Rename NetflixSession.parse_season_entry() to _parse_season_entry()...
[plugin.video.netflix.git] / resources / lib / NetflixSession.py
index 8c3172482c2348cd3f9c771de01f7c5d671758da..a4c48744ef7a933cd6e0cc18ef2dc8b9a0dc4929 100644 (file)
@@ -3,21 +3,18 @@
 # Module: NetflixSession
 # Created on: 13.01.2017
 
-import sys
 import os
-import base64
-import time
-import urllib
 import json
-import requests
-import platform
+from requests import session, cookies
+from urllib import quote, unquote
+from time import time
+from base64 import urlsafe_b64encode
+from bs4 import BeautifulSoup, SoupStrainer
+from utils import noop, get_user_agent_for_current_platform
 try:
    import cPickle as pickle
 except:
    import pickle
-from bs4 import BeautifulSoup, SoupStrainer
-from pyjsparser import PyJsParser
-from utils import noop
 
 class NetflixSession:
     """Helps with login/session management of Netflix users & API data fetching"""
@@ -27,15 +24,16 @@ class NetflixSession:
 
     urls = {
         'login': '/login',
-        'browse': '/browse',
-        'video_list_ids': '/warmer',
+        'browse': '/profiles/manage',
+        'video_list_ids': '/preflight',
         'shakti': '/pathEvaluator',
-        'profiles':  '/browse',
+        'profiles':  '/profiles/manage',
         'switch_profiles': '/profiles/switch',
         'adult_pin': '/pin/service',
         'metadata': '/metadata',
         'set_video_rating': '/setVideoRating',
-        'update_my_list': '/playlistop'
+        'update_my_list': '/playlistop',
+        'kids': '/Kids'
     }
     """:obj:`dict` of :obj:`str` List of all static endpoints for HTML/JSON POST/GET requests"""
 
@@ -103,9 +101,9 @@ class NetflixSession:
         self.log = log_fn
 
         # start session, fake chrome on the current platform (so that we get a proper widevine esn) & enable gzip
-        self.session = requests.session()
+        self.session = session()
         self.session.headers.update({
-            'User-Agent': self._get_user_agent_for_current_platform(),
+            'User-Agent': get_user_agent_for_current_platform(),
             'Accept-Encoding': 'gzip'
         })
 
@@ -151,13 +149,13 @@ class NetflixSession:
             :obj:`list` of :obj:`dict`
                 List of all the serialized data pulled out of the pagws <script/> tags
         """
-        scripts = page_soup.find_all('script', attrs={'src': None});
-        self.log('Trying sloppy inline data parser')
+        scripts = page_soup.find_all('script', attrs={'src': None})
+        self.log(msg='Trying sloppy inline data parser')
         inline_data = self._sloppy_parse_inline_data(scripts=scripts)
         if self._verfify_auth_and_profiles_data(data=inline_data) != False:
-            self.log('Sloppy inline data parsing successfull')
+            self.log(msg='Sloppy inline data parsing successfull')
             return inline_data
-        self.log('Sloppy inline parser failed, trying JS parser')
+        self.log(msg='Sloppy inline parser failed, trying JS parser')
         return self._accurate_parse_inline_data(scripts=scripts)
 
     def is_logged_in (self, account):
@@ -275,7 +273,7 @@ class NetflixSession:
         """
         payload = {
             'switchProfileGuid': profile_id,
-            '_': int(time.time()),
+            '_': int(time()),
             'authURL': self.user_data['authURL']
         }
 
@@ -283,14 +281,9 @@ class NetflixSession:
         if response.status_code != 200:
             return False
 
-        # fetch the index page again, so that we can fetch the corresponding user data
-        browse_response = self._session_get(component='browse')
-        only_script_tags = SoupStrainer('script')
-        browse_soup = BeautifulSoup(browse_response.text, 'html.parser', parse_only=only_script_tags)
         account_hash = self._generate_account_hash(account=account)
         self.user_data['guid'] = profile_id;
-        self._save_data(filename=self.data_path + '_' + account_hash)
-        return True
+        return self._save_data(filename=self.data_path + '_' + account_hash)
 
     def send_adult_pin (self, pin):
         """Send the adult pin to Netflix in case an adult rated video requests it
@@ -453,17 +446,22 @@ class NetflixSession:
         for key in self.video_list_keys:
             video_list_ids[key] = {}
 
