Making Hangouts Chat bots with Python

Introduction:
Prior today at Google I/O, the Hangouts Chat group (counting yours really) conveyed an included chat on the bot structure for Hangouts Chat (talk video here [~40 mins]). On the off chance that you missed it a while back, Google propelled the new Hangouts Chat application for G Suite clients (yet not purchaser Gmail clients as of now). This cutting edge coordinated effort stage has a few key highlights not accessible in "exemplary Hangouts," including the capacity to make talk rooms, search, and the capacity to connect records from Google Drive. Anyway for engineers, the greatest new element is a bot structure and API enabling designers to make bots that interface with clients inside "spaces," i.e., talk rooms or direct messages (DMs).
Before beginning with bots, how about we audit average API use where Python Training in Bangalore your application must be approved for information get to, i.e., OAuth2. When consent is in all actuality, your application calls the API, the API benefits the solicitation and reacts back with the ideal outcomes, as showed here:
Bot engineering is very extraordinary. With bots, perceive that solicitations originate from clients in talk rooms or DMs. Clients direct messages towards a bot, at that point Hangouts Chat transfers those solicitations to the bot. The bot plays out all the vital handling (conceivably calling different APIs and administrations), examines the reaction, and returns it to Chat, which thus, shows the outcomes to clients. Here's the condensed bot work process:
A key take away from this discourse is that regularly in your applications, you call APIs. For bots, Hangouts Chat calls you. Hence, the administration is incredibly adaptable for designers: you can make bots utilizing basically any language (not simply Python), utilizing your pile of decision, and facilitated on any open or private cloud. The main prerequisite is that as long as Hangouts Chat would http be able to POST to your bot, you're ready.
Moreover, on the off chance that you believe there's something missing in the outline since you don't see OAuth2 nor API calls, you'd likewise be right. Search for both in a future post where I center around offbeat reactions from Hangouts Chat bots. Python Training will give a vast knowledge about this topic.
Types and Bot Preparing:
So what does the payload resemble when your bot gets a call from Hangouts Chat? The primary thing your bot needs to do is to decide the kind of occasion got. It would then be able to process the solicitation dependent on that occasion type, lastly, return a fitting JSON payload back to Hangouts Chat to render inside the space. There are presently four occasion types in Hangouts Chat:
• ADDED_TO_SPACE
• REMOVED_FROM_SPACE
• MESSAGE
• CARD_CLICKED
The main pair is for when a bot is added to or expelled from a space. In the principal case, the bot would almost certainly send an invite message like, "A debt of gratitude is in order for adding me to this room," and maybe give a few guidelines on the most proficient method to speak with the bot. At the point when a bot is expelled from a space, it can never again speak with talk room members, so there's no reaction message sent for this situation; the probably activity here is log that the bot was expelled from the space.
The third message type is likely the most widely recognized situation, where a bot has been added to a room, and now a human client is sending it a solicitation. The other regular sort would be the last one. Instead of a message, this occasion happens when clients click on a UI card component. The bot's activity here is to summon the "callback" related with the card component clicked. Various things can occur here: the bot can restore an instant message (or nothing by any means), a UI card can be refreshed, or another card can be returned.
All of what we depicted above is acknowledged in the pseudocode underneath (regardless of whether you execute in Python or some other upheld language) and accommodating in setting you up to audit the official documentation:
def process_event(req, rsp):
occasion = json.loads(req['body']) # occasion got
in the event that event['type'] == 'REMOVED_FROM_SPACE':
# no reaction as bot expelled from room
return
elif event['type'] == 'ADDED_TO_SPACE':
# bot added to room; send welcome message
msg = {'text': 'A debt of gratitude is in order for adding me to this room!'}
elif event['type'] == 'MESSAGE':
# message got during typical activities
msg = respond_to_message(event['message']['text'])
elif event['type'] == 'CARD_CLICKED':
# result from client click on card UI
activity = event['action']
msg = respond_to_click( action['actionMethodName'], action['parameters'])
else:
return
rsp. Send(json.dumps(msg))

Notwithstanding usage language, regardless you have to make the fundamental changes to keep running on the facilitating stage you pick. Our model uses Google App Engine (GAE) for facilitating; underneath you'll see the undeniable similitudes between our real bot application and this pseudocode. For more details refer Python Training in Bangalore.