每日新聞摘要

The author describes a difficult migration of a large KMP project from AGP 8 to AGP 9.0 and provides a practical step-by-step guide. Major breaking changes include removal of build-type support (which disables BuildConfig-based static flags), requiring replacement with DI-provided configuration; prohibition of NDK/JNI in library modules, necessitating interfaces in libraries and native implementations in the app module; and a required plugin/DLS migration to com.android.kotlin.multiplatform.library and kotlin.android { } setup. Other changes: replace androidUnitTest with androidHostTest, enable android resources where Compose Multiplatform needs them, add a Kotlin buildscript dependency to align versions, and apply consumer keep rules per library. The post warns that some plugins (Hilt, KSP) initially lacked AGP 9 support and that the migration can be time-consuming, and offers the author’s consulting help for teams needing assistance.

作者描述将大型 KMP 项目从 AGP 8 迁移到 AGP 9.0 的艰难过程，并给出可操作的迁移步骤。关键变更包括：取消 build type 支持导致不能再用 BuildConfig 静态标志，需将这些标志改为通过依赖注入提供的 AppConfiguration；库内的 NDK/JNI 代码不得继续存在，需在应用模块实现并通过接口调用；移除旧的 Kotlin Android 插件，改用 com.android.kotlin.multiplatform.library 并将原有顶层 android 配置合并到 kotlin.android { }；同时需要将 androidUnitTest 替换为 androidHostTest、为使用 Compose Multiplatform 的模块显式启用 androidResources、并在 buildscript 中添加 Kotlin 插件以保持版本一致。文章提醒部分官方插件（如 Hilt、KSP）在早期可能未支持 AGP 9，并提供作者的迁移咨询服务。

This article provides a complete migration guide for Android Gradle Plugin (AGP) 9.0.0, released January 2026. It identifies AGP 9.0.0 as a major release that introduces breaking changes and aims to help Android developers update projects and build configurations safely. The guide outlines the scope of the release, recommends step-by-step migration actions, and emphasizes checking compatibility, running tests, and preparing developer tools and CI before upgrading. It warns developers to validate changes in non-production branches or staging environments to avoid build or runtime regressions. For detailed procedures, examples, and troubleshooting, the post directs readers to continue reading on the linked Dev Genius blog.

本文为 Android Gradle Plugin (AGP) 9.0.0 的完整迁移指南，发布于 2026 年 1 月，说明该版本为重大更新并包含破坏性变更，旨在帮助开发者安全升级项目与构建配置。文章概述了版本范围，建议逐步迁移操作，强调进行兼容性检查、运行测试并准备好开发工具与 CI，提醒在功能分支或预发布环境验证后再推向生产以避免构建或运行时回归。更多详细步骤、示例与疑难排解请参考 Dev Genius 原文。

This excerpt from "Kotlin: 10 Must-Know Concepts" highlights data classes as one of the essential Kotlin topics. It states plainly that the primary purpose of a data class is to store data and that you create one by prefixing a regular class declaration with the keyword data. The passage is short and serves as an introductory note rather than a full explanation, ending with a prompt to continue reading on MobilePeople for the remainder of the discussion. Positioned within a broader list of ten important Kotlin concepts, the snippet emphasizes declaration syntax and frames data classes as a key construct worth exploring further in the complete article.

本节摘自《Kotlin：10个必知概念》，聚焦于数据类。文中说明数据类的主要目的在于存储数据，并且可以通过在普通 class 前添加关键字 data 来创建数据类。该段落较为简短，作为入门说明并提示读者前往 MobilePeople 阅读完整文章。置于十个重要概念的整体框架内，这一节强调了数据类的声明语法，并将其定位为一个值得进一步学习的 Kotlin 基础构件。

This Medium post by Jan Wylegala presents an Android project that builds on MediaPipe's real-time hand-tracking to add gesture recognition and a playful emoji-rain animation. The project demonstrates how hand-tracking output can be used to detect gestures and trigger an animated visual effect as feedback, illustrating integration of ML-based tracking with interactive UI elements. Aimed at developers interested in on-device real-time image processing and gesture interaction, the article serves as a practical demonstration of combining gesture detection with visual animation on Android. It highlights the concept and a working demo of emoji rain to showcase how gesture events can drive engaging visual responses in an application.

