[Lldb-commits] [PATCH] D138539: Use std::nullopt_t instead of NoneType (NFC)

2023-03-02 Thread Balázs Benics via Phabricator via lldb-commits
steakhal added a comment. Herald added a subscriber: thopre. This patch breaks `llvm::StringSet` equality. The following code would no longer compile: llvm::StringSet LHS; llvm::StringSet RHS; bool equal = LHS == RHS; Such code might be used as gtest assertions like `EXPECT_EQ(LHS, RHS)`.

[Lldb-commits] [PATCH] D138539: Use std::nullopt_t instead of NoneType (NFC)

2023-03-02 Thread Balázs Benics via Phabricator via lldb-commits
steakhal added a comment. Oh, I forgot to mention why this change breaks the equality operator of `llvm::StringSet`. It's because the `StringMap::operator==` will try to compare the `value` as well, which is of type `std::nullopt_t` and that has no equality comparison operator. Previously, the

[Lldb-commits] [PATCH] D138539: Use std::nullopt_t instead of NoneType (NFC)

2023-03-02 Thread Balázs Benics via Phabricator via lldb-commits
steakhal added a comment. Hm, it would be easier to use `llvm::StringMap` instead of `std::nullopt_t`, or some adhoc empty struct. Repository: rG LLVM Github Monorepo CHANGES SINCE LAST ACTION https://reviews.llvm.org/D138539/new/ https://reviews.llvm.org/D138539

[Lldb-commits] [PATCH] D139881: [clang] Use a StringRef instead of a raw char pointer to store builtin and call information

2022-12-13 Thread Balázs Benics via Phabricator via lldb-commits
steakhal added a comment. The `StaticAnalyzer` portion looks good to me AFAICT. Comment at: clang/lib/StaticAnalyzer/Core/CallDescription.cpp:39 ento::CallDescription::CallDescription(CallDescriptionFlags Flags, - ArrayRef QualifiedName, +