Implement bellagrams

README.md

@@ -5,6 +5,7 @@ using [HTTP API](https://gitea.brno.mraveniste.cc/turbotraktor/ladylilia.com/src
 
 ## Supported commands
 
+* `!bellagram`, `!bellapics`, `!instabella`, `!instagram` - posts a link to a random Instagram picture of Bella
 * `!lastquote` - requests the most recent quote  
 * `!findquote PATTERN` - searches for quotes matching `PATTERN` and in case multiple matches are found, requests one of them randomly  
     - `PATTERN` has to be at least 3 characters long and it can be enclosed in double quotes in case it contains spaces

bot.py

@@ -1,6 +1,7 @@
 import configparser
 import logging
 import os
+import random
 import re
 import string
 
@@ -8,6 +9,7 @@ import dateutil.parser
 import irc.bot
 import requests
 
+from instagram import Instagram
 from twitch import Twitch
 
 
@@ -53,6 +55,10 @@ class TwitchBot(irc.bot.SingleServerIRCBot):
             (QUOTE_REMOVED_PATTERN, self.remove_quote),
         ]
         self.commands = [
+            (re.compile(r'^!bellagram$'), self.bellagram),
+            (re.compile(r'^!bellapics$'), self.bellagram),
+            (re.compile(r'^!instabella$'), self.bellagram),
+            (re.compile(r'^!instagram$'), self.bellagram),
             (re.compile(r'^!lastquote$'), self.last_quote),
             (re.compile(r'^!findquote\s+(?P<q>")?(?P<filter>.+)(?(q)")$'), self.find_quote),
             (re.compile(r'^!sync(\s+(?P<since>.+))?$'), self.sync),
@@ -65,10 +71,19 @@ class TwitchBot(irc.bot.SingleServerIRCBot):
         self.master_user_id = config['Twitch'].getint('master_user_id')
         self.api_url = config['Quotes'].get('api_url')
         self.api_key = config['Quotes'].get('api_key')
+        self.bellagrams = self._get_bellagrams()
         log.info('Connecting to %s:%d', self.server, self.port)
         super(TwitchBot, self).__init__([(self.server, self.port, self.token)],
                                         self.nickname, self.nickname)
 
+    def _get_bellagrams(self):
+        username = config['Instagram'].get('username')
+        keywords = config['Instagram'].get('keywords').split(',')
+        media = Instagram(username, log).get_media()
+        if not media:
+            return None
+        return [m for m in media if [k for k in keywords if k.lower() in m['text'].lower()]]
+
     def on_welcome(self, connection, event):
         connection.cap('REQ', ':twitch.tv/membership')
         connection.cap('REQ', ':twitch.tv/tags')
@@ -111,6 +126,13 @@ class TwitchBot(irc.bot.SingleServerIRCBot):
         r.raise_for_status()
         return r
 
+    def bellagram(self, connection, tags, **kwargs):
+        if not self.bellagrams:
+            msg = 'Sorry @{0}, couldn\'t get any media from Instagram'.format(tags['display-name'])
+        else:
+            msg = random.choice(self.bellagrams)['url']
+        connection.privmsg(self.channel, msg)
+
     def last_quote(self, connection, tags, **kwargs):
         try:
             quotes = self.get(dict(

instagram.py

@@ -0,0 +1,83 @@
+import hashlib
+import json
+import re
+
+import requests
+
+
+BASE_URL = 'https://www.instagram.com'
+
+QUERY_HASH = '42323d64886122307be10013ad2dcc44'
+
+SHARED_DATA = re.compile(r'window\._sharedData = (\{.*\});</script>')
+
+
+class Instagram(object):
+
+    def __init__(self, username, log=None):
+        self.username = username
+        self.log = log
+        shared_data = self._get_shared_data()
+        try:
+            self.user_id = shared_data['entry_data']['ProfilePage'][0]['graphql']['user']['id']
+            self.rhx_gis = shared_data['rhx_gis']
+        except (IndexError, KeyError, TypeError):
+            self.user_id = None
+            self.rhx_gis = None
+
+    def _get_shared_data(self):
+        r = requests.get('{0}/{1}'.format(BASE_URL, self.username))
+        if not r.ok:
+            return None
+        m = SHARED_DATA.search(r.text)
+        if m:
+            return json.loads(m.group(1))
+        return None
+
+    def _get_media(self):
+        def request(count, cursor):
+            variables = json.dumps(dict(
+                id=self.user_id,
+                first=count,
+                after=cursor))
+            gis = '{0}:{1}'.format(self.rhx_gis, variables)
+            gis = hashlib.md5(gis.encode('UTF-8')).hexdigest()
+            url = '{0}/graphql/query'.format(BASE_URL)
+            params = dict(
+                query_hash=QUERY_HASH,
+                variables=variables)
+            r = requests.get(url, params=params, headers={'X-Instagram-GIS': gis})
+            r.raise_for_status()
+            return r.json()
+        result = []
+        try:
+            count = 50
+            cursor = ''
+            while True:
+                data = request(count, cursor)
+                data = data['data']['user']['edge_owner_to_timeline_media']
+                for edge in data['edges']:
+                    result.append(dict(
+                        type=edge['node']['__typename'].split('Graph')[1],
+                        text=edge['node']['edge_media_to_caption']['edges'][0]['node']['text'],
+                        url='{0}/p/{1}'.format(BASE_URL, edge['node']['shortcode']),
+                        display_url=edge['node']['display_url']))
+                cursor = data['page_info']['end_cursor']
+                if not cursor:
+                    break
+        except KeyError:
+            return None
+        else:
+            return result
+
+    def get_media(self):
+        if not self.user_id:
+            return None
+        try:
+            result = self._get_media()
+        except requests.exceptions.HTTPError as e:
+            if self.log:
+                self.log.error('Failed to retrieve media: %s', str(e))
+            return None
+        else:
+            return result

settings.cfg.example

@@ -4,6 +4,10 @@ port = 6667
 nickname = spooky_lurker
 channel = lilialil
 
+[Instagram]
+username = lilialovesgames
+keywords = bella,teeny,kitty,cat,😹,😻,🐱,🐈
+
 [Twitch]
 api_url = https://api.twitch.tv/v5
 client_id = __TWITCH_CLIENT_ID__