+        # check if the list items are hidden behind a `value` sub key
+        # this is the case when we fetch the lists via POST, not via a GET preflight request
+        if 'value' in response_data.keys():
+            response_data = response_data['value']
+
         # subcatogorize the lists by their context
         video_lists = response_data['lists']
         for video_list_id in video_lists.keys():
             video_list = video_lists[video_list_id]
-            if video_list['context'] == 'genre':
-                video_list_ids['genres'].update(self.parse_video_list_ids_entry(id=video_list_id, entry=video_list))
-            elif video_list['context'] == 'similars' or video_list['context'] == 'becauseYouAdded':
-                video_list_ids['recommendations'].update(self.parse_video_list_ids_entry(id=video_list_id, entry=video_list))
-            else:
-                video_list_ids['user'].update(self.parse_video_list_ids_entry(id=video_list_id, entry=video_list))
-
+            if video_list.get('context', False) != False:
+                if video_list['context'] == 'genre':
+                    video_list_ids['genres'].update(self.parse_video_list_ids_entry(id=video_list_id, entry=video_list))
+                elif video_list['context'] == 'similars' or video_list['context'] == 'becauseYouAdded':
+                    video_list_ids['recommendations'].update(self.parse_video_list_ids_entry(id=video_list_id, entry=video_list))
+                else:
+                    video_list_ids['user'].update(self.parse_video_list_ids_entry(id=video_list_id, entry=video_list))
         return video_list_ids
 
     def parse_video_list_ids_entry (self, id, entry):
@@ -776,7 +774,7 @@ class NetflixSession:
                 'synopsis': video['synopsis'],
                 'regular_synopsis': video['regularSynopsis'],
                 'type': video['summary']['type'],
-                'rating': video['userRating']['average'],
+                'rating': video['userRating'].get('average', 0) if video['userRating'].get('average', None) != None else video['userRating'].get('predicted', 0),
                 'episode_count': season_info['episode_count'],
                 'seasons_label': season_info['seasons_label'],
                 'seasons_count': season_info['seasons_count'],
@@ -1076,14 +1074,29 @@ class NetflixSession:
                 }
             }
         """
-        seasons = {}
         raw_seasons = response_data['value']
+        videos = raw_seasons['videos']
+
+        # get art video key
+        video = {}
+        for key, video_candidate in videos.iteritems():
+            if not self._is_size_key(key):
+                video = video_candidate
+
+        # get season index
+        sorting = {}
+        for idx, season_list_entry in video['seasonList'].iteritems():
+            if self._is_size_key(key=idx) == False and idx != 'summary':
+                sorting[int(season_list_entry[1])] = int(idx)
+
+        seasons = {}
+
         for season in raw_seasons['seasons']:
             if self._is_size_key(key=season) == False:
-                seasons.update(self.parse_season_entry(season=raw_seasons['seasons'][season], videos=raw_seasons['videos']))
+                seasons.update(self._parse_season_entry(season=raw_seasons['seasons'][season], video=video, sorting=sorting))
         return seasons
 
-    def parse_season_entry (self, season, videos):
+    def _parse_season_entry (self, season, video, sorting):
         """Parse a season list entry e.g. rip out the parts we need
 
