Only Custom action names are different, one can be executed and the other cannot

Rasa Open Source
1

@nik202

It’s work if story settings like this

stories.yml (successful)

version: "3.1"

stories:
- story: start path
  steps:
  - intent: start_kmamiz_bot
  - action: utter_start_kmamiz_bot

- story: insights path
  steps:
  - intent: chose_insights
  - action: utter_chose_insights
  - intent: chose_type_of_insights
  - action: action_chose_type_of_insights
  - intent: chose_type_of_display
  - action: action_hello_world
  # - action: action_display_insight

Screenshot from 2023-04-27 18-21-00
Screenshot from 2023-04-27 18-21-00386×645 35.6 KB

And it isn’t work if story settings like this

stories.yml (failed)

version: "3.1"

stories:
- story: start path
  steps:
  - intent: start_kmamiz_bot
  - action: utter_start_kmamiz_bot

- story: insights path
  steps:
  - intent: chose_insights
  - action: utter_chose_insights
  - intent: chose_type_of_insights
  - action: action_chose_type_of_insights
  - intent: chose_type_of_display
  # - action: action_hello_world
  - action: action_display_insight

Screenshot from 2023-04-27 18-22-57
Screenshot from 2023-04-27 18-22-57386×645 34.1 KB

My action.py is setting like this action.py

from typing import Any, Text, Dict, List

from rasa_sdk import Action, Tracker
from rasa_sdk.executor import CollectingDispatcher

import requests
import json

class ActionHelloWorld(Action):

    def name(self) -> Text:
        return "action_hello_world"

    def run(self, dispatcher: CollectingDispatcher,
            tracker: Tracker,
            domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:
            
        if display == "text":
            dispatcher.utter_message(text=f"show in text")
        elif display == "image":
            dispatcher.utter_message(text=f"show in image")
        elif display == "url":
            dispatcher.utter_message(text=f"show in url")
        else:
            dispatcher.utter_message(text=f"the way you chose is not exist.")

        return []


class ActionDisplayInsight(Action):

    def name(self) -> Text:
        return "action_display_insight"

    def run(self, dispatcher: CollectingDispatcher,
            tracker: Tracker,
            domain: Dict[Text, Any]) -> List[Dict[Text, Any]]:

        display = tracker.get_slot('display')

        if display == "text":
            dispatcher.utter_message(text=f"show in text")
        elif display == "image":
            dispatcher.utter_message(text=f"show in image")
        elif display == "url":
            dispatcher.utter_message(text=f"show in url")
        else:
            dispatcher.utter_message(text=f"the way you chose is not exist.")

        return []



@nik202 I have no idea what’s going on!!!
The difference between them is the action name!!
One can trigger intent and can response, another can’t.
Maybe I miss something???


The whole project is in click me , and you need to open two terminal to run.
One input rasa run -m models --enable-api --cors "*" --debug
Another input rasa run actions
Then open index.html.

Thank you very much :smiling_face_with_three_hearts:

2

Why you have:

display = tracker.get_slot(‘display’)

in the second case on which it’s not working?

Check the code please and you will good to go, I guess.