4 days ago

Summary of Progress (October 20 - October 27)

This week is a continuation of the work done to break the dependency of dmd/mtype.d on semantic functions. mtype.d (as we speak) is essentially dependent on three semantic files: dsymbolsem, enumsem, and typesem. The first two are almost trivial (one function is imported from each) and they have already been eliminated in the following PRs:

The final dependency (on typesem) is tricky not in the complexity of the refactoring but in the number of changes needed. The work has already started with moving the static function Type._init to typesem (see Merged PR) and I should continue eliminating the rest of imports over the course of the next week or so.