(KernelDSL) MethodError raises uninformative error message

[femtomc]

WARNING: replacing module GenInferenceTunerTour.
ERROR: LoadError: MethodError: no method matching var"##reverse_kernel#304"(::Gen.DynamicDSLTrace{Gen.DynamicDSLFunction{Any}}, ::Int64, ::Matrix{Float64})┌ Error: Error showing method candidates, aborted
│   exception =
│    could not determine location of method definition
│    Stacktrace:
│      [1] error(s::String)
│        @ Base ./error.jl:33
│      [2] functionloc
│        @ ./methodshow.jl:164 [inlined]
│      [3] show_method_candidates(io::IOContext{Base.TTY}, ex::MethodError, kwargs::Any)
│        @ Base ./errorshow.jl:499
│      [4] showerror(io::IOContext{Base.TTY}, ex::MethodError)
│        @ Base ./errorshow.jl:318
│      [5] showerror(io::IOContext{Base.TTY}, ex::MethodError, bt::Vector{Base.StackTraces.StackFrame}; backtrace::Bool)
│        @ Base ./errorshow.jl:88
│      [6] showerror(io::IOContext{Base.TTY}, ex::LoadError, bt::Vector{Base.StackTraces.StackFrame}; backtrace::Bool)
│        @ Base ./errorshow.jl:96
│      [7] show_exception_stack(io::IOContext{Base.TTY}, stack::Vector{Any})
│        @ Base ./errorshow.jl:866
│      [8] display_error(io::IOContext{Base.TTY}, stack::Base.ExceptionStack)
│        @ Base ./client.jl:104
│      [9] #invokelatest#2
│        @ ./essentials.jl:716 [inlined]
│     [10] invokelatest
│        @ ./essentials.jl:714 [inlined]
│     [11] print_response(errio::IO, response::Any, show_value::Bool, have_color::Bool, specialdisplay::Union{Nothing, AbstractDisplay})
│        @ REPL /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/REPL.jl:288
│     [12] (::REPL.var"#45#46"{REPL.LineEditREPL, Pair{Any, Bool}, Bool, Bool})(io::Any)
│        @ REPL /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/REPL.jl:277
│     [13] with_repl_linfo(f::Any, repl::REPL.LineEditREPL)
│        @ REPL /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/REPL.jl:510
│     [14] print_response(repl::REPL.AbstractREPL, response::Any, show_value::Bool, have_color::Bool)
│        @ REPL /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/REPL.jl:275
│     [15] (::REPL.var"#do_respond#66"{Bool, Bool, REPL.var"#77#87"{REPL.LineEditREPL, REPL.REPLHistoryProvider}, REPL.LineEditREPL, REPL.LineEdit.Prompt})(s::REPL.LineEdit.MIState, buf::Any, ok::Bool)
│        @ REPL /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/REPL.jl:846
│     [16] #invokelatest#2
│        @ ