Unifying RTParserBuilder and CTParserBuilder

[ratmice]

This writing down some thoughts so they don't get lost.

in PR #550 we had when trying to test the change, briefly toyed around with adding a callback mechanism which
returned a result and all the parameters needed to build an RTParserBuilder, this was dropped from the patch pretty
quickly in favor of just having a #[cfg(test)] method on the builder, as it was fairly beyond the scope of the patch.

The impression that the implementation gave me personally was that it required a lot of arguments to build a RTParserBuilder.
One way that that could be alleviated is just building an RTParserBuilder, and passing it directly to the callback.

On top of that though this involves linker/parser interaction with the token map, so it isn't clear to me that just reducing the number of parameters is going to get things to a mergeable quality.

[ltratt]

Working out a nice link/API between parser and lexer has been an ongoing challenge in grmtools.

In retrospect, a possibly fateful decision was to consider the two things separate. This means that the API of one would, in a sense, have to lord it over the other. Perhaps -- and I haven't thought this through! -- the better solution might be to have some sort of independent, intermediary thingy which coordinates the two. It might not actually do much beyond that coordination.

[ratmice]

Yeah, I also want to say this isn't something I'm too worried about, in case we just want to actually close this.
But still wanted to remind myself of giving one more shot at.

[ratmice]

Yeah I kind of get what you mean, the big issue with all this more than anything else seems like it should be the
lack of circular dependencies, but I guess what I am wondering is since the token_map seems like it should operate entirely though core:: types it perhaps could be done through a CTLexerBuilder -> CTParserBuilder callback-like interface.

I should say though I struggle with resolve the circular dependency problem with the independent, intermediary thingy idea.

Anyhow, I'll do a little bit of playing to see if I can't find something that feels good and doesn't require any acts contortion.
Otherwise I'll probably try and not worry about it.