antangelo wrote: > LGTM modulo nits. Can we observe that through sema/pch tests?
This functionality isn't yet observable in code (i.e. through lit tests), so it can't be tested through PCH tests on its own (which is primarily why I haven't split this code off sooner). In the CTAD from inherited constructors patch, there is observable behavior determined by the deduction guide source, so there are PCH tests in that patch for this functionality. When the remaining functionality for alias template CTAD is implemented to create alias deduction guides for explicit deduction guides declared after the initial CTAD usage, this functionality will also be visible there. It is possible to test this as-is through unit tests (like the coverage in the AST importer). Are there other spots in the unit test suite that this should be tested? https://github.com/llvm/llvm-project/pull/123875 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
