augusto2112 added inline comments.
================ Comment at: lldb/include/lldb/Target/Target.h:496 + std::once_flag target_once_flag; // These two functions fill out the Broadcaster interface: ---------------- kastiglione wrote: > did you mean to use this, or should it be deleted? Leftover from a different implementation, I'll delete it. ================ Comment at: lldb/source/Commands/CommandObjectDWIMPrint.cpp:129-130 + static llvm::SmallPtrSet<Target *, 4> targets_warned; + if (targets_warned.contains(target_ptr)) + return; + ---------------- kastiglione wrote: > question: would people want to be warned once per target, or once per type? > > If I do like the warning, I might want to know every time I `po` a type that > doesn't support it. In my opinion once per target should be enough, if we do it for every type that could potentially be too noisy, and hopefully after seeing the hint a few times users will start remembering to try `p` too. ================ Comment at: lldb/source/Commands/CommandObjectDWIMPrint.cpp:158-162 + StreamString temp_result_stream; + valobj_sp->Dump(temp_result_stream, dump_options); + llvm::StringRef output = temp_result_stream.GetString(); + maybe_add_hint(output); + result.GetOutputStream() << output; ---------------- kastiglione wrote: > what do you think of passing in the `result`'s stream into `maybe_add_hint`? > Perhaps I am overlooking something, but I wonder if it would simplify the > code to reuse the one stream, instead of separating and then combining two > streams. I need the two streams to print it in the correct order (hint first, result later) Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D153489/new/ https://reviews.llvm.org/D153489 _______________________________________________ lldb-commits mailing list lldb-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
