Okay so I think the best thing to do for parsing to just have a single class for parsing then just build up on it perhaps. That way there is just say a single parser that can parse multiple things and just call each other as needed.
Okay so, what if I split into packages the different parts of the file. Basically I have a package for the class part, then the code parts, and other such things. This way it is not everything thrown into one package.
What if I made like a static class parser kind of thing. Give it a bunch of tokens and it just tries to parse everything and such based on a given structure.
I think splitting the parser from the structure would be the best thing to do so that it is not really that ugly. Might be best with static methods although single objects could be used.
I do not really know the best way to structure the parser so that the code is clean. And today I just feel completely horrible and worn out due to stress induced by "family". I just cannot think clearly at all.
Okay, how about instead there is just a "simple" class layout. basically I cut out a bunch of the lexical structure stuff to simplify it as much as possible. This way at least initially I only need to worry about the initial structure of a file and not all the possible sub-expressions. I think this would be the easiest. Then what I can do instead is to have it where I just worry about modifiers, parameters, and members. This would simplify things greatly. When I parse a member I could just ignore everything inside of it for the most part until those things, if they actually are, needed.