Hi guys! So i wanted to find a way to store some information from my web application to my RASA BOT that does not affect the prediction. eg. if a person starts my bot’s widget on our website. It should send some information about the user like his unique_id and something else. So I want a way to store these parameters. Can someone please tell me a detailed way how to do it? I’m using the deprecated way of calling API’s right now! But i’ll go to REST soon. So please help Thanks @akelad @Juste @souvikg10
Hey @Akshit. Sorry for the late repy. Could you clarify a bit more what resulat are you looking for: is it about storing these detais in memory (as a slot) or do you want to persist such information in a database so that you could reuse the same information later?
Yes! So we pass parameters in our REST api call as
{“user_id”:default, “message”:“hi”}
So is it possible that I can send more parameters with this call and store those or maybe use them in custom action like I can get sender_id in custom action as
tracker.sender_id right? So whenever a user clicks on my widget, I’m planning to call an API to get information about him. So after click some basic information about our client goes to chat bot so user doesn’t have to enter those! like his name,user_id and some other information I have through which I need to make API calls to our DB to get info! So
2ND problem! Sorry for asking other one! but i’m kinda stuck on this right now!
So when a user says this:
q. How is Akshit doing today? I need to extract this name in nlu as an entity and I have added lot of names in my lookup table but it doesn’t seem to work! Where do we store that lookup table bcoz during running I don’t see any link to it. I have made a text file in which lot of names are there separated by a newline! So i need to extract this name, match it to a user_id and send it back to DB for information! check out this link as well ! Thanks a lot
[ my table is in data folder, a text file named as “name” which has all names separated by nextline]