大家好！欢迎来到程序视点，我是你们的老朋友.安戈。

前言

目前IntelliJ IDEA 正是进入2025.2最新版本！

以下是IntelliJ IDEA 2025.2第一个独立修复版本的概述！

We are always working hard to make the IDE even more stable, reliable, and enjoyable to use. In previous releases, information about this work was blended into the What’s New content, making it very long and sometimes hard to grasp. To address this issue, we’ve decided to focus the What’s New only on the most impactful and noticeable changes and to introduce a dedicated What’s Fixed article that focuses on the stability and quality improvements.

我们一直在努力使 IDE 更加稳定、可靠和使用起来愉快。（激活请看下一步的内容）

在以前的版本中，有关此作品的信息被混合到“新增功能”内容中，使其非常长，有时难以掌握。为了解决这个问题，我们决定将新功能仅关注最具影响力和最显着的变化，并引入一篇专门的修复问题文章，重点介绍稳定性和质量改进。

The key areas we have been working on in this release are:

我们在此版本中一直致力于的关键领域是：

Spring support（Spring支撑）
Kotlin support in Spring projects
（Spring 项目中的 Kotlin 支持）
Spring Security and Spring Web
（Spring Security 和 Spring Web）
Remote development（远程开发）
Support for WSL（对 WSL 的支持）
Kotlin K2 mode（Kotlin K2 模式）
Kotlin Notebook（Kotlin 笔记本）
Maven execution（Maven 执行）
Cloud tooling（云工具）
Performance（性能）

But that’s not everything. For the full list of fixes and improvements, please refer to the release notes.

但这还不是全部。有关修复和改进的完整列表，请参阅发行说明。立即更新到 IntelliJ IDEA 2025.2，关注薇亻言工号【程序视点】，回复vip，领取优惠激活！

Improved Kotlin support for Spring projects(改进了对 Spring 项目的 Kotlin 支持)

With Kotlin proving itself as a good fit for the Spring Framework, more and more people are using it in Spring projects. To deliver the best developer experience, we formed a partnership between Kotlin and Spring, and we are actively improving IntelliJ IDEA’s Kotlin-Spring support. Version 2025.2 brings:

随着 Kotlin 证明自己非常适合 Spring 框架，越来越多的人在 Spring 项目中使用它。为了提供最佳的开发者体验，我们在 Kotlin 和 Spring 之间建立了合作伙伴关系 ，并且我们正在积极改进 IntelliJ IDEA 的 Kotlin-Spring 支持。版本 2025.2 带来：

Live templates and context menu actions for creating Spring beans quickly, along with improvements to the Logical structure view to make understanding the relations between application components easier.

用于快速创建 Spring Bean 的实时模板和上下文菜单作，以及对逻辑结构视图的改进，以使理解应用程序组件之间的关系更容易。

Smarter bean support: No more false injection points for constructor parameters for @SpringBootTest or for parameters with defaults, and improved handling of @ConditionalOnProperty beans.

更智能的 bean 支持： 对于 @SpringBootTest 的构造函数参数或具有默认值的参数，不再有错误的注入点，并改进了对 @ConditionalOnProperty bean 的处理。

Configuration made clear: The ability to go from @ConfigurationProperties to YAML or properties with a click, default values in code completion, and full support for nested properties.

配置明确：只需单击一下即可从 @ConfigurationProperties 转到 YAML 或属性，代码完成中的默认值，以及对嵌套属性的完全支持。

Spring MVC improvements: An easier experience with path variables in request handlers, thanks to clear feedback when something doesn’t match.

Spring MVC 改进：在请求处理程序中更轻松地使用路径变量，这要归功于在某些内容不匹配时的清晰反馈。

Spring Data fixes: Better repository method completion for is properties, as well as cleaner support for varargs, projections, and named queries.

Spring Data 修复：更好地完成 is 属性的存储库方法，以及对 varargs、投影和命名查询的更清晰的支持。

Spring Framework support(Spring 框架支持)

Spring Framework support is a key focus area for IntelliJ IDEA. In addition to new functionality such as Spring Debugger, we continue to enhance existing features.

