每日新聞摘要

Java Weekly Issue 624 highlights developments across the Spring and Java ecosystems. The lead coverage examines the push behind JEP 500 and the “integrity by default” rationale for hardening final against reflective modification to enable future JVM optimizations and correct concurrent behavior. Another featured article recommends rethinking Spring application integration tests by focusing on vertical, functional slices rather than brittle horizontal mocks. The issue aggregates additional recommended reads ranging from Spring Tools, IntelliJ changes, Java language previews, to Micrometer and MongoDB guidance, plus webinars and podcasts. A sizeable “Time to upgrade” section lists recent releases—Spring gRPC 1.0.0, IntelliJ IDEA 2025.3, Spring AI 1.1.1, and many framework and tooling updates. Short technical links and musings close the edition, and the pick of the week recommends Paul Graham’s essay “Good Writing.”

本期 Java Weekly（第624期）聚焦 Spring 与 Java 生态的重要动向。专题文章探讨 JEP 500 背后的“默认完整性”理念，说明为何强化 final 以阻止反射修改有助于未来 JVM 优化与并发行为的正确性。另一篇重点文章建议在 Spring 应用整合测试中以垂直、功能切片为主，避免脆弱的横向 mock。期刊同时汇整延伸阅读、线上研讨與播客，並列出眾多升級提醒（如 Spring gRPC 1.0.0、IntelliJ IDEA 2025.3、Spring AI 1.1.1 等），最後以技術雜談與 Paul Graham 的《Good Writing》作為本週精選結束。

The article opens with the scenario of deciding to build a modern application using Compose Multiplatform and immediately cautions against expecting miracles. While Compose Multiplatform can simplify UI development across platforms and is an attractive option for many projects, it does not eliminate the need for native code. Platform-specific APIs, integrations, and certain behaviors still demand native implementations or interop. The piece advises developers to be pragmatic: embrace multiplatform tooling where it fits, but plan and reserve effort for native code when necessary. In short, multiplatform is a powerful approach, yet realistic expectations and native integration remain essential for shipping complete, polished apps.

文章以「决定用 Compose Multiplatform 打造一个超现代化应用」为切入点，提醒读者不要期望出现奇迹。Compose Multiplatform 在简化跨平台界面开发方面很有价值，但并不能完全替代原生代码。许多平台专属 API、集成需求或特定行为仍需原生实现或调用原生接口。作者建议采取务实的做法：在合适的场景采用跨平台方案，同时为必要的原生开发留出规划与资源，以确保最终交付的应用功能完整且表现良好。

This ProAndroidDev article by Sriyank Siddhartha is a hands-on guide showing how developers can use Meta’s MCP server together with AI-assisted tools to accelerate Android app development for Meta Quest. It focuses on practical techniques that reduce iteration time by enabling faster debugging, streamlining the integration of 3D assets, and introducing automation to handle repetitive tasks. The guide emphasizes applied workflows and tool combinations that help teams move more quickly from development to testing on Meta Quest hardware. Readers can expect actionable guidance on leveraging MCP alongside AI features to simplify common pain points in VR app development and improve overall development efficiency.

这篇由 Sriyank Siddhartha 在 ProAndroidDev 上发表的实践指南，介绍了如何将 Meta 的 MCP 服务器与 AI 辅助工具配合使用，以加快为 Meta Quest 开发 Android 应用的速度。文章强调通过更快速的调试、即时的 3D 资产整合以及对重复任务的自动化来减少迭代时间并简化开发流程。读者可从中获得可操作的流程与工具组合建议，帮助开发团队更高效地从开发阶段推进到测试与部署。

This article is a practical, example-rich guide that explains every Compose Effect API in Jetpack Compose, detailing when to use each one, common pitfalls to avoid, and recommended best practices. It walks through the different Effect APIs with code examples and explanations so that developers can choose the right effect for lifecycle, side-effect management, state synchronization, and interaction with external systems. The piece emphasizes practical guidance over theory, highlighting typical mistakes and patterns to prefer, and offers actionable tips to write safer, more predictable Compose code. Intended for Compose users who want a comprehensive reference to side effects, the guide aims to shorten learning curves and help maintain robust UI logic by clarifying trade-offs and correct usage scenarios for each effect.