这篇由 Jan Wylegala 在 Medium 发布的文章介绍了一个在 Android 平台上的项目：以 MediaPipe 的实时手部追踪为基础，加入手势识别功能，并以趣味的表情符号雨（emoji rain）动画作为视觉反馈。文章演示如何把手部追踪结果用于检测手势并触发动画，展示了将基于机器学习的追踪与互动动画结合的实用示例，适合对实时图像处理与手势交互感兴趣的开发者参考和学习。

This short piece recalls the early days of AI coding assistants, when asking an assistant to "write a function" often returned perfectly valid code. The author explains that, despite those capabilities, integrating AI-generated code into real Android projects highlighted collaboration gaps between automated tools and human developers. To address those gaps, the AGENTS.md file was born: its purpose is to help AI systems and Android engineers work together by clarifying roles, expectations, and workflows so that generated code can be more effectively integrated into a project. The article outlines the motivation and practical intent behind AGENTS.md as a lightweight coordination mechanism aimed at improving how AI outputs are used in Android development contexts.

文章回顾了早期 AI 编程助手的情景：当你让它们“写一个函数”时，往往会得到看似正确的代码；但在把这些输出实际用于 Android 项目时，仍会出现协作上的不足。为了解决这些不足并明确彼此的期望，作者介绍了 AGENTS.md 的诞生原因与目标：作为一个轻量的协作与协调文件，帮助定义角色与工作流，使 AI 生成的代码更容易被 Android 开发者采用并顺利整合到项目中。

In this article, Kabi argues that in fast-paced Android development it's common to favor delivering features quickly at the cost of long-term maintainability, resulting in 'spaghetti' code in the data layer. The piece advocates applying SOLID design principles and decoupling strategies in Kotlin to separate concerns, reduce tight coupling between components (repositories, sources, and consumers), and make testing and future changes easier. It outlines the benefits of a cleaner, modular data layer—improved readability, easier unit testing, and safer evolution as the app grows—while urging developers to balance maintainability with functionality. The article also offers practical examples and Kotlin snippets that illustrate how to refactor common data-access patterns into interfaces and layered components, enabling teams to adopt improvements incrementally and lower technical debt.

Kabi 指出，在快节奏的 Android 开发中，优先交付功能往往以牺牲长期可维护性为代价，使数据层演变为混乱的“意大利面”代码。文章主张在 Kotlin 中采用 SOLID 设计原则与解耦策略，通过分层与接口隔离数据访问、数据源和使用方，减少紧耦合并提升单元测试与后续变更的可行性。文中阐述了更清晰、模块化的数据层能带来的好处——更易阅读、更易测试以及随着应用增长更安全的演进，并通过示例与重构建议，说明如何逐步将临时性的代码组织转为责任明确的架构，从而降低技术债务并提升团队維護效率。

The article examines the Google Play Integrity API as a practical tool for Android app security and notes that many developers postpone security work until incidents such as fraud, cracked APKs, or fake installs occur. Drawing on real-world examples, it explains how integrity checks can help identify tampered APKs, unauthorized execution environments, and suspicious installs, while also highlighting practical implementation considerations and common challenges like complexity, edge cases, and the need for complementary server-side logic. The piece advocates a layered defense strategy—combining the Play Integrity API with other protections, monitoring, and proactive testing—and emphasizes treating integrity verification as an ongoing process rather than a one-time feature to reduce fraud and improve app trustworthiness.

本文审视 Google Play Integrity API 在 Android 应用安全中的实用性，指出许多开发者会推迟安全工作，直到发生欺诈、被破解的 APK 或虚假安装等问题才重视。文章以真实案例说明完整性检查如何帮助识别被篡改的 APK、未授权的执行环境与可疑安装，并讨论整合该 API 时的实际考量与常见挑战，如实现复杂性、边界情形及需配合后端逻辑。作者建议采用分层防御，结合其他防护、监控与主动测试，并将完整性检测视为持续性流程，以降低欺诈风险并提升应用信任度。

This brief announcement covers the release of Kotlin Compose Multiplatform 1.10.0 and highlights two core promises: “Write once, preview everywhere” and “Hot reload with a single click.” The message positions 1.10.0 as a release focused on improving developer productivity by enabling unified UI authoring and immediate feedback across targets through centralized previews and quick reloads. By reducing iteration time and simplifying the preview workflow, the release aims to make cross-platform UI development with Compose more efficient and pleasant. Readers are invited to continue reading on Medium for full details, examples, and guidance on adopting the new version in their projects.

