=?utf-8?q?Balázs_Kéri?= <[email protected]> Message-ID: In-Reply-To: <llvm.org/llvm/llvm-project/pull/[email protected]>
https://github.com/necto requested changes to this pull request. I agree with the approach overall, but I am worried that the particular implementation might make the code less maintainable. It will surprise me that `MapExtDefNamesConsumer` operate on a global shared state. At the same time I believe it is not much more complicated to keep state of each `MapExtDefNamesConsumer` private and ephemeral, and aggregate them into a local variable in `HandleFiles`. Alternatively, they can hold an explicit pointer to the shared collection local in `HandleFiles`. Additionally, I think it would be useful to add a couple of tests demonstrating the following behaviors: - does a symbol defined in another TU override a weak symbol in main TU? - what happens when there are two strong symbols (unless there is already a "duplicate symbol" test, but I failed to find it after a quick look). https://github.com/llvm/llvm-project/pull/222010 _______________________________________________ cfe-commits mailing list [email protected] https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
