A Custom MVVM‑DataBinding Framework for Android: Design, Implementation, and Benefits

MVVM and DataBinding have become popular patterns for Android UI development; this article presents a custom MVVM‑DataBinding framework that combines the strengths of both and simplifies implementation using Android Architecture Components.

The author first reviews the evolution from MVC to MVP to MVVM, highlighting that MVVM separates UI logic into a ViewModel, improving code robustness and testability, while DataBinding enables automatic UI updates without manual findViewById calls.

Three main pain points of raw DataBinding are identified: complex binding rules, difficulty migrating from MVP, and cumbersome manual binding code such as DatabindingUtil . To address these, the framework defines five concise conventions for placing ViewModel, Floor, and Item data classes and for annotating layout files (using "viewModel", "floor", and "item" variables).

Usage steps are:

Create ViewModel/Floor/ViewHolder classes that extend the provided base classes.

Define layout XML files following the naming and variable conventions, e.g., android:text="@{viewModel.pageTitle}" .

Extend BaseBindingMvvmActivity or BaseBindingMvvmFragment and specify the generated binding class and ViewModel type.

The framework’s internal structure consists of four parts: a common ViewBindingAdapter , base classes for Activity/Fragment/ViewModel, base classes for Floor, and base classes for RecyclerView items. The implementation details show how BR files are generated, how binding classes are instantiated, and how executePendingBindings() triggers UI refreshes.

Advantages include reduced findViewById overhead, fewer thread switches, less boilerplate code, clearer binding rules, and easier unit testing. Potential drawbacks are limited XML expression debugging (often requiring Android Studio restarts) and challenges in dynamic or plugin‑based projects due to compile‑time code generation.

The author also notes context‑handling concerns in ViewModels and suggests using the AAC branch of the repository for a solution.

In conclusion, the framework aims to provide a more concise, performant, and testable approach to Android UI development, encouraging team adoption for better code maintainability and scalability.

GitHub repository: https://github.com/ganlinux/JDMvvmDatabinding.git