GitLab

Add a workaround in case pattern compilation failed (after typechecking) due to the presence of type variables in some types.

Fix a stupid typo which made the constraint solver prefer monomorphic variables over polymorphic ones.
This prevented one from applying a polymorphic function in the body of another function.

Make it so that the substitution function preserves as much sharing as possible from the original type.

This reverts commit a8ba6ab6.

This commit introduces a regression where suprious type variables are introduced in the final type.

Seal the representation of SortedList.Make(X).t (by making the type private). Expose Var.Set as a SortedList.S

Perform a more aggressive memoization of substituted types and register the name if the original type was named.
Make the pretty-printer re-entrant.

containing variables but no toplevel variables would not be printed as
regular expressions but nested pairs.

sequence type would not show up (e.g. 'a & [ Any *] would be displayed
as [ Any * ], while the 'a would still be present internally).

Positive.solve(Positive.decompose t) was not the identity but added spurious types at toplevel.

Added more test cases in poly-ok.cd to cover such regressions

type has already been normalized. Without this patch, in the following usage:

norm(T1, delta)
norm(<a>[ T1 ], delta)

the fact that T1 has already been normalized against some delta is not taken into
account while normalizin <a>[T1].

table instead of a 'recursion set' like in the paper.  The hash table
is tricky:

 - its keys should be both the type we are normalizing and
delta

 - we should store a flag together with the constraint set, indicating
whether the computation of the normalization for the corresponding
type has finished. If it has, we can use the associated constraint set
instead of CS.sat and stop recursion. If it has not, we can return
CS.sat (like the previous base case).

We also update the test case files to check that everything is in order:

- part2.cd has been rewritten to make use of the new syntax and remove
  the red-black trees examples that are now in a separate file

- red-black.cd is a fully typechecking file
- rb-fail.cd has the type definition and the wrong balance function.

Fix variable ordering problem by adding a 'kind' field inside variables so that variables of the function type are processed before arguments types and after Gamma.

This reverts commit c9ad3e5c.

Conflicts:
	types/var.ml
	typing/typer.ml

All problems related to the introduction of the new syntax for
constructed types

Add a global variable table, so when we call "Var.mk id" twice, the
same variable is returned and when we call "Var.fresh v" we always
return a truly globally fresh variable.

- type t 'a = ('a,'a)
- type t ('a,'b) = ('a,'b)

Fix Typer.pp_env printer for types
Minor code refactring

Conflicts:
	Makefile.distrib
	depend

Change $$ in >> in types.ml for the substitution operator

Do not generate equations for monomorphic variables, during the generation of type equation from constraints.

- Var identifiers are now of type U.t instead of string
- Remove TVar from ast. Polymorphic variables are just types

- New fresh variables now share the same str, but different freshness index
- Remove the function is_internal from var module

Now gamma is correctly propagated in the compilation Enviroment
and substitions are correctly applied to abstractions