Spring 框架支持是 IntelliJ IDEA 的一个重点领域。除了 Spring Debugger 等新功能外，我们还继续增强现有功能。

In this release, we’ve resolved a couple of minor but frustrating issues related to rule parsing in Spring Security:

在此版本中，我们解决了与 Spring Security 中的规则解析相关的几个小但令人沮丧的问题：

Custom login and logout URLs are now processed correctly.

自定义登录和注销 URL 现在可以正确处理。

Complex security configurations involving multiple role assignments for a single endpoint are now properly parsed.

现在可以正确解析涉及单个终结点的多个角色分配的复杂安全配置。

We’ve also enhanced support for the widely used Spring Web library. URL references – especially in MockMvc tests – are now properly recognized and autocompleted.

我们还增强了对广泛使用的 Spring Web 库的支持。URL 引用（尤其是在 MockMvc 测试中）现在可以正确识别和自动完成。

Remote development(远程开发)

Remote development – connecting to a powerful remote host where the IDE runs from a thin frontend client – is becoming the norm, with many enterprises relying on it.

远程开发——连接到一个功能强大的远程主机，其中 IDE 从瘦前端客户端运行——正在成为常态，许多企业都依赖它。

We are fully committed to making the remote development experience in IntelliJ IDEA equal to the local one. While it has proven to be a hard endeavor for such a complex product as IntelliJ IDEA, and achieving this goal is taking time, version 2025.2 marks a huge step in this direction:

我们完全致力于使 IntelliJ IDEA 中的远程开发体验与本地体验相同。虽然事实证明，对于像 IntelliJ IDEA 这样复杂的产品来说，这是一项艰巨的任务，并且实现这一目标需要时间，但 2025.2 版本标志着朝着这个方向迈出了一大步：

The editor is now as responsive when working remotely as it is in a local setup.

编辑器现在在远程工作时与在本地设置中一样响应。

This is also true for the Search Everywhere, Find in Files, and Git Branches popups.

“ 随处搜索 ”、“ 在文件中查找 ”和“Git 分支 ”弹出窗口也是如此。

The debugger has been reworked so that its backend is isolated from its frontend, making debugging work natively.

调试器已经过重新设计，使其后端与前端隔离，从而使调试在本机工作。

Plugins and settings are properly synchronized between the client and host.

插件和设置在客户端和主机之间正确同步。

Last but not least – remote development connections are now available directly in the Toolbox app, with support for all OS and OpenSSH connections. And you know what? They are at least 1.5x faster!

最后但并非最不重要的一点是，远程开发连接现在可以直接在 Toolbox 应用程序中使用，支持所有作系统和 OpenSSH 连接。你知道吗？它们至少快 1.5 倍！

Support for WSL(对 WSL 的支持)

The majority of our users are on Windows, which is known to have some performance limitations and specific toolchains. That is where WSL comes in handy, and roughly one out of every five Windows users works with WSL eventually.

我们的大多数用户都在 Windows 上，众所周知，Windows 有一些性能限制和特定的工具链。这就是 WSL 派上用场的地方，大约五分之一的 Windows 用户最终会使用 WSL。

IntelliJ IDEA has offered WSL support for some time. However, the way we chose to provide the integration proved unscalable and did not work well for large projects – with issues ranging from performance problems to the inability to run or even open projects.

IntelliJ IDEA 提供 WSL 支持已有一段时间了。然而，事实证明，我们选择提供集成的方式无法扩展，并且不适用于大型项目——问题包括性能问题到无法运行甚至无法打开项目。

To ensure you can stay productive regardless of where your project is, we started a rework of the WSL integration from the ground up. The IntelliJ Platform team created a new API that dramatically reduces communication overhead between the IDE and WSL and allows you to work with WSL projects natively as though they were in the same system as the host IntelliJ IDEA – without needing to spin up a full backend inside WSL.