         Parameters
@@ -1109,16 +1122,6 @@ class NetflixSession:
                 }
             }
         """
-        # get art video key
-        video_key = ''
-        for key in videos.keys():
-            if self._is_size_key(key=key) == False:
-                video_key = key
-        # get season index
-        sorting = {}
-        for idx in videos[video_key]['seasonList']:
-            if self._is_size_key(key=idx) == False and idx != 'summary':
-                sorting[int(videos[video_key]['seasonList'][idx][1])] = int(idx)
         return {
             season['summary']['id']: {
                 'idx': sorting[season['summary']['id']],
@@ -1126,10 +1129,10 @@ class NetflixSession:
                 'text': season['summary']['name'],
                 'shortName': season['summary']['shortName'],
                 'boxarts': {
-                    'small': videos[video_key]['boxarts']['_342x192']['jpg']['url'],
-                    'big': videos[video_key]['boxarts']['_1280x720']['jpg']['url']
+                    'small': video['boxarts']['_342x192']['jpg']['url'],
+                    'big': video['boxarts']['_1280x720']['jpg']['url']
                 },
-                'interesting_moment': videos[video_key]['interestingMoment']['_665x375']['jpg']['url'],
+                'interesting_moment': video['interestingMoment']['_665x375']['jpg']['url'],
             }
         }
 
@@ -1242,6 +1245,11 @@ class NetflixSession:
           },
         }
         """
+        mpaa = ''
+        if episode.get('maturity', None) is not None:
+            if episode['maturity'].get('board', None) is not None and episode['maturity'].get('value', None) is not None:
+                mpaa = str(episode['maturity'].get('board', '').encode('utf-8')) + '-' + str(episode['maturity'].get('value', '').encode('utf-8'))
+
         return {
             episode['summary']['id']: {
                 'id': episode['summary']['id'],
@@ -1252,10 +1260,10 @@ class NetflixSession:
                 'title': episode['info']['title'],
                 'year': episode['info']['releaseYear'],
                 'genres': self.parse_genres_for_video(video=episode, genres=genres),
-                'mpaa': str(episode['maturity']['rating']['board']) + ' ' + str(episode['maturity']['rating']['value']),
+                'mpaa': mpaa,
                 'maturity': episode['maturity'],
                 'playcount': (0, 1)[episode['watched']],
-                'rating': episode['userRating']['average'],
+                'rating': episode['userRating'].get('average', 0) if episode['userRating'].get('average', None) != None else episode['userRating'].get('predicted', 0),
                 'thumb': episode['info']['interestingMoments']['url'],
                 'fanart': episode['interestingMoment']['_1280x720']['jpg']['url'],
                 'poster': episode['boxarts']['_1280x720']['jpg']['url'],
@@ -1277,8 +1285,9 @@ class NetflixSession:
         response = self._session_get(component='browse')
         return BeautifulSoup(response.text, 'html.parser')
 
-    def fetch_video_list_ids (self, list_from=0, list_to=50):
+    def fetch_video_list_ids_via_preflight (self, list_from=0, list_to=50):
         """Fetches the JSON with detailed information based on the lists on the landing page (browse page) of Netflix
+           via the preflight (GET) request
 
         Parameters
         ----------
@@ -1298,12 +1307,36 @@ class NetflixSession:
             'toRow': list_to,
             'opaqueImageExtension': 'jpg',
             'transparentImageExtension': 'png',
-            '_': int(time.time()),
+            '_': int(time()),
             'authURL': self.user_data['authURL']
         }
+
         response = self._session_get(component='video_list_ids', params=payload, type='api')
         return self._process_response(response=response, component=self._get_api_url_for(component='video_list_ids'))
 
+    def fetch_video_list_ids (self, list_from=0, list_to=50):
+        """Fetches the JSON with detailed information based on the lists on the landing page (browse page) of Netflix
+
+        Parameters
+        ----------
+        list_from : :obj:`int`
+            Start entry for pagination
+
+        list_to : :obj:`int`
+            Last entry for pagination
+
+        Returns
+        -------
+        :obj:`dict` of :obj:`dict` of :obj:`str`
+            Raw Netflix API call response or api call error
+        """
+        paths = [
+            ['lolomo', {'from': list_from, 'to': list_to}, ['displayName', 'context', 'id', 'index', 'length']]
+        ]
+
+        response = self._path_request(paths=paths)
+        return self._process_response(response=response, component='Video list ids')
+
     def fetch_search_results (self, search_str, list_from=0, list_to=10):
         """Fetches the JSON which contains the results for the given search query
 
