The C Preprocessor also does all of those things. That’s expected of a preprocessor. If you say “you can write java code with a bit of extra sugar and JPlus turns it back into ‘regular’ java,” then that’s a preprocessor.

A simple preprocessor only performs code transformation and cannot analyze the meaning of the code or ensure type safety. However, JPlus goes beyond mere transformation by providing static analysis capabilities, such as generating a parse tree and checking nullability. It also includes functionality to automatically generate necessary Java code through the apply syntax. As a result, the combination of performing nullability static analysis and generating code via apply cannot be expressed or handled using any existing Java syntax alone.

JPlus follows standard Java syntax while aiming to be an “Upgraded Java” by adding features that Java developers consider practically useful in real-world projects. This allows existing Java developers to utilize new features with minimal learning curve. All features are added on top of Java syntax. For example, null-safety syntax (type?, ?.) and boilerplate code generation syntax (apply). As a result, developers can experience an enhanced version of Java while continuing to use existing Java code and libraries without modification. This should clarify exactly where JPlus fits in the ecosystem.

Thank you for the excellent question.

1. How does JPlus handle null safety in Java library interfaces?

• JPlus is currently implemented at the MVP level. Therefore, all Java library interfaces are treated as non-null by default. As a result, developers need to manually check for nullability when interacting with Java libraries.

• When referencing Java library objects, always declare the variables as nullable (type?) and use the null-safe operator (?.) when accessing them to prevent null pointer exceptions (NPEs).

• In future implementations, we plan to leverage nullability annotations in Java library code to add null-checking logic. However, since not all Java libraries include such annotations, developers will still need to manually handle null checks. The same applies to Kotlin: platform types are used, so developers are responsible for performing null checks.

2. If nullability information is a type metadata extension for compile-time checking, does that inevitably break on library interfaces when I create both library and consuming app with JPlus?

• In the current JPlus MVP, when converting to Java code, nullability information specified in the JPlus files is not converted into @Nullable/@Nonnull annotations.

• Therefore, using static analysis tools that rely on these annotations may cause issues at compile time.

• This feature will be supported in the next version, which will include conversion to @Nullable/@Nonnull. Thank you for your valuable feedback.

Finally, for faster responses, please post any future questions on the JPlus GitHub Discussions page: https://github.com/nieuwmijnleven/JPlus/discussions.

Didn’t I make myself clear? If you find it uncomfortable because you think it’s an advertisement, just move along. What do you think you’re doing? Look at the project I’m working on before you speak. You have no idea how much effort has gone into it. Calling it a scam is truly absurd.

The idea might be enough. Lots of companies running legacy code would be interested in this idea since it would make maintaining/patching it easy.

Thank you for your response. I will take your valuable feedback into careful consideration.

First of all, thank you for your interest in the JPlus project. I also apologize if anything I said earlier came across as rude.

The point I think was misunderstood is that, while the content of the posts is similar, the links they contain are different. To clarify once again, the link in this post directs to the JPlus GitHub repository, whereas the link in the post you mentioned points to the JPlus blog on Hashnode. Please check again.

Yoy won’t find your target audience here on lemmy.

Instead you should look for companies that have open job/freelancer positions for maintaining legacy java code and pitch your project to them.

That’s a great idea. Thank you. However, I’m not sure if such opportunities would be available at my current stage.

You can probably adopt both for a large existing project. Old files get compiled with JPlus, new files will be written in Kotlin. Old files can also gradually be ported to kotlin.

Exactly, that could work. You can keep the existing files as they are and compile them with JPlus, while writing new modules in Kotlin to adopt it gradually. JPlus can serve as a stepping stone before moving fully to Kotlin. However, converting all Java code in an existing project to Kotlin would not only carry significant risks but also be costly. With JPlus, you can fully leverage the proven existing Java codes.

Please check this post: Making your Java code null-safe without rewriting

Is this your first time here? The link in this post leads to the GitHub JPlus repository, while the other posts ultimately link to explanations on how to handle null safety using the JPlus IntelliJ plugin. Aside from the brief introduction, check out the additional links.

Groovy doesn’t have null safety or boilerplate code generation features. That’s why JPlus was created to address what Java developers truly feel is missing. Go post that comment in the Groovy community.

“Yeah, Kotlin. The real The End”, just kidding :)

As the title suggests, this page is an introduction to the project, while the other posts focus on how to use it.

AFAIK Kotlin and Java code can co-exist as source level. Never tried it though. I guess it depends on the end case scenario what to use.

JPlus retains almost all of Java’s syntax while providing conveniences like null checks, allowing Java developers to use it immediately without additional learning. In contrast, Kotlin can utilize Java classes but requires learning Kotlin syntax and its libraries, and it cannot use Java syntax directly or automatically guarantee null safety. Therefore, JPlus offers a distinct advantage over Kotlin.

Things can get confusing if the explanation is too long, so here’s a concise way to put it.

It’s not entirely clear whether Groovy includes 100% of Java syntax, but in my view, Groovy focuses on enhancing the language itself, addressing areas where it falls short compared to modern languages.

On the other hand, JPlus is similar to how TypeScript addresses JavaScript’s lack of type safety: it focuses on strengthening Java’s safety while improving developer convenience through features like boilerplate code generation. Importantly, JPlus code always compiles down to standard Java, which means it can be integrated seamlessly into existing Java projects without changing the build environment or toolchain.

In short:

Groovy: Focuses on enhancing the expressive power of Java, adding modern language features and syntactic flexibility.

JPlus: Focuses on enhancing language safety and developer convenience, while maintaining full compatibility with Java. This allows developers to adopt JPlus without worrying about breaking existing Java projects.

From Wikipedia:

In computer science, a preprocessor (or precompiler)[1] is a program that processes its input data to produce output that is used as input in another program. […], which is often used by some subsequent programs like compilers.

The emphasis is mine.

Both Typescript and SASS are examples of preprocessors.

By that logic, the C compiler would also be a preprocessor since it converts C code into assembly. Simply calling something a preprocessor just because its output is source code is not logically correct. The same applies to JPlus: the fact that it ultimately produces Java source code does not make it a preprocessor. Internally, it performs compiler-level processes such as AST generation, null-safety checks, and boilerplate code generation, so it should be regarded as a proper compiler.

This isn’t an accusation, but was this comment written with AI? There’s a glaring logical error here which I think a human would catch easily, but an LLM (which is just a natural language generator, not a logic processor) could possibly overlook.

Specifically, your arguments don’t really make a lot of sense. They’re also not targeted at my claim. It reads more like a defense of JPlus. To which I want to clarify, I merely took issue with the specific claim I quoted, I wasn’t trying to say there’s no point to JPlus. There’s no need to defend JPlus in general. So I’m going to dismiss runtime behaviors since that has nothing to do with the syntax.

Things can get confusing if the explanation is too long, so here’s a concise way to put it.

It’s not entirely clear whether Groovy includes 100% of Java syntax, but in my view, Groovy focuses on enhancing the language itself, addressing areas where it falls short compared to modern languages.

On the other hand, JPlus is similar to how TypeScript addresses JavaScript’s lack of type safety: it focuses on strengthening Java’s safety while improving developer convenience through features like boilerplate code generation. Importantly, JPlus code always compiles down to standard Java, which means it can be integrated seamlessly into existing Java projects without changing the build environment or toolchain.

In short:

Groovy: Focuses on enhancing the expressive power of Java, adding modern language features and syntactic flexibility.

JPlus: Focuses on enhancing language safety and developer convenience, while maintaining full compatibility with Java. This allows developers to adopt JPlus without worrying about breaking existing Java projects.

Kotlin is great for null-safety, but Kotlin isn’t a superset, you can’t just compile a java file with kotlin and have it work.

JPlus allows you to enforce null-safety without rewriting your existing Java code, which can be easier for teams working in legacy projects or who prefer staying in pure Java.

You say:

Groovy is not 100% identical

Then you say:

JPlus aims to keep Java syntax almost intact

Which implies that it is not 100% identical either.

So why do you apply that same argument against one but not the other? Seems like marketing bs.

As you mentioned, the phrasing might make it seem inconsistent. The key difference here is not whether they are “100% identical,” but the degree and nature of the changes.

Groovy introduces dynamic typing, optional syntax, runtime behaviors, and other language features that can fundamentally alter how Java code executes. Even if the code compiles, subtle differences in behavior can occur. In contrast, JPlus preserves nearly all standard Java syntax while only adding compile-time null-safety checks and boilerplate code generation. These changes have minimal impact on the execution of existing Java code and are primarily intended to improve safety and developer convenience.

Additionally, JPlus is currently developed based on Java 20 syntax, with added null-safety and boilerplate code generation features. Therefore, code up to Java 20 is fully compatible. Since future updates will support the latest Java syntax as well, the phrasing “almost identical” is used intentionally. In practice, JPlus will continue to be Java syntax + α.

Technically, neither language is “100% identical” to Java, but the practical impact is very different. Groovy can alter runtime behavior, whereas JPlus preserves existing behavior almost entirely while adding safety and developer convenience. This is not marketing exaggeration, but a reflection of the difference in the invasiveness of changes.

Kotlin isn’t a superset, you can’t just compile a java file with kotlin and have it work afaik. That seems to be the point here.

This is more like a preprocessor anyway, like SASS to CSS. The compiler spits out Java source code, not jvm bytecode.

Exactly, as you said, Kotlin isn’t a Java superset. you can’t just compile a Java file with Kotlin and have it work. JPlus works similarly in that it outputs standard Java source code rather than JVM bytecode.

However, JPlus is not merely a “preprocessor.” It actually parses Java source like a compiler, performs null-safety checks and boilerplate code generation on the generated parse tree, and finally produces standard Java code. In that sense, JPlus should be considered a compiler. The only difference is that its output is Java code; if the code generation step were extended to produce JVM bytecode directly, it could bypass the Java source entirely and generate bytecode straightaway.

The key point is that, just like TypeScript addresses JavaScript’s lack of type safety, JPlus fills in some of Java’s gaps. It allows you to keep almost all of your existing Java code while adding features like null-safety and automatic boilerplate generation, improving both safety and developer convenience.