本文是一篇实用且包含大量示例的指南，逐一解释 Jetpack Compose 中的各种 Effect API，并说明何时使用、常见陷阱及推荐的最佳实践。通过示例与讲解，帮助开发者在生命周期管理、状态同步和与外部系统交互等情境中，选择合适的 side effect 管理方式。文章侧重可操作性的指导以避免常见错误，提高 Compose 应用的可预测性与稳定性，适合作为希望全面掌握 Compose 副作用处理的开发者参考。

Media3 2.2 updates focus on improving playback reliability, expanding DRM capabilities, and adding AI-powered tools to simplify media app development. For developers building modern audio and video apps with Kotlin and Jetpack Compose, the release promises smoother, more resilient playback experiences, better support for protected content through advanced DRM integrations, and AI features that can assist with tasks such as metadata handling, recommendations, or automated processing. The update emphasizes practical considerations for integrating Media3 into Compose-based UIs and for ensuring secure, performant playback across devices. Developers should review compatibility notes, test DRM workflows thoroughly, and evaluate how AI features can be applied safely and effectively in their apps. Overall, Media3 2.2 aims to reduce implementation friction and empower teams to deliver richer media experiences with fewer low-level concerns.

Media3 2.2 带来无缝播放、进阶 DRM 支持与 AI 驱动功能，适用于使用 Kotlin 与 Jetpack Compose 的音视频应用。此版本强调播放稳定性与受保护内容处理，提升不同设备上的安全与性能；AI 功能可协助元数据处理、推荐与自动化流程，减少开发者处理低阶细节。开发者应检视兼容性与 DRM 工作流程，并评估 AI 功能的实用性与安全性，以利在 Compose UI 中顺畅整合与可靠播放体验。

CameraX 1.4 is presented as a significant step in Android imaging, reshaping how developers capture, process, and deliver photos on mobile devices. The article outlines upcoming upgrades in CameraX 1.4 and explains their importance for mobile photography workflows: streamlining capture pipelines, improving image processing paths, and simplifying delivery to apps and services. For developers, these changes promise a better balance between image quality and implementation complexity, while enabling more consistent results across devices. For product teams and photographers using Android apps, the update could mean richer imaging capabilities and smoother integration. The piece gives context about why the upgrade matters and what developers should expect, encouraging readers to explore the detailed breakdown on Medium to understand specific changes and migration considerations.

CameraX 1.4 被视为 Android 摄影的重要进展，改变了开发者捕捉、处理与传递影像的方式。文章概述了 CameraX 1.4 即将带来的升级，并解释这些改变对移动摄影工作流的意义：简化捕捉流程、强化影像处理链路以及改善交付到应用和服务的方式。对开发者而言，更新有助于在影像质量与实现复杂度之间取得更好平衡，并提升跨设备的一致性；对于产品团队和用户，这可能带来更丰富的拍摄能力与更顺畅的整合体验。作者建议讀者前往 Medium 阅读完整细节与迁移建议。

Android Studio 2025 introduces an updated suite of performance-profiling features aimed at helping developers find and fix runtime issues more quickly. The release highlights AI-powered bottleneck detection that surfaces likely performance hotspots and a real-time memory analysis capability that monitors allocations and retention while the app runs. Together, these features reduce manual investigation by providing automated guidance and live metrics during profiling sessions, helping teams prioritize fixes and validate improvements faster. The Medium article outlines how these additions integrate into familiar profiler workflows, enabling developers to detect regressions earlier, iterate with greater confidence, and maintain app performance throughout development and testing.

