Ethin Probst

> Interesting task you set yourself! Will this be limited to UTF-8 encoding only? Yes; I could expand it to UTF-16 and UTF-32 later, though that wouldn't be very hard...

Yeah so I don't know how well this is going to work. Is there a way I can include rules from other parsers? Or write rules in a separate file?...

This is particularly problematic since the Ada specification uses the graphic character alias category for character literals, which, according to Wikipedia, is "those with General Category Letter, Mark, Number, Punctuation,...

@zevv Oh good. Though figuring out how to debug this is going to be an utter pain... Especially if it doesn't compile. I really hope I got the syntax right...

Most likely, yes. I've written a compiler before (not the cleanest, and for University) but we didn't cover ASTs/semantic analysis/code generation as it is done now. I don't know how...

@zevv Well we might have an issue: when I build this small parser I get this error in debug builds of the compiler, and in release builds it crashes: Error:...

I would love if this were added. Particularly since to my knowledge the Intel C/C++ compilers are better at codegen for x86/x86-64 than MSVC is.

@badicsalex I would think it would, but I've found that for the particular grammar I'm working with it only makes the resulting AST look really weird. For example, if I...

It works for simple things, e.g. `SubtypeMark = > Name`, but not for ordered choice productions, which is probably where it'd get used most since that's where the AST gets...

@badicsalex Post-processing on directives and rejecting unrecognized directives and saying "Unknown directive x" would be better.