How do I extract entire user messages in a form?

I am building my RASA assistant for a personal project, and I need the RASA assistant to grab the entire user input and store it. However, I’m unsure how to do so.

On the form I made, the RASA assistant asks for a user e-mail subject input, but I don’t know how I’m going to extract the entire user message and store it as a slot considering that I can’t define a specific intent for the message the user is going to input.

Is there a work-around this issue? Or does RASA have a “questionnaire” functionality that allows me to store whatever the user gives me?

P.S.: I know that there’s a functionality in actions.py (tracker.latest_message.get(‘text’)) to obtain the latest message, I just don’t know how to trigger the function since the user input can be anything at all as it is not tied to a specific intent.

1 Like

Hi @DARKSUPLA, welcome to the community.

You can use the Text Slot to store the entire user utterance in the slot. You can also use the Mapping Conditions to specify the context of the slot mapping.

Anyone else with a more specific answer? No matter how I use Mapping Conditions with Forms it seems that it can’t detect the message since the intent is very broad and could be anything…

Maybe the next Rasa version should allow a type of functionality where it facilitates just grabbing the user input without detecting any type of intent to trigger it as a Forms functionality? That’s something I’d look forward to working with.

1 Like

Thank you for your response, but I don’t think that’s what I’m looking for. If it helps, this person is having the same issue as me: How to take an input from the user and save it in a slot

Maybe they explained it better in their post. Could you please take a look at it and let me know if you know a work-around this issue?

@rasa_learner answered your question already but you can find a working example in the helpdesk-assistant. Take a look at the problem_description slot and how it works in the open_incident_form.

1 Like

Thank you, Greg. This might be what I’m looking for. If it works as I intend it, I’ll come back to your response and mark it as a solution.

@stephens Even after using text slot and mapping conditions it is not working that text is falling under nlu fallback only, any suggestion?

@DARKSUPLA Could you please share how did you solve it

I suspect you aren’t using a form. Take a look at how problem_description is collected in the example I pointed to above.