11:07:10scymtymbeach: maybe the s-expression-syntax library can offer a builder mixin or a builder "wrapper" that canonicalizes declaration specifiers. that solution would get independence of the input expression representation for free. one downside i see is that there would have to be at least one new generic function that would not be fully coherent with the rest of the library
11:07:44scymtyma builder based solution would also work as a separate library, of course, but the coupling between the two libraries would be pretty tight
11:08:14beachI see yes. I am working on a separate library (and I am probably going to finish it later today).
11:08:23beachWe will see where to put it ultimately.
11:10:54beachIt abstracts out FIRST, REST, etc. So it can handle both s-expressions and CSTs.