Closed kjcjohnson closed 3 years ago
Breaking changes:
Semgus.Syntax.SemanticRule
's Predicate
member is now Predicates
and is a list of possible predicates.Semgus.Syntax.SynthFun
now has all variables in its closure marked as auxiliary variables, instead of inputs or outputs.I think everything else should function as expected, although the usage of the constraints may need to change.
Note: this is pretty hacked together. Once v1.0 is released, we'll need to do a restructuring of how the LanguageEnvironment
and VariableClosure
s co-exist with the SemgusProblem
.
This has partial, experimental support for annotations on the semantic relation instances (a.k.a. CHC conclusions). Not attaching #10 because it's not good enough yet, but still passable for now.
Main updates to support the v1.0 syntax.