This repository contains university project which is a web application designed for improving interaction beetween university and students or teachers. It uses conversation with bot as main communication interface. Bot recognize questions about plan, grades etc. Student cannot ask about grades for all students and therefore teacher cannot ask about concrete student grades. Also question about consultation is only available for student. Project is divided into backend made in Spring framework for Java language and frontend made in Angular framework for JavaScript language. Backend side is designed as REST API and uses Postgres database for storing all neccessary data. Backend also uses docker because of easy way to set up developer environment. Frontend side is mostly created by us but we used outside technologies like Dialogflow and jqWidgets. Dialogflow is technology which use machine learning to recognize conversation context and necceserry information like role or action type while having conversation with user. We also used jqWidgets Angular 6 UI Components for displaying plan, scores.
This project require a few tool to run:
- Java 8
- Maven
- Docker
- Docker Compose
- Angular 6 with CLI
- NodeJS (npm)
If you are using Windows or Mac, you will additionally need:
- VirtualBox
- Docker Machine
We need to create virtual machine for docker:
# run this command only once
docker-machine create --driver virtualbox --virtualbox-cpu-count "1" --virtualbox-memory "2048" docker-bot
Remember, when you create docker-bot machine, it remains on you computer until you delete it. You need to run above command only once.
Once docker-machine is created, we should start it:
# run this command every time after restart of your computer
docker-machine start docker-bot
# WINDOWS : run this command and follow instructions
docker-machine env docker-bot
# MAC : run this command every time after opening new terminal
eval $(docker-machine env docker-bot)
To build project you need to run below command in these subdirectories: \BotRest
# run this command after any changes in application code
mvn clean install
Once you app is successfully built, you need to proceed to parrent directory and execute below command:
docker-compose up
# run these commands every time you want to restart an application
docker-compose build
docker-compose up -d
You can check IP address of you docker machine by command:
docker-machine ip docker-bot
You can access your app by typing in you browser:
<your docker-machine ip address>:8080
# for example: 192.168.99.100:8080
To build project you need to run below command in these subdirectories: \bot-front
npm install
ng serve
You can now use our application. To log in we prepared following test credentials:
Role: Teacher, Index: adasow, Password: test
Role: Student, Index: 253233, Password: test
The best date period for testing our application is 04.06.2018 - 10.06.2018.
Requirements:
- Postgres
psql -h <docker machine ip address> -p <database port> -U postgres
# for example: psql -h 192.168.99.100 -p 5432 -U postgres
Then you can type SQL directly to database.
- Bartłomiej Łazarczyk
- Jakub Kacorzyk