Connecting Rasa to a database to store responses - python-3.x

Question

I'm trying to connect my rasa open source chatbot to a local database and store response from the user. But I couldn't succeed in the process of integrating database to my chatbot. I tried using google but nothing seems to be working. I hope I'll get a solution from community members. Attaching the data for your reference.
nlu.yml
nlu:
- intent: painting_request
examples: |
- im interested in buying a painting
- i'm interested in buying a painting
- I'm interested in buying a painting
- interested in buying canvas
- id like to request a painting
- id like to buy a painting
- im looking for a painting that have a beautuful background
- i would like to get a sketch of my photograph
- i would like to get a sketch of my photo
- intent: art_type_entry
examples: |
- im looking for a [graphite](art_type) painting
- i'm looking for a [graphite](art_type) sketch
- i want to buy a [graphite](art_type) painting
- i want to buy a [graphite](art_type) sketch
- can i able to get a [graphite](art_type) art
- i want my picture to be drawn in [graphite](art_type)
- im looking for a [charcoal](art_type) painting
- i'm looking for a [charcoal](art_type) sketch
- i want to buy a [charcoal](art_type) painting
- i want to buy a [charcoal](art_type) sketch
- can i able to get a [charcoal](art_type) art
- i want my picture to be drawn in [charcoal](art_type)
- im looking for a [coloured pencil art](art_type) painting
- i'm looking for a [coloured pencil art](art_type) sketch
- i want to buy a [coloured pencil art](art_type) painting
- i want to buy a [coloured pencil art](art_type) sketch
- can i able to get a [coloured pencil art](art_type) art
- i want my picture to be drawn in [coloured pencil art](art_type)
- i want my picture to be drawn in [oil painting](art_type)
- im looking for a [oil painting](art_type) painting
- i'm looking for a [oil painting](art_type) sketch
- i want to buy a [oil painting](art_type) painting
- i want to buy a [oil painting](art_type) sketch
- can i able to get a [oil painting](art_type) art
- i want my picture to be drawn in [oil painting](art_type)
- intent: size_entry
examples: |
- size should be of [a4](size)
- size should be of [a3](size)
- size should be of [a2](size)
- size should be of [a1](size)
- painting should be of [a4](size)
- painting should be of [a3](size)
- painting should be of [a2](size)
- painting should be of [a1](size)
- intent: frame_entry
examples: |
- my frame should be of [standard](frame)
- my frame should be of [wooden](frame)
- my frame should be of [designer](frame)
- my canvas should be fitted with a [standard](frame)
- my canvas should be fitted with a [wooden](frame)
- my canvas should be fitted with a [designer](frame)
- frame should be of [standard](frame)
- frame should be of [Wooden](frame)
- frame should be of [designer](frame)
- intent: finishing_entry
examples: |
- the outer finishing should be done with [maple](finishing)
- the outer finishing should be done with [walnut](finishing)
- the outer finishing should be done with [black](finishing)
- the outer finishing should be done with [white](finishing)
- the outer finishing should be done with [glass](finishing)
- outer finishing should be done with [brass](finishing)
- outer finishing should be done with [maple](finishing)
- outer finishing should be done with [walnut](finishing)
- outer finishing should be done with [black](finishing)
- outer finishing should be done with [white](finishing)
- outer finishing should be done with [glass](finishing)
- outer finishing should be done with [brass](finishing)
- intent: orientation_entry
examples: |
- orientation of canvas should be in [landscape](orientation)
- orientation of canvas should be in [potrait](orientation)
- orientation of canvas should be in [designer choice](orientation)
rules.yml
rules:
- rule: activate painting form
steps:
- intent: painting_request
- action: painting_form
- active_loop: painting_form
- rule: submit form
condition:
- active_loop: painting_form
steps:
- action: painting_form
- active_loop: null
- action: utter_submit
- action: utter_slots_values
stories.yml
stories:
- story: confirmation
steps:
- intent: affirm
- action: utter_affirm
- story: stop form + continue
steps:
- intent: painting_request
- action: painting_form
- active_loop: painting_form
- intent: out_of_scope
- action: utter_ask_continue
- intent: affirm
- action: painting_form
- active_loop: null
- action: utter_submit
- action: utter_slots_values
- story: stop form + stop
steps:
- intent: painting_request
- action: painting_form
- active_loop: painting_form
- intent: out_of_scope
- action: utter_ask_continue
- intent: deny
- action: action_deactivate_loop
- active_loop: null
domain.yml
intents:
- greet
- goodbye
- affirm
- reject
- mood_great
- mood_unhappy
- bot_challenge
- askname
- help
- out_of_scope
- art_type_entry
- size_entry
- frame_entry
- finishing_entry
- orientation_entry
- painting_request:
use_entities: []
entities:
- art_type
- size
- frame
- finishing
- orientation
actions:
- validate_painting_form # From actions.py page
slots:
model:
type: text
influence_conversation: false
auto_fill: false
frame_size:
type: text
influence_conversation: false
auto_fill: false
frame_type:
type: text
influence_conversation: false
auto_fill: false
frame_orientation:
type: text
influence_conversation: false
auto_fill: false
frame_finishing:
type: text
influence_conversation: false
auto_fill: false
requested_slot:
type: text
influence_conversation: false
forms:
painting_form:
required_slots:
model:
- type: from_entity
entity: art_type
frame_size:
- type: from_entity
entity: size
frame_type:
- type: from_entity
entity: frame
frame_orientation:
- type: from_entity
entity: orientation
frame_finishing:
- type: from_entity
entity: finishing
responses:
utter_greet:
- text: "Hey! Welcome to My Website, how can I help you? "
- text: "Hi Welcome to My Website, how can I help you?"
utter_reject:
- text: "Sorry to hear it,How can I help you?"
- text: "Did that help you?"
utter_default:
- text: "I am not sure what you're aiming for."
- text: "I am sorry but I am not able to get you."
- text: "My appologies but I am not able to get you."
utter_happy:
- text: "Great, carry on!"
utter_cheer_up:
- text: "Sorry to hear that."
utter_affirm:
- text: "Nice to hear it"
utter_goodbye:
- text: "Bye"
utter_name:
- text: "I am a bot, powered by Rasa."
utter_bot_help:
- text: "I'm here to help you in customising your painting."
- text: "I'm powered by A.I here to helps you in guide to buy a beautiful painting with your desired customisation."
- text: "I'm here to gather information of how you want your painting to be."
utter_ask_continue:
- text: "Do you want to continue?"
utter_wrong_art_type:
- text: "Sorry we are unable to proceed with your art type, please try again"
utter_wrong_size_type:
- text: "Sorry we are unable to proceed with your size type, please try again"
utter_wrong_frame_type:
- text: "Sorry we are unable to proceed with your frame type, please try again"
utter_wrong_finishing_type:
- text: "Sorry we are unable to proceed with your finishing type, please try again"
utter_wrong_orientation_type:
- text: "Sorry we are unable to proceed with your orientation type, please try again"
utter_ask_model:
- text: "What is your desired art model"
utter_ask_frame_size:
- text: "Please enter your required size"
utter_ask_frame_type:
- text: "Please enter your desired frame type"
utter_ask_frame_finishing:
- text: "Please enter your deisired finishing"
utter_ask_frame_orientation:
- text: "Please enter orientation"
utter_submit:
- text: "Thanks for the information provided."
utter_slots_values:
- text: "I am going to run a search using the following parameters:\n
- art_type: {model}\n
- size: {frame_size}\n
- frame: {frame_type}\n
- finishing: {frame_finishing}\n
- orientation: {frame_orientation}"
session_config:
session_expiration_time: 60
carry_over_slots_to_new_session: true
actions.py
from database_connector import DataUpdate
from typing import Any, Text, Dict, List, Union
from rasa_sdk import Tracker
from rasa_sdk.executor import CollectingDispatcher
from rasa_sdk.forms import FormValidationAction
class PaintingFormValidation(FormValidationAction):
"""Example of a form validation action."""
def name(self) -> Text:
return "validate_painting_form"
def slot_mappings(self) -> Dict[Text, Union[Dict, List[Dict]]]:
return{
"model": [self.from_entity(entity='art_type', intent='art_type_entry')],
"frame_size": [self.from_entity(entity='size', intent='size_entry')],
"frame_type": [self.from_entity(entity='frame', intent='frame_entry')],
"frame_finishing": [self.from_entity(entity='finishing', intent='finishing_entry')],
"frame_orientation": [self.from_entity(entity='orientation', intent='orientation_entry')]
}
#staticmethod
def art_type_db() -> List[Text]:
"""Database of supported cuisines."""
return [
"graphite",
"charcoal",
"sketching",
"oilPainting",
"colored pencil"
]
#staticmethod
def size_db() -> List[Text]:
"""Database of supported sizes"""
return [
"a1",
"a2",
"a3",
"a4"
]
#staticmethod
def frame_db() -> List[Text]:
"""Database of supported frame types"""
return [
"no frame",
"standard",
"designer"
]
#staticmethod
def finishing_db() -> List[Text]:
"""Database of supported finishing"""
return [
"maple",
"wood",
"metal",
"glass",
"walnut"
]
#staticmethod
def orientation_db() -> List[Text]:
"""Database of supported orientations."""