为了确保无论项目位于何处，都能保持高效，我们从头开始对 WSL 集成进行返工。IntelliJ 平台团队创建了一个新的 API，可显著减少 IDE 和 WSL 之间的通信开销，并允许你本机处理 WSL 项目，就好像它们与主机 IntelliJ IDEA 位于同一系统中一样，而无需在 WSL 中启动完整的后端。

While this is still a work in progress, we have migrated significant parts of the integration to this new API. As a result, in 2025.2:

虽然这仍在进行中，但我们已经将集成的重要部分迁移到这个新的 API。因此，在 2025.2 中：

Maven and Gradle projects can be reliably built and tested, with proper JDK detection and better overall performance.

Maven 和 Gradle 项目可以可靠地构建和测试，具有适当的 JDK 检测和更好的整体性能。

You now have a consistent experience with application servers, whether you’re running Tomcat or WildFly servers locally or on WSL.

现在，无论你是在本地还是在 WSL 上运行 Tomcat 或 WildFly 服务器，都可以获得一致的应用程序服务器体验。

Further improvements to Kotlin K2 mode(Kotlin K2 模式的进一步改进)

K2 mode, the new default implementation of Kotlin support that improves Kotlin code analysis stability and overall performance, continues to evolve. In this release, we’ve:

K2 模式是 Kotlin 支持的新默认实现 ，可提高 Kotlin 代码分析的稳定性和整体性能，该模式仍在不断发展。在此版本中，我们：

Provided support for Kotlin 2.2 features.

提供对 Kotlin 2.2 功能的支持。

Improved the performance of code completion.

改进了代码补全的性能。

Added type-smart completion support.

添加了类型智能补全支持。

Added new inspections with corresponding coroutine-specific quick-fixes and refactorings, like Convert Enum to Sealed Class, Extract Interface, Rearrange Code, and Create Test and Navigate for files and functions.

添加了新的检查，其中包含相应的特定于协程的快速修复和重构，例如将枚举转换为密封类 、 提取接口 、 重新排列代码以及为文件和函数创建测试和导航 。

Resolved several false-positive errors in inspections.

解决了检查中的几个误报错误。

Improved the reliability of Kotlin script execution.

提高了 Kotlin 脚本执行的可靠性。

Learn more about the state of K2 mode in our recent blog post.

在我们最近的博客文章中了解有关 K2 模式状态的更多信息。

Quality upgrades for Kotlin Notebook（Kotlin Notebook 的质量升级）

Kotlin Notebook is a go-to tool for performing ad hoc and prototyping tasks, presenting, writing documentation, and visualizing data, all while enjoying access to your project context and easy dependency management. Since its recent release, interest in it has been growing, and we’ve received a lot of feedback that we are now addressing, focusing on stability and the editing experience.

Kotlin Notebook 是执行临时和原型设计任务、演示、编写文档和可视化数据的首选工具，同时可以访问您的项目上下文并轻松管理依赖项。自最近发布以来，人们对它的兴趣一直在增长，我们收到了很多反馈，我们现在正在处理这些反馈，重点关注稳定性和编辑体验。

AI Assistant cells now allow you to choose a model, view the prompt history, and get automatic suggestions.

AI Assistant 单元格现在允许您选择模型、查看提示历史记录并获取自动建议。

Incorrect syntax highlighting has been fixed.

不正确的语法突出显示已修复。

Dependency management has been improved.

依赖关系管理已得到改进。

Kotlin intentions and inspections have been added to make the editing experience close to what you get in a regular .kt file.

添加了 Kotlin 意图和检查，使编辑体验接近常规 .kt 文件 。

Maven execution(Maven 执行)

Maven integration is used by half of our users. Historically, IntelliJ IDEA has additional configuration layers on top of Maven, and while they bring speed and control, some differences between execution from the IDE and from the CLI could occur.

我们一半的用户使用 Maven 集成。从历史上看，IntelliJ IDEA 在 Maven 之上有额外的配置层，虽然它们带来了速度和控制，但从 IDE 和 CLI 执行之间可能会出现一些差异。

One of the common reasons for such problems was that the specific environment configuration – the .m2/settings file, .mavenrc, or environment variables – was not properly inherited. Starting from 2025.2, IntelliJ IDEA starts Maven goals in such a way that the environment is fully respected.

