Possible to map trace_state output back to code?

Stream: new members

[Alex Gu]

I want to see the state at several points in a Lean proof. I'm putting trace_state to capture the states, then executing the code with lake env lean xxx.lean, and the outputs all get printed together (as well as other error messages, etc).

What's the nicest way to be able to differentiate which part of the output came from each trace_state block?