Re: [Haskell-cafe] Generic Traversal of ASTs

Multiplate actually looks very likely for my use case - I figured there must be some prior work in this area that I had not found yet, so thanks! As sort of a follow up, is there a quick way to try out all of these libraries without having to setup a new stack project every time? I used to use the arch packages to manually invoke ghc on small files, but that only works if the package is on the aur so I was wondering if anyone had any thoughts. Thanks, Richard On Tue, Jun 1, 2021 at 1:17 PM Mario Blazevic wrote:

Have you tried Multiplate?

If Multiplate is not powerful enough and you're willing to make your types more complex, have a look at deep-transformations.

On Mon, May 24, 2021 at 4:52 PM Richard Yu wrote:

...

I have a question about generic traversal/transformation of nested data structures. From what I understand the main options available are SYB, Uniplate, GHC.Generics and Data.Lens (although I am under the impression this requires record types). The data structure I am attempting to traverse is similar to this simplified version:

data SExpr = Expr | Const data Const = B bool | I int data Expr = Lambda String SExpr | If Expr ...

I have attempted to use Uniplate and GHC.Generics to traverse the built AST and modify for example, the string of a Lambda type, but have not been able to figure out how to get any of the libraries to typecheck. The examples or docs I found were not quite enough. Does anyone know of examples that might help or explain how I should be using the generic libraries? In addition if I moved to writing the AST with GADTs, do some generic traversal strategies/libraries stop working?

Many thanks, Richard

_______________________________________________ Haskell-Cafe mailing list To (un)subscribe, modify options or view archives go to: http://mail.haskell.org/cgi-bin/mailman/listinfo/haskell-cafe Only members subscribed via the mailman list are allowed to post.

-- Richard Yu B.S Aerospace Engineering | B.S Computer Science Gemstone Honors Program University of Maryland, College Park https://github.com/beForged