造成此类问题的常见原因之一是特定的环境配置（.m2/settings 文件、.mavenrc 或环境变量）没有正确继承。从 2025.2 开始，IntelliJ IDEA 以充分尊重环境的方式启动 Maven 目标。

Thanks to this change, it is now possible to use the Maven Daemon for goal execution. If you have one installed, specify it as a Maven home path in File | Settings | Build, Execution, Deployment | Build Tools | Maven, and all goals will be run using it, speeding things up a bit and reducing resource usage.

由于这一变化，现在可以使用 Maven 守护进程来执行目标。如果您安装了一个，请在文件 | 设置 | 构建、执行、部署 | 构建工具 |Maven，所有目标都将使用它运行，从而加快速度并减少资源使用。

Project sync and goal execution now also support plugins that supply custom M2E lifecycles, for example, Byte Buddy. When such plugins are present, the required pre-build and post-build steps will be configured and executed as necessary.

项目同步和目标执行现在还支持提供自定义 M2E 生命周期的插件，例如 Byte Buddy。当存在此类插件时，将根据需要配置和执行所需的构建前和构建后步骤。

Cloud tooling(云工具)

Interaction with cloud services is now a major part of software development workflows. This includes the building, deployment, and sometimes managing remote environments. Terraform and Kubernetes are key tools in these scenarios, which is why they are our focus areas.

与云服务的交互现在是软件开发工作流程的主要部分。这包括构建、部署，有时还包括管理远程环境。Terraform 和 Kubernetes 是这些场景中的关键工具，这就是为什么它们是我们的重点领域。

For Terraform, the 2025.2 update brings:

对于 Terraform，2025.2 更新带来了：

Better support for provider-defined functions:

更好地支持提供程序定义的函数：

Correct syntax highlighting.

正确的语法突出显示。

Quick navigation to provider definitions.

快速导航到提供程序定义。

On-hover documentation.

悬停文档。

Default code formatter.

默认代码格式化程序。

Auto-completion for object variables.

对象变量的自动完成。

Specific support for the Keycloak provider.

对 Keycloak 提供程序的特定支持。

Support for resources in Google Cloud Beta.

支持 Google Cloud Beta 版中的资源。

Ability to use for_each in import blocks.

能够在导入块中使用 for_each。

Accurate binary detection on Windows.

Windows 上的准确二进制检测。

Terraform file recognition.

Terraform 文件识别。

For Kubernetes, there are:
对于 Kubernetes， 有：

Improved port-forwarding with automatic suggestions for ports and real-time checks for availability.

改进了端口转发，自动建议端口并实时检查可用性。

Better performance when working with large log files.

处理大型日志文件时性能更好。

More information about pods.

有关 Pod 的更多信息。

Auto-marking of the active cluster.

活动集群的自动标记。

On top of that, this update improves the GitLab CI integration with:

最重要的是，此更新改进了 GitLab CI 与以下集成：

Smarter support for CI language variable expressions.

更智能地支持 CI 语言变量表达式。

Rich code completion.

丰富的代码补全。

On-hover documentation.

悬停文档。

Support for injecting shell code directly into your pipelines.

支持将 shell 代码直接注入管道。

Debugger(调试器)

Debugging and profiling are essential tools for developing high-quality software, and we strive to make sure that IntelliJ IDEA’s debugger remains best in class. On top of the huge rework to provide a native debugging experience when working remotely, we’ve also fixed issues and provided additional capabilities for existing use cases.

调试和分析是开发高质量软件的重要工具，我们努力确保 IntelliJ IDEA 的调试器保持同类最佳。除了在远程工作时提供原生调试体验的大规模返工之外，我们还修复了问题并为现有用例提供了额外的功能。

You can now evaluate code on any application pause during execution, not just at breakpoints.

现在，您可以在执行期间的任何应用程序暂停时评估代码，而不仅仅是在断点处。

The Analyze Stack Trace or Thread Dump action can now analyze outputs from the jstack and jcmd tools, including in the new JSON format.

