K.S.R.V. Njord App

De Flut-ter versie.

How do I?

• Follow https://docs.flutter.dev/get-started/install
• Clone the repository
• Start working in your own branch (git checkout -b)
• Download the dependencies: flutter pub get
• Start an emulator or connect your phone
• Run the app: flutter run

For running the app on web, use flutter run -d chrome --web-browser-flag "--disable-web-security"

Code Generation

Code generation for JSON serialization and deserialization provides an automated approach to converting objects to and from JSON. It not only saves valuable development time by avoiding manual code creation but also ensures consistent and error-free data processing. This technique is particularly useful when data models evolve, as changes in serialization and deserialization code can be automatically implemented, ensuring consistency between the schema and code. Furthermore, code generation can apply optimizations that outperform manual implementations, while also ensuring type safety in strongly typed programming languages. When dealing with multiple platforms or languages, it also guarantees uniform serialization/deserialization logic. In short, code generation for JSON interactions enhances efficiency, reduces errors, and facilitates maintenance and debugging. We use json_serializable for code generation. Run the code generation with:

dart run build_runner watch

Automatically Running Static Analysis

When you open a Pull Request, a static analysis is automatically performed, which may take a few minutes. We recommend running this locally before opening a Pull Request.

• Run the following command in the root of this project to install this pre-push hook:

chmod +x run_static_analysis.sh; cp run_static_analysis.sh .git/hooks/pre-push; chmod 700 .git/hooks/pre-push

This runs all the static analysis tools we use.

Architecture

You can navigate between different pages using "named routes" (Navigator).

Project Structure

The folder structure is feature-based. This means that all code related to a feature is in a separate folder. This makes it easier to navigate the code and to know where to go to make changes. The code for features is further divided into folders each serving a specific role. The structure looks like this:

• lib
  • src
    • features
      • announcements
        • api
        • models
        • pages
        • widgets
      • events
        • ...
      • ...

A feature thus has an api, models, pages, and widgets folder. The api folder contains the code the app uses to fetch data from the server. The models folder contains the models used to fetch and send data to the server. The pages folder contains the code for the various pages associated with the feature. The widgets folder contains the code for widgets used on the various pages.

---

Continuous Integration / Continuous Deployment

How do I release a new version in the App Store / Google Play Store?

Builds are automatically created by Codemagic based on git tags. The tags are in turn automatically generated based on the version in pubspec.yaml on the main branch using Github Actions.

Suppose you have a branch with a new feature that you want to release. You have the branch feature/new-feature and you have increased the version in pubspec.yaml to 1.0.0+1. You create a pull request to the main branch and wait until it is approved and merged.

1. As soon as your branch is merged, a new git tag is automatically created based on the version in pubspec.yaml. The tag will be 1.0.0.
2. Codemagic will then create a new build based on the tag. This build will then become available in the App Store / Google Play Store for the internal test group.

Code Quality & Style

Code quality is checked by various tools. If there are errors, the pull request will fail and you will need to resolve the errors. You can also run the tools locally to detect errors. There are various checks that are performed as part of the CI/CD pipeline. These checks are performed on every push and pull request to main.

Dart Code Linter

We use the Dart Code Linter to analyze the code for errors and to format the code. You run the linter locally with:

$ flutter analyze

Dart Code Metrics

We use Dart Code Metrics to analyze the code for code smells.
You run dart code metrics locally with:

$ dcm analyze lib

This will analyze the lib folder.

Flutter Format

We use Flutter Format to format the code. You can run the formatter locally with:

$ dart format .

Acknowledgments

Dart Code Metrics

We are grateful to Dmitry Zhifarsky for generously providing us with a Dart Code Metrics Teams version free of charge. As a nonprofit, resources like these are invaluable in helping us achieve our mission.

Dmitry Zhifarsky's support plays a crucial role in our development process, enabling us to perform static analysis more effectively. His commitment to supporting nonprofit initiatives aligns closely with our organizational values and objectives.

For more information about Dmitry Zhifarsky and Dart Code Metrics, visit his website: https://dcm.dev/

---