[Lldb-commits] [lldb] [LLDB] Add `ScalarLiteralNode` and literal parsing in DIL (PR #152308)

[Ilia Kuklin via lldb-commits]

kuilpd wrote:

> I have a feeling that this would actually be better off as two node types (float and integer), as each stage (lexer, parser and evaluator) handles the two differently. Maybe then the integer node could use llvm::APInt directly, which would make the `is_unsigned` field look less out of place.
> 
> The implementation makes sense to me, but I don't know if we've reached consensus that this is the way this should work.

The node will also have boolean literals and nullptr later, they are all stored in a `Scalar`, so I don't think we should do 4 nodes for this. I can add them in this PR as well.

We didn't really discuss how literals are parsed and stored, only the math and type promotion part, which I will get to in the next PR.

https://github.com/llvm/llvm-project/pull/152308