ojhunt wrote: > @ojhunt but the expression here seems valid? it just does not have a type > yet. So the fix seems reasonable?
I thought this was incorrect handling of the injected code_completion pseudo instruction, but it's not. I feel like the fix belongs in `CodeCompleteObjCInstanceMessage`, but this is the only code path that can hit this null - every other path has real types assigned to the subject so maybe that isn't warranted? https://github.com/llvm/llvm-project/pull/180343 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