return [
"landscape",
"potrait",
"designer choice"
]
#staticmethod
def is_int(string: Text) -> bool:
"""Check if a string is an integer."""
try:
int(string)
return True
except ValueError:
return False
def validate_model(
self,
value: Text,
dispatcher: CollectingDispatcher,
tracker: Tracker,
domain: Dict[Text, Any],
) -> Dict[Text, Any]:
"""Validate art type value."""
if value.lower() in self.art_type_db():
# validation succeeded, set the value of the "art_type" slot to value
return {"model": value}
else:
dispatcher.utter_message(response="utter_wrong_art_type")
# validation failed, set this slot to None, meaning the
# user will be asked for the slot again
return {"model": None}
def validate_frame_size(
self,
value: Text,
dispatcher: CollectingDispatcher,
tracker: Tracker,
domain: Dict[Text, Any],
) -> Dict[Text, Any]:
"""Validate size value."""
if value.lower() in self.size_db():
# validation succeeded, set the value of the "size" slot to value
return {"frame_size": value}
else:
dispatcher.utter_message(response="utter_wrong_size_type")
# validation failed, set this slot to None, meaning the
# user will be asked for the slot again
return {"frame_size": None}
def validate_frame_type(
self,
value: Text,
dispatcher: CollectingDispatcher,
tracker: Tracker,
domain: Dict[Text, Any],
) -> Dict[Text, Any]:
"""Validate frame type value."""
if value.lower() in self.frame_db():
# validation succeeded, set the value of the "frame" slot to value
return {"frame_type": value}
else:
dispatcher.utter_message(response="utter_wrong_frame_type")
# validation failed, set this slot to None, meaning the
# user will be asked for the slot again
return {"frame_type": None}
def validate_frame_finishing(
self,
value: Text,
dispatcher: CollectingDispatcher,
tracker: Tracker,
domain: Dict[Text, Any],
) -> Dict[Text, Any]:
"""Validate finishing value."""
if value.lower() in self.finishing_db():
# validation succeeded, set the value of the "finishing" slot to value
return {"frame_finishing": value}
else:
dispatcher.utter_message(response="utter_wrong_finishing_type")
# validation failed, set this slot to None, meaning the
# user will be asked for the slot again
return {"frame_finishing": None}
def validate_frame_orientation(
self,
value: Text,
dispatcher: CollectingDispatcher,
tracker: Tracker,
domain: Dict[Text, Any],
) -> Dict[Text, Any]:
"""Validate orientation value."""
if value.lower() in self.orientation_db():
# validation succeeded, set the value of the "orientation" slot to value
return {"frame_orientation": value}
else:
dispatcher.utter_message(response="utter_wrong_orientation_type")
# validation failed, set this slot to None, meaning the
# user will be asked for the slot again
return {"frame_orientation": None}
def submit(
self,
dispatcher:CollectingDispatcher,
tracker: Tracker,
domain: Dict[Text, Any]
) -> List[Dict]:
dispatcher.utter_message(template="utter_submit")
DataUpdate(tracker.get_slot("model"), tracker.get_slot("frame_size"), tracker.get_slot("frame_type"), tracker.get_slot("frame_finishing"), tracker.get_slot("frame_orientation"))
dispatcher.utter_message("Your response has been loaded.")
return []
database_connector.py
import psycopg2
def DataUpdate(art_type_entry, size_entry, frame_entry, finishing_entry, orientation_entry):
'''
Pushes Descriptive Analytics Data to the Database
'''
db = psycopg2.connect(
host="localhost",
database="Rasa_Chatbot",
user="postgres",
password="postgres"
)
mycursor = db.connect()
postgres_insert_query = """INSERT INTO rasainfo(model,size,type,orientation,finishing) VALUES (%s,%s,%s,%s,%s);""".format(art_type_entry,size_entry, frame_entry, finishing_entry, orientation_entry)
mycursor.execute(postgres_insert_query)
db.commit()
print("Record inserted successfully into table")

Answer 1

You need to define this in your endpoints.yml
tracker_store:
type: SQL
dialect: "postgresql" # the dialect used to interact with the db
url: "postgres"
db: "rasa" # path to your db
username: # username used for authentication
password: # password used for authentication
query: # optional dictionary to be added as a query string to the connection URL
driver: my-driver