Home.md

Welcome to our Anwendungsprojekt 22-23 Wiki.

Overview

• Introduction and Motivation
• Getting Started
• How it Works
• Sensors

Introduction and Motivation

Study apps can be helpful for the successful conduct of medical and psychological studies.
Especially the use of sensors, for example from smartphones or wearables, can provide useful data to support the study. Therefore, the development of such apps should be as simple as possible in order to be able to use them without great effort.

That's why we develop a smart sensing library, which can be used to easily implement cross-platform apps with Flutter. This library will be able to track sensors from the smartphone and some wearables, collect the sensor data and save it. Additionally, it will be possible to either get live data or store and process the data.

All further information, about how you can use and test the library and a demo app can be found in the smart-sensing-library repository.

Development hints

Keeping Pigeon code up to date

Pigeon is a plugin that generates API code for all three of our platforms (Dart, Android, iOS). The API code is defined in the ./pigeons directory. If the content changes, run the following command to keep the code up to date.

On Windows run:

.\run_pigeon.ps1

On Linux/macOS run:

bash run_pigeon.sh

Working on Android

We use Android Studio to write code for our Android sensor implementations. In order for Android Studio to recognize all dependencies, it is mandatory to open the ./example/android/ directory with Android Studio. If any other directory is opened, the dependencies won't be recognized.

Linting

We use Ktlint to lint our Kotlin code.

For a simple check, run:

./gradlew ktlintCheck

For Ktlint to format the code automatically, run:

./gradlew ktlintFormat

Working on iOS

For the development of the iOS platform code, which is written in Swift, we use Xcode. In order for Xcode to recognize all dependencies and correctly set up the project structure, it is mandatory to open the ./example/ios/ directory with Xcode.

Linting

To lint the Swift code we use Swiftlint (see here for installation guide). If Swiftlint is installed successfully, you can manually check the Swift files by running

swiftlint

and correct fixable linting issues with

swiftlint --fix

For further configuration of the swiftlint command, it is recommended to use a .swiftlint.yml file, which already exists in this repository and can be extended.

---

If you want Swiftlint to be automatically run by Xcode, you need to set up a new "Run script phase" in the "Build phases" of your target and execute the swiftlint command there, as described here.

How it Works

Software Architecture

This section covers the software architecture. It is split into two core components:
Smart Sensing Library and the Sensing Plugin.

The reason for this is that we have one component to read and standardize sensor data in a platform-independent way, while the other component represents most of the smart features like storage and filtering of values.

More information can be found in the software design document in Documentation.

Smart Sensing Library

Architectur

Class Diagram