You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Switching the definitions of torus and ellipsoid to be SymbolicObj3
primitives rather than being `implict`ly defined. The maths functions
haven't changed, only been moved around.
One test was updated as some of the STL values changed, but to the eye
it looks similar, and a random check of values indicated that they were
small.
These are currently implemented "inline" using
implicit
which makes them unique and opaque to our AST.This reminded me of the Haskell expression problem https://eli.thegreenplace.net/2018/more-thoughts-on-the-expression-problem-in-haskell/ and one of the solutions Data Types a la Carte
Search for DTalC led me to familiar site https://reasonablypolymorphic.com/blog/better-data-types-a-la-carte/ which offers improved version.
Switching to this representation would allow us and our users to extend our current monolithic AST types with custom ones.
The text was updated successfully, but these errors were encountered: