Why is modular code so hard to implement? - Programming On Unix

Users browsing this thread: 1 Guest(s)
venam
Administrators
It's sometimes called the microkernel architecture, related related to the easiness to which you can add modules, and that the compiler/preprocessor will automatically pick-up when they are added.


I think it comes from the phases in the compiler/interpreter process and the early design decisions in the language. It's still easier to do this in Rust than in many other languages, like you mention.
In the Java world it is popular to do it through dependency injection (DI), which as I'm discovering now there's a bunch of languages built precisely around the concept. While it's not really necessary to have DI to do automatic plug-in discovery, it's one of the way I can think of that makes it automatic.


Messages In This Thread
RE: Why is modular code so hard to implement? - by venam - 12-11-2020, 06:18 AM