本文简要宣布 Kotlin Compose Multiplatform 1.10.0 的发布，突出“一次编写，处处预览”和“单击热重载”两大要点。该版本旨在通过统一的预览体验和即时重载缩短开发与验证反馈周期，从而提升跨平台界面开发的效率与流畅度。文章建议读者在 Medium 上继续阅读以获取详细说明、示例和采用建议，帮助开发者评估并将新版本整合到其多平台 UI 开发流程中。

This article explains how to use Koin for dependency injection in Kotlin, with emphasis on defining dependencies behind interfaces and selecting among multiple implementations using qualifiers. It describes the rationale for injecting interfaces to keep code decoupled and testable, and explains how Koin’s module bindings and qualifier labels let you register and retrieve different implementations cleanly. The article highlights common patterns for registering interface implementations, resolving dependencies by qualifier at injection sites, and the benefits such as clearer architecture and easier testing. It targets developers seeking practical guidance rather than deep theoretical treatment, and points readers to the full source code available at the end of the article for hands‑on examples and reference. Examples illustrate registration patterns and runtime selection to help developers adapt the approach to their app needs.

本文介绍在 Kotlin 项目中使用 Koin 进行依赖注入的做法，重点在于通过接口抽象并使用 qualifier 区分多个实现。文章说明为何注入接口有助于降低耦合并提升测试性，阐述 Koin 的模块绑定与 qualifier 标签如何注册和解析不同实现的基本方法。文章列出常见的注册与解析模式及其带来的架构与测试优势，并以实务为导向提供说明，文末附有完整的示例源代码以便读者参考和实践。

This article examines the runBlocking coroutine function from the perspective of Android development. It explains what runBlocking does, why developers often learn it early, and the trade-offs involved when it is used in Android apps. The author highlights scenarios where runBlocking can be appropriate (for example, simple entry points or bridging blocking and suspending code) and scenarios where it is harmful, particularly when it blocks the Android main thread and risks UI hangs. The piece emphasizes practical lessons drawn from real development experience and discusses safer alternatives and patterns—such as using lifecycle-aware coroutine scopes, suspend functions, and structured concurrency—to avoid blocking behavior on Android while preserving predictable concurrency and responsiveness.

本文从 Android 开发者的视角审视 runBlocking 函数，说明其功能与为何常被初学者接触到，并讨论在不同场景下的利弊。作者指出 runBlocking 在某些简单入口或需要将阻塞代码与挂起函数连接时可能有用，但在 Android 主线程上使用会导致阻塞并可能引发界面卡顿等问题。文章以实际开发经验为基础，提出避免阻塞主线程的替代方案和模式，例如使用与生命周期绑定的 coroutine scope、采用 suspend 函数与结构化并发等，以维持应用的响应性与并发的可预测性。

This piece presents higher-order functions (HOFs) as a fundamental mechanism in modern Kotlin, arguing they streamline behavioral architecture and underpin the declarative nature of Jetpack Compose. By treating behavior as first-class values that can be passed, composed, and reused, HOFs reduce coupling between components, make intent explicit, and simplify event and state handling. In the context of Jetpack Compose, HOFs help express UI logic declaratively: composables receive behavior via function parameters, enabling clearer component boundaries, easier testing, and more predictable rendering. The article positions HOFs not merely as a language feature but as an architectural tool that promotes composability, readability, and maintainability in modern Kotlin applications.

本文强调高阶函数（HOF）在现代 Kotlin 中的基础性作用，阐述其如何通过将行为作为可传递、可组合的值来简化整体架构。HOF 有助于降低组件耦合、明确行为意图并简化事件与状态管理；在 Jetpack Compose 场景下，将行为以函数参数传入 composable，可以让 UI 逻辑以声明式表达，使组件边界更清晰、测试更便捷且渲染更可预测。文章将 HOF 看作一种架构工具，能够提升可组合性、可读性和可维护性，而不仅仅是语言的语法特性。

In this short piece, Riyas Pullur argues that Android development has evolved beyond traditional concerns like Activities, ViewModels, and APIs, and that prompt engineering is becoming an essential part of the developer toolkit. The article's title and opening line signal a broader shift in the platform’s expectations, indicating that developers must engage with new techniques for interacting with intelligent systems. While the full discussion is available on Medium, the core message is clear: building modern Android applications increasingly involves designing effective prompts and interfaces for AI-driven features alongside conventional app architecture considerations.

作者 Riyas Pullur 在这篇短文中指出，Android 开发已不再仅仅围绕 Activities、ViewModels 和 APIs，提示工程（prompt engineering）正在成为开发者工具箱中不可忽视的一项技能。标题与开头句表明了平台期望的扩展，意味着开发者需要掌握与智能系统交互的新方法。虽然完整论述可在 Medium 阅读，但文章传达的核心观点很明确：构建现代 Android 应用除了传统架构考量外，也应重视提示设计与 AI 交互策略。