djypanda / rasa3-for-botfront

Slightly change to fit 'botfront' frontend project.

Before GO !!!

This rasa fork can not run in standalone mode.
It's a part of botfront-for-rasa3 project.
It requests configurations from botfront-for-rasa3 backend service. And also use botfront backend service as a response and track server.

Run in docker

1. Get the sourcecode

git clone https://github.com/djypanda/rasa3-for-botfront.git
cd rasa3-for-botfront

2. Build the docker image If you run into some build errors due to the network connection problem (especially in cn), just run this command again.

docker build -t rasa3-for-bf:v0.1 -f docker/Dockerfile.botfront .

3. Check the image

docker images

You should get a output like this:

REPOSITORY             TAG       IMAGE ID       CREATED          SIZE
rasa3-for-bf           v0.1      3e604ad10384   43 seconds ago   2.33GB

Now rasa docker image is ready, go to botfront part for the next step;

Development Installation

1. Download sourcecode and create virtual environment:

git clone https://github.com/djypanda/rasa3-for-botfront.git
cd rasa3-for-botfront
python3 -m venv ./venv
source ./venv/bin/activate

2. Install poetry:

curl -sSL https://install.python-poetry.org | python3 -

3. Install rasa:

poetry install

4. Run rasa:

# Change this project id to the project id created in botfront frontend
export BF_PROJECT_ID=chitchat-9D1ct-aMp
# This URL is for rasa service to get endpoint/credential config from botfront service
export BF_URL=http://localhost:3000/graphql 
rasa run --cors * --debug --enable-api

Official rasa instruction goes here !!!