Android Studio 2025 推出了一套更新的性能分析功能，重点包括 AI 驱动的瓶颈检测和实时内存分析，旨在帮助开发者更快地发现并修复运行时问题。AI 功能可自动提示潜在性能热点，实时内存分析则在应用运行期间跟踪内存分配与保留情况，二者协同能减少人工调查并提供即时指标支持。Medium 文章概述了这些新增项及其如何整合到熟悉的 Profiler 流程中，帮助团队更早检测回归、优先处理修复，并在开发与测试阶段持续保持性能。

Google's Privacy Sandbox Phase 3 marks a shift toward a cookie-less mobile ecosystem with direct implications for Android apps and their measurement systems. The article outlines why this change matters for app developers and product teams, urging them to reassess analytics, update SDKs, and explore privacy-preserving alternatives to traditional tracking. It emphasizes practical preparation: audit current data collection, validate critical metrics under new constraints, and plan migrations or instrumentations that rely less on third-party cookies. By testing changes early, coordinating with vendors, and prioritizing user privacy while preserving essential measurement needs, teams can reduce disruption when the new privacy protections roll out. The overall message is that proactive adaptation and careful testing are key to maintaining reliable analytics in a cookie-less mobile world.

文章指出 Google 的 Privacy Sandbox 第 3 阶段将使移动生态迈向无 Cookie 的现实，对 Android 应用的数据收集与分析带来直接影响。作者建议开发和产品团队审查现有的数据采集流程与 SDK，调整分析策略，探索在隐私保护下的替代方案，并对关键指标进行验证与测试。通过及早准备、与供应商协同并减少对第三方 Cookie 的依赖，团队可在隐私新政推出时维持重要的测量能力并降低业务中断风险。核心在于主动应对及周密测试，以在无 Cookie 的移动环境中保留可靠的分析能力。

In Jetpack Compose, remember { mutableListOf() } and mutableStateListOf() both yield mutable lists but behave differently with respect to UI updates. remember preserves the same regular MutableList instance across recompositions, which is useful to keep a stable reference, but Compose will not automatically recompose when you add, remove, or change items in that list because it is not observable. By contrast, mutableStateListOf() returns a SnapshotStateList that participates in Compose’s snapshot/state system: modifications to that list (such as add/remove) are observed and will trigger recomposition of dependent composables. For UI-backed collections, prefer mutableStateListOf() so the UI reflects changes; use remember { mutableListOf() } only when you manage updates manually or do not need automatic recomposition.

在 Jetpack Compose 中，remember { mutableListOf() } 和 mutableStateListOf() 都能创建可变列表，但对 UI 更新的处理不同。remember 会在重组之间保留相同的普通 MutableList 实例，适用于需要稳定引用的场景，但对该列表的增删改不会被 Compose 观察到，因此不会自动触发重组。mutableStateListOf() 返回的是 SnapshotStateList，它参与 Compose 的 snapshot/state 系统，列表发生变化（例如添加或删除元素）时会通知依赖的 composable 并触发重组。若列表内容会影响界面显示，应优先使用 mutableStateListOf()；仅在不需要自动重组或由你自行管理更新时使用 remember { mutableListOf() }。

The author warns developers to stop writing "Java in Kotlin" and to focus on idiomatic Kotlin by addressing five prevalent anti-patterns. Rather than translating Java patterns directly, the article advocates refactoring code into Kotlin-native constructs that leverage concise syntax, null-safety, and expressive standard-library functions. By recognizing these recurring mistakes and applying targeted refactorings, engineers can make code more readable, maintainable, and less error-prone. The piece serves as a concise call to action: invest time in learning Kotlin idioms, replace Java-centric habits with Kotlin-appropriate alternatives, and iteratively improve code quality and team productivity through clearer, more idiomatic implementations.

作者提醒开发者不要把 Kotlin 当成 Java 来写，应该采用惯用的 Kotlin 写法来重构常见反模式。文章指出五种普遍错误，并建议通过重构改用 Kotlin 专有的语法和标准库功能。通过识别这些反模式并替换为更简洁、符合语言精神的实现，可以提升代码的可读性、可维护性并降低错误风险。核心信息是停止“在 Kotlin 中写 Java”，投入学习 Kotlin 的 null 安全、简洁语法与富表达力的函数式工具，从而逐步改善代码质量与团队开发效率。

Prahalad Sharma's post "ExoPlayer setup for Android TV — Media Playback on TV (Compose) Part-6" is a short entry in a series about media playback on Android TV with Jetpack Compose. The visible content indicates this installment focuses on configuring ExoPlayer for Android TV and directs readers to a GitHub repository containing the full source code. The article functions as Part 6 of the series and includes a link to continue reading on Medium for the complete explanation. The preview itself is brief and primarily acts as an index pointing to the GitHub code and the full Medium article for implementation details, examples, and step‑by‑step instructions.

