

User Chat Conversation API

This endpoint handles the conversation and natural language processing from the front-end, Watson Assitant is used here to understand and classify the question user ask. The raw question acquired from front-end speech to text would be passed into Watson, Watson would then return the response type or response text back after processing, the backend would then process the information from Watson again to add user profile and finally send back to the client.

Get question response

POST - /chat

Use this request to send a question and get the response.

Query Parameter(s): none

Request Body:

{ 
    "userid": String,
    "content": String,
    "senderUsername": String
}

Key	Type	Description	Constraints	Default	Required
userid	string	MongoDb ObjectID of the profile the user is chatting (asking question) to	MongoDb ObjectID Type	N/A	Yes
content	string	The question content	N/A	N/A	Yes
senderUsername	string	The username of the sender, for record purposes. Leave null if no user is logged in.	N/A	N/A	Yes

---

Success

The history can be added.

Status Code: 200 OK

Response Body:

{
    "type": "<`text`, `audio`, `image` or `url`>",
    "content": "<String of content type>"
}

Key	Type	Description
type	string	One of text, audio, image or url
content	string	String of content type

Bad Request

Status Code: 400 Bad Request

Response Body:

{
    "error": "<Error message>"
}