This repository contains several components: api, backend, frontend, genie_script, docs and files. Some of them comes from previous teams, and our team has added more functions. During sprint1, we made some significant modifications mainly in both frontend and solved bugs to make the whole front application work smoothly and meet the client's expectation. For back end, at this stage, we mainly focus on the datebase deployment using ASW EC2.
Here is the brief introduction of each part of our project.
"api" is based on the Apache Tomcat server. It not only contains the RESTful API for handling the HTTP requests and includes the socket server listener for listening the data from Genie Script program. What is more, it also can do the DML statements on our MySQL database. We add more APIs, fix some bugs, make it deal with more types of data and so on to make the server work well.
"backend" is a docker file including the Adminer, MySQL database (and the SQL script) and Tomcat server (a .war file generated from api). And as we said before, all the backend resources are now deployed on the AWS EC2 successfully.
"frontend" is the mobile app frontend using flutter which can be deployed on both Android and iOS devices. We modify a lot of interfaces and add more functions in the frontend, and it is similar to the prototype.
"genie_script" is a TCP client that can upload certain type files which are used to add or update the data in MySQL database. Since we cannot use the Genie database API directly at present, this is an approach to add or update data in our own MySQL database.
"docs" includes the documentations from previous teams and the documentations of API.
"files" includes sample files which can be uploaded by Genie Script program. In details, there are some exported sample files from Genie software, some files for uploading certain modules, and the pdf report files which can also be handed out in each appointment by our program.
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes. See deployment for notes on how to deploy the project on a live system.
Preferably linux system, the following configuration has been tested, though other configs will most likely work with minimal change.
git version 2.21.0
Ubuntu 18.04
Docker version 19.03.8, build afacb8b
docker-compose version 1.25.5, build 8a1c60f6
It is possible to run this on a mac or windows version of docker.
The following will assume a linux server.
Installing prerequisites:
Follow your platform specific install for docker compose and docker
Getting the repo:
git clone {remote url}
Navigate to backend directory:
cd {projectname}/backend/
Optionally change the clinic name in the ~/{projectname}/backend/.env file to name the clinics appropriately.
Spin up server use the following command:
docker-compose up -d
This will install and run each container, it may take some time to download images. Please ensure that the ports specified in docker-compose.yml are not being used up by other processes on your server.
Now we need to seed the database server, this needs to be done manually, as we want the database to remain persistent from now on.
First we need to find what the hash of the running container for the database is. On each system, the container id will be different. If you restart the container, the id will change.
docker ps
Take note of the first few numbers of the container id.
Next we need to actually seed the database.
cd db/seed/
docker cp db_with_data.sql {containerid}:/db_with_data.sql
The file is inside the running container, we now execute a bash inside the container.
docker exec -it {containerid} bash
Seed the db (while in container bash)
mariadb -u root -p < db_with_data.sql
We just need to type in the password, which by default is root. (These settings are specified in docker-compose.yml)
Can also check database we create
mariadb -u root -p
Enter the default password: root (These settings are specified in docker-compose.yml)
Show databases:
SHOW DATABASES;
And you can check the data in the database:
USE medsec;
SHOW TABLES;
SELECT * FROM User;
Exit MariaDB:
exit
Exit docker bash:
exit
Confirm setup by typing in localhost/adminer.php into a browser, then login using:
server: DB
username: root
password: root
(These settings are specified in docker-compose.yml)
You should be able to see a database named medsec. You can also manage the database using this interface if needed.
In order to deploy the app, configure your router settings such that the server exposes the api port on the internet. To find the port, view the docker-compose yml port mapping.
Edit the ~/{projectname}/frontend/lib/util/serverDetails.dart file to reflect your public ip address.
After deploye the backend on your own computer, you may want to deploy it on the AWS EC2 just like we did. For more information about AWS EC2 deployment, please see: AWS Deployment
While the api will be automatically deployed via docker-compose, the following details are about how to rebuild the api code, should changes be required.
The code is built using Maven and the documentation for building can be found in the api folder README.md file.
Essentially mvn clean install will build a .war file in the target directory. To deploy, move this file into ~/{projectname}/backend/tomcat/webapps then rebuild the docker containers with:
docker-compose stop
docker-compose --build
docker-compose up -d
One important note is that the Java version Maven uses must match with the Java version Tomcat uses, the following configuration works without issues:
Apache Maven 3.6.3
Java version: 1.8
For more detailed instructions, please see the documentations from previous two teams and repeat those steps (while making sure to use the updated version of Java and Maven), and obviously no need to set up Tomcat because docker will handle that.
The frontend project is located in the ~/{projectname}/frontend directory, please refer to your platform specific guide on how to set up flutter on your system and import a project.
The following config was used successfully to build flutter code:
Flutter 1.17.1 • channel stable • https://github.com/flutter/flutter.git (The latest version for flutter is 1.20.4)
Engine • revision 6bc433c6b6
Tools • Dart 2.8.2
There are 2 more configurations you may want to change.
Firstly, the app points to an ip address and port which is located in a filed called ~/{projectname}/frontend/lib/util/serverDetails.dart - change the relevant fields to your server. Or you can skip this step, because we have deployed it using AWS EC2 and our server can keep running currently.
Secondly, the fcm token setup should be changed over to your account. Set up a firebase account, then follow the steps required to activate firebase for your app. This is a very involved process which included generating key files from firebase and changing configurations for both the android and ios subdirectories of the project. Consult the official documentation.
The new changes are mainly in the these files：google-services.jason, login.dart, register.dart, appointmentfile.dart, doctordetail.dart and so on. For more information, please see the README.md in the frontend file.
From the previous group: Wenkai Huang - wenkai.huang@hotmail.com Callum Dowling - callum.dowling@gmail.com
From the group Wombat: Lin Luo - linluo@student.unimelb.edu.au
Previous MH-Bilby team who built docker and part of frontend.
Previous medical-secretary team who built the genie script and api.