Peter Hill

I removed `flibs` from my corpus as it has too many files using a custom preprocessor. `WRF` also uses a custom preprocessor for at least one of its submodules, so...

Great idea!

I think the extra node here would actually be useful in order to have the `(size)` node be a child. For example: ```fortran integer :: a(n), b ``` gets parsed...

@stadelmanma I finally revisited this to rebase it onto `master`. I'm not sure about the `variable` node -- it does make it nicer when grouping name + size for arrays,...

I also have something for parsing string literals with kinds, except that it eats the `_` as part of the kind identifier, which is a little annoying

> I think the benefit of having (variable (identifier) (size (number_literal))) is worth the added noise you get when seeing (variable (identifier)) in the parse tree. It also adds a...

Ok, this took way, way too much effort for the actual payoff, but I now have working string literal kind parameters. We're now parsing 99% of files without error (97%...

Ok, trying not to fiddle with this any more! I just removed the `(argument_list)` node from appearing in `(kind)`. Basically, parse the following: real(real64) :: foo as (variable_declaration (intrinsic_type) (kind...

Having used this in practice a little, I'm still not completely happy: ```fortran real(real64) :: foo = 1.0_real64 ``` gets parsed as ```text (variable_declaration type: (intrinsic_type) type: (kind (identifier)) declarator:...

After playing around with trying to use various versions, I've settled on this: ``` (variable_declaration type: (intrinsic_type kind: (kind (keyword_argument name: (identifier) value: (number_literal)) (keyword_argument name: (identifier) value: (identifier)))) declarator:...