@@ -1324,7 +1357,7 @@ class NetflixSession:
             Raw Netflix API call response or api call error
         """
         # properly encode the search string
-        encoded_search_string = urllib.quote(search_str)
+        encoded_search_string = quote(search_str)
 
         paths = [
             ['search', encoded_search_string, 'titles', {'from': list_from, 'to': list_to}, ['summary', 'title']],
@@ -1337,7 +1370,7 @@ class NetflixSession:
         response = self._path_request(paths=paths)
         return self._process_response(response=response, component='Search results')
 
-    def fetch_video_list (self, list_id, list_from=0, list_to=20):
+    def fetch_video_list (self, list_id, list_from=0, list_to=26):
         """Fetches the JSON which contains the contents of a given video list
 
         Parameters
@@ -1427,7 +1460,7 @@ class NetflixSession:
         payload = {
             'movieid': id,
             'imageformat': 'jpg',
-            '_': int(time.time())
+            '_': int(time())
         }
         response = self._session_get(component='metadata', params=payload, type='api')
         return self._process_response(response=response, component=self._get_api_url_for(component='metadata'))
@@ -1571,8 +1604,6 @@ class NetflixSession:
         })
 
         params = {
-            'withSize': True,
-            'materialize': True,
             'model': self.user_data['gpsModel']
         }
 
@@ -1816,10 +1847,10 @@ class NetflixSession:
             return False
 
         with open(filename) as f:
-            cookies = pickle.load(f)
-            if cookies:
-                jar = requests.cookies.RequestsCookieJar()
-                jar._cookies = cookies
+            _cookies = pickle.load(f)
+            if _cookies:
+                jar = cookies.RequestsCookieJar()
+                jar._cookies = _cookies
                 self.session.cookies = jar
             else:
                 return False
@@ -1852,24 +1883,7 @@ class NetflixSession:
         :obj:`str`
             Account data hash
         """
-        return base64.urlsafe_b64encode(account['email'])
-
-    def _get_user_agent_for_current_platform (self):
-        """Determines the user agent string for the current platform (to retrieve a valid ESN)
-
-        Returns
-        -------
-        :obj:`str`
-            User Agent for platform
-        """
-        if platform == 'linux' or platform == 'linux2':
-            return 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
-        elif platform == 'darwin':
-            return 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
-        elif platform == 'win32':
-            return 'Mozilla/5.0 (Windows NT 6.1; WOW64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
-        else:
-            return 'Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/56.0.2924.87 Safari/537.36'
+        return urlsafe_b64encode(account['email'])
 
     def _session_post (self, component, type='document', data={}, headers={}, params={}):
         """Executes a get request using requests for the current session & measures the duration of that request
@@ -1897,10 +1911,10 @@ class NetflixSession:
                 Contents of the field to match
         """
         url = self._get_document_url_for(component=component) if type == 'document' else self._get_api_url_for(component=component)
-        start = time.time()
+        start = time()
         response = self.session.post(url=url, data=data, params=params, headers=headers, verify=self.verify_ssl)
-        end = time.time()
-        self.log('[POST] Request for "' + url + '" took ' + str(end - start) + ' seconds')
+        end = time()
+        self.log(msg='[POST] Request for "' + url + '" took ' + str(end - start) + ' seconds')
         return response
 
     def _session_get (self, component, type='document', params={}):
@@ -1923,10 +1937,10 @@ class NetflixSession:
                 Contents of the field to match
         """
         url = self._get_document_url_for(component=component) if type == 'document' else self._get_api_url_for(component=component)
-        start = time.time()
+        start = time()
         response = self.session.get(url=url, verify=self.verify_ssl, params=params)
-        end = time.time()
-        self.log('[GET] Request for "' + url + '" took ' + str(end - start) + ' seconds')
+        end = time()
+        self.log(msg='[GET] Request for "' + url + '" took ' + str(end - start) + ' seconds')
         return response
 
