JDevlieghere wrote: I'm torn: on the one hand this is obviously making things better, on the other, it undermines what an assertion and has the potential to be used where error handling is necessary. The idea behind the latter is that if your assertion fails, you're already in an unrecoverable state (and if it were recoverable, error handling would have taken care of that). Not coincidentally I had a similar discussion with @labath at the dev meeting last month in the context of `lldbassert`.
The [contribution guidelines](https://lldb.llvm.org/resources/contributing.html) say that `lldbassert` "[...] should be replaced by other means of error handling." Is this sufficiently different that we think this is warranted? https://github.com/llvm/llvm-project/pull/71175 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