“ 分析堆栈跟踪”或“线程转储 ”作现在可以分析 jstack 和 jcmd 工具的输出，包括新的 JSON 格式的输出。

The Collect Thread Dump action now works with the Eclipse OpenJ9 JVM.
C
ollect Thread Dump 作现在可以与 Eclipse OpenJ9 JVM 一起使用。

GraalVM debugger(GraalVM 调试器)

Compiling JVM applications to native code significantly reduces CPU and memory usage, but debugging native images remains challenging. In this release, we’ve improved the native image debugger:

将 JVM 应用程序编译为本机代码可显着降低 CPU 和内存使用量，但调试本机映像仍然具有挑战性。在此版本中，我们改进了本机映像调试器：

Variable rendering provides richer information, for example, full object content and details about proxy objects.

变量渲染提供更丰富的信息，例如，完整的对象内容和有关代理对象的详细信息。

Expression evaluation has been enhanced, making reflection calls such as clazz.getMethod(“myMethod”).invoke(instance) work correctly.

表达式评估已得到增强，使反射调用（例如 clazz.getMethod(“myMethod”).invoke(instance) 正常工作）已得到增强。

Profiler(分析器)

The profiling experience depends on the Java version, and there used to be cases when it didn’t work with Java 24 and the upcoming Java 25. We fixed the failures and implemented support for new language features. You can now reliably profile your application on any Java version and also easily integrate Async-Profiler 4 into your workflow.

分析体验取决于 Java 版本，过去曾有过不适用于 Java 24 和即将推出的 Java 25 的情况。我们修复了故障并实现了对新语言功能的支持。现在，您可以在任何 Java 版本上可靠地分析您的应用程序，并轻松地将 Async-Profiler 4 集成到您的工作流程中。

Performance(性能)

Terminal(终端)

In 2025.2, we’ve enabled the reworked terminal for all users. This change brings significant improvements in rendering and overall terminal performance. For bash and zsh, it also brings some minor visual changes, but without compromising the shells’ behavior.

在 2025.2 中，我们为所有用户启用了重新设计的终端 。这一变化带来了渲染和整体终端性能的显著提升。对于 bash 和 zsh，它还带来了一些细微的视觉变化 ，但不会影响 shell 的行为。

Working with data sources(使用数据源)

When working with large databases, the IDE used to load the entire schema, which produced overhead and was rather slow. Now, PostgreSQL and Microsoft SQL Server are introspected by levels, so only specific parts of the schema are loaded, reducing this overhead.

在处理大型数据库时，IDE 过去常常加载整个模式，这会产生开销并且速度相当慢。现在，PostgreSQL 和 Microsoft SQL Server 是按级别进行内省的，因此仅加载模式的特定部分，从而减少了这种开销。

PostgreSQL users also benefit from smart refresh, which updates only the objects affected by DDL changes.

PostgreSQL 用户还受益于智能刷新，它仅更新受 DDL 更改影响的对象。

Developer experience(开发人员体验)

We have also invested effort into small UX improvements, especially in areas related to reading code. For example, we’ve reworked the Parameter Info popup to make it cleaner, more readable, and easier to use across all supported languages.

我们还在小的用户体验改进上投入了精力，特别是在与阅读代码相关的领域。例如，我们重新设计了 “参数信息 ”弹出窗口，使其更干净、更易读，并且更易于在所有支持的语言中使用。

其他

IntelliJ IDEA 2025.2 版本的概述就先这些了。立即更新到 IntelliJ IDEA 2025.2，关注薇亻言工号【程序视点】，回复vip，领取优惠激活！大家对此版本中的修复和优先级的看法，都可以向官方反馈！

---

最后

【程序视点】助力打工人减负，从不是说说而已！

如果你觉得这篇教程有帮助，别忘了【点赞+收藏+关注】三连支持！

后续安戈会持续分享更多开发工具和技巧，敬请期待！如果有其他工具需求，欢迎留言讨论~ 🚀