     def _sloppy_parse_user_and_api_data (self, key, contents):
@@ -2073,16 +2087,17 @@ class NetflixSession:
             :obj:`dict` of :obj:`str`
                 Dict containing user, api & profile data
         """
-        inline_data = [];
+        inline_data = []
+        from pyjsparser import PyJsParser
         parser = PyJsParser()
         for script in scripts:
-            data = {};
+            data = {}
             # unicode escape that incoming script stuff
             contents = self._to_unicode(str(script.contents[0]))
             # parse the JS & load the declarations we´re interested in
             parsed = parser.parse(contents)
             if len(parsed['body']) > 1 and parsed['body'][1]['expression']['right'].get('properties', None) != None:
-                declarations = parsed['body'][1]['expression']['right']['properties'];
+                declarations = parsed['body'][1]['expression']['right']['properties']
                 for declaration in declarations:
                     for key in declaration:
                         # we found the correct path if the declaration is a dict & of type 'ObjectExpression'
@@ -2184,10 +2199,9 @@ class NetflixSession:
         important_fields = [
             'profileName',
             'isActive',
-            'isFirstUse',
-            'isAccountOwner'
+            'isAccountOwner',
+            'isKids'
         ]
-
         # values are accessible via dict (sloppy parsing successfull)
         if type(netflix_page_data) == dict:
             for profile_id in netflix_page_data.get('profiles'):
@@ -2197,7 +2211,7 @@ class NetflixSession:
                         profile.update({important_field: netflix_page_data['profiles'][profile_id]['summary'][important_field]})
                     avatar_base = netflix_page_data['nf'].get(netflix_page_data['profiles'][profile_id]['summary']['avatarName'], False);
                     avatar = 'https://secure.netflix.com/ffe/profiles/avatars_v2/320x320/PICON_029.png' if avatar_base == False else avatar_base['images']['byWidth']['320']['value']
-                    profile.update({'avatar': avatar})
+                    profile.update({'avatar': avatar, 'isFirstUse': False})
                     profiles.update({profile_id: profile})
             return profiles
 
@@ -2272,11 +2286,31 @@ class NetflixSession:
             :obj:`str` of :obj:`str
             ESN, something like: NFCDCH-MC-D7D6F54LOPY8J416T72MQXX3RD20ME
         """
-        esn = '';
+        # we generate an esn from device strings for android
+        import subprocess
+        try:
+            manufacturer = subprocess.check_output(["/system/bin/getprop", "ro.product.manufacturer"])
+            if manufacturer:
+                esn = 'NFANDROID1-PRV-'
+                input = subprocess.check_output(["/system/bin/getprop", "ro.nrdp.modelgroup"])
+                if not input:
+                    esn = esn + 'T-L3-'
+                else:
+                    esn = esn + input.strip(' \t\n\r') + '-'
+                esn = esn + '{:5}'.format(manufacturer.strip(' \t\n\r').upper())
+                input = subprocess.check_output(["/system/bin/getprop" ,"ro.product.model"])
+                esn = esn + input.strip(' \t\n\r').replace(' ', '=').upper()
+                self.log(msg='Android generated ESN:' + esn)
+                return esn
+        except OSError as e:
+            self.log(msg='Ignoring exception for non Android devices')
+
         # values are accessible via dict (sloppy parsing successfull)
         if type(netflix_page_data) == dict:
             return netflix_page_data.get('esn', '')
 
+        esn = ''
+
         # values are stored in lists (returned from JS parser)
         for item in netflix_page_data:
             if 'esnGeneratorModel' in dict(item).keys():
@@ -2297,5 +2331,5 @@ class NetflixSession:
         self.esn = self._parse_esn_data(netflix_page_data=netflix_page_data)
         self.api_data = self._parse_api_base_data(netflix_page_data=netflix_page_data)
         self.profiles = self._parse_profile_data(netflix_page_data=netflix_page_data)
-        self.log('Found ESN "' + self.esn + '" for platform "' + str(platform.system()) + '"')
+        self.log(msg='Found ESN "' + self.esn + '"')
         return netflix_page_data