labath wrote: I can sort of see your point, but I also disagree with it. :)
It's true that in this case (the blocking version of Socket::Accept), a timeout argument to MainLoop::Run would be sufficient (and probably look cleaner), but it's not very generic. Like, imagine you had a "proper" long-lived MainLoop object (like the one we have in lldb-server), and you wanted to use that to wait for a connection. You couldn't pass a timeout to the Run function there. Instead, you'd probably schedule a callback to unregister the accept handle after a certain amount of time. In this mindset, I view the blocking Accept function as a sort of a convenience wrapper on top of that functionality and its goal is to provide a simple interface for the cases where all you want to do wait for a connection. https://github.com/llvm/llvm-project/pull/117691 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