Prahalad Sharma 的文章《在 Android TV 上设置 ExoPlayer — 电视媒体播放（Compose）第6部分》属于关于在 Android TV 使用 Jetpack Compose 进行媒体播放的系列文章。可见内容显示本篇侧重于 ExoPlayer 的配置，并指向一个包含完整源码的 GitHub 仓库。文章作为系列的第6部分，提供了继续在 Medium 阅读完整内容的链接。预览信息有限，想获取实现细节和代码示例请访问文章中提供的 GitHub 仓库与 Medium 正文。

This article shows how to set up and write JUnit 5 tests in a Micronaut application. It explains required Maven dependencies—micronaut-test-junit5, junit-jupiter-engine, junit-jupiter-api, and mockito-core—and their roles for integration, test execution, annotations, and mocking. Core concepts cover annotating test classes with @MicronautTest to start the ApplicationContext (and EmbeddedServer for HTTP tests), injecting beans with @Inject, and avoiding manual lifecycle management. The tutorial demonstrates testing a simple AdditionService implementation, using startApplication=false for non-HTTP unit tests, and replacing beans with Mockito mocks via @MockBean. It also shows testing an HTTP controller by injecting an HttpClient, sending real requests to the embedded server, and asserting responses while the service is mocked for isolation.

本文介绍如何在 Micronaut 应用中配置并编写 JUnit 5 测试。先列出并说明关键 Maven 依赖：micronaut-test-junit5（整合与上下文管理）、junit-jupiter-engine（测试执行）、junit-jupiter-api（注解與断言）及 mockito-core（创建 mock）。核心概念包括在测试类上使用 @MicronautTest 来启动 ApplicationContext（需要时自动启动 EmbeddedServer）、通过 @Inject 注入 Bean，以及通过 @MockBean 替换 Bean 以隔离被测单元。文章通过 AdditionService 与 AdditionController 的示例展示非 HTTP 单元测试（startApplication=false）、使用 Mockito stub 行为，以及通过 HttpClient 向嵌入式服务器发送真实 HTTP 请求来验证控制器响应。

In this tutorial Kai Yuan demonstrates multiple ways to insert BLOB data with Spring’s JdbcTemplate, using an H2 in-memory table and focused JUnit tests. The article shows three practical approaches: passing a byte[] via PreparedStatement.setBytes for small binary payloads; streaming data through an InputStream with setBinaryStream to avoid high memory usage for large files; and using Spring-specific helpers—historically SqlLobValue with a LobHandler (now deprecated in Spring 6.2) and the newer SqlBinaryValue wrapped in SqlParameterValue for typed binary parameters. Example code illustrates insertion and retrieval assertions for each method. The guide emphasizes choosing the method that matches file size, data source, and JDBC driver behavior, recommending streams for large content and SqlBinaryValue for modern Spring applications. Full example code is available on GitHub.

本文由 Kai Yuan 演示如何使用 Spring 的 JdbcTemplate 插入 BLOB，透过 H2 内存数据库与独立 JUnit 测试展示三种实作：将 byte[] 直接写入适用于小型二进制数据；以 InputStream 流式写入适用于大型文件以降低内存使用；以及 Spring 专属方案：历史上的 SqlLobValue/LobHandler（在 Spring 6.2 被标为弃用）与新版的 SqlBinaryValue（搭配 SqlParameterValue 指定 JDBC 类型）。文章通过示例说明如何插入与检索数据，并强调应根据文件大小、数据来源与驱动特性选择合适方案，完整示例代码托管于 GitHub。

This Medium post by Erwin Elder outlines the process of deploying a basic distributed system using multiple locally running virtual machines. The preview indicates the article will describe how to set up and deploy components across different virtual hosts on the same physical machine, presenting a practical walkthrough of the deployment process. Intended for readers who want to experiment with distributed architectures without relying on remote cloud infrastructure, the piece frames local virtualization as a way to simulate distributed environments and validate deployment steps. The author positions the write-up as an introductory, hands-on guide for those wishing to run a simple distributed system locally; readers are directed to the full Medium article for the complete instructions and examples.

