Access Control Syntax

Summary

I’m still tinkering on a scripting language for my hobby fantasy console project. I’m ashamed to admit this, but up to this point, the language had absolutely no notion of modules. Literally every source file is dumped into one big global namespace and compiled together. I always planned to have some sort of module system. I just hadn’t figured it out yet because I had other, harder language design problems to solve. I assumed that the module system mostly didn’t interact with other language features, so I could kick it down the road for now. That was true until it wasn’t. I’ve been beating my head against the wall around generics for.. oh God I just checked the Git history and it’s three years now. I still don’t have that pinned down. Parametric types are hard. Anyway, one of the approaches I’m exploring does get tangled up in modules and scoping so now I have to figure modules out. This post is about one little syntax design question I ran into: how do you distinguish public and private declarations? A basic module system Since my language is a scripting language, my ambitions for the module system are pretty minimal. Think more like Python or Dart than Java or C#. Every file has its own top-level scope that isn’t shared with others. If you want to access top-level declarations from another file, you import that file. That makes its top-level declarations available in the importing file. Of course, a module might have some declarations that are only for its own internal use and should not be made available when you import it. A module should be able to encapsulate parts of its implementation. Thus, I need a way for users to indicate which declarations are private and which are public. What other languages do Every language out there has some kind of module system and an ability to control access (though it did take JavaScript about 20 years to get there, bless its heart). In some sense, it’s a solved problem. But they don’t all solve it the same way, especially if...