shafik wrote: > > Static analysis flagged that the non-static member Semantics was not > > initialized by the default default constructor. Fix is to initialize it > > using in class member initialization. > > Yeah, this might be one better to mark as Intentional in the static analysis > tool; there really isn't a sensible default value to begin with, so getting > sanitizer coverage seems like a better approach IMO.
Yes but we have a default constructor, then maybe we should not have one then? It looks like it is only used for `TYPE_SWITCH` and in that case if we need it then maybe there should be some explanatory comments detailing why this is so folks finding this in the future can understand the use. https://github.com/llvm/llvm-project/pull/153671 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