Erwin Elder 在 Medium 上的这篇文章将说明如何通过多个在本机运行的虚拟机部署一个基本的分布式系统。预览显示文章会描述在不同虚拟主机上设置与启动系统组件的部署流程，并以实践导向的方式示范如何在同一硬件上模拟分布式环境。此文定位为入门说明，适合希望在本地进行分布式架构实验的开发者或学生；完整内容与步骤请在 Medium 原文继续阅读，文中将提供条理化的部署说明与操作要点。

This article identifies eight common Jetpack Compose recomposition mistakes that harm app performance and provides a production‑grade guide to diagnosing and fixing them. It focuses on stability and state management patterns to reduce unnecessary recompositions, explains how careless use of lambdas and captured state can trigger frequent UI updates, and gives practical strategies for optimizing LazyColumn usage to lower UI churn and memory churn. The guide emphasizes profiling and real‑world fixes — recognizing problematic composition scopes, using remember and keys smartly, avoiding re-creating objects on every frame, and batching state updates. It also covers trade-offs and recommends measurement-driven decisions so fixes target real bottlenecks rather than premature optimization.

本文指出八种常见的 Jetpack Compose 重组（recomposition）错误，说明它们如何削弱应用性能，并提供生产级的诊断与修复建议。内容侧重于稳定性与状态管理模式，解释不当使用 lambda 或捕获状态会导致频繁重组，并提出 LazyColumn 优化、使用 remember 与 key、避免在每帧重建对象、批量化状态更新等实务策略。文章强调以测量为依据定位瓶颈，并提供可操作的修补方法，帮助开发团队在不大幅改动架构的情况下，减少 UI 抖动、降低内存与 CPU 开销，提升响应与稳定性。

This article by Jack Henry offers a concise, practical guide for fresh Android developers preparing for their first job interviews. It focuses on essential interview questions, core concepts, and pragmatic tips to help candidates present themselves confidently. Topics include common technical areas, how to structure answers, and strategies to demonstrate problem-solving and learning ability even with limited experience. The guide emphasizes prioritizing fundamentals over memorizing obscure details, practicing clear explanations, and showcasing project work or internships. It also provides advice on handling behavioral questions, interview logistics, and follow-up etiquette. Aimed at newcomers, the no-nonsense tone helps readers create a focused study plan, reduce anxiety, and improve their chances of converting interviews into offers by combining technical readiness with communication and professionalism.

作者 Jack Henry 提供一份直接實用的求職指南，面向首次求職的 Android 新手，整理常見面試題目與核心概念，並給予務實的答題與展示建議。文章強調掌握基礎勝於背誦冷僻細節，建議透過專案成果與清晰表達展現學習與解題能力，並涵蓋行為面試準備、面試流程與後續禮節。目的是幫助求職者制定重點學習計畫、減少緊張，並提高把面試轉化為錄用的機會。

This brief first article in a three-part series by Thomas Martin announces that the team moved away from Spring Security’s SPEL for permission evaluation. After searching for a better-fitting off-the-shelf solution and finding none, they opted to design and build a custom permission system. The post functions as an introduction to that decision and signals that subsequent installments will cover the new system in greater detail. Published on WhozApp, the piece summarizes the motivation to replace SPEL and invites readers to continue with the follow-up posts to learn about the design, implementation choices, and rationale behind the bespoke permission system.

Thomas Martin 在这篇三部分系列的第一篇简短文章中说明，团队决定不再使用 Spring Security 的 SPEL 来处理权限问题。因未能找到更合适的现成替代方案，团队选择自行设计并实现一套权限系统。文章作为该决策的引言，表明后续文章将对新系统的设计与实现做更详细的阐述。该公告发布于 WhozApp，主要用于说明改动动机并引导读者阅读后续更深入的内容。