
+            'AWS4-HMAC-SHA256 Credential=%s' % '/'.join(
+                [access_key_id, date, self._AWS_REGION, self._AWS_SERVICE, self._AWS_REQUEST]),
+            'SignedHeaders=%s' % self._AWS_SIGNED_HEADERS,
+            'Signature=%s' % signature,
+        ])
+
+        mcp_id = self._download_json(
+            'https://%s%s' % (self._SNI_HOST, uri), video_id, headers={
+                'Accept': '*/*',
+                'Referer': url,
+                'Authorization': auth_header,
+                'X-Amz-Date': datetime_now,
+                'X-Amz-Security-Token': session_token,
+                'X-Api-Key': self._AWS_API_KEY,
+            })['results'][0]['mcpId']
+
+        return self.url_result(
+            'anvato:anvato_scripps_app_web_prod_0837996dbe373629133857ae9eb72e740424d80a:%s' % mcp_id,
+            AnvatoIE.ie_key(), video_id=mcp_id)