ajordanr-google wrote: This apparently broke a downstream edge case with fuschia: https://issues.fuchsia.dev/397455029, where they set all but the `TINFO_LIBRARIES` variable.
I have given them a downstream fix, but wondering if I can rework the patch so that this isn't an issue. I don't think we need to revert, given this only happens with edge-case defines and it's easily fixed downstream by setting `TINFO_LIBRARIES = CURSES_LIBRARIES`. I'm thinking we could make it easier for folks who hit similar errors by doing one of the following: 1. Checking that when `CURSES_LIBRARIES`, see if it has the symbols. If not, then check if `TINFO_LIBRARIES` is set. Error out if not. 2. Don't check for the existence of `TINFO_LIBRARIES` at all, assume `CURSES_LIBRARIES` is set "correctly" for their use case. I don't really have a preference here. I also notice with this commit that we're changing `CURSES_LIBRARIES` to be a path to a list of paths. Not sure if anyone depends on that fact? Do we care? https://github.com/llvm/llvm-project/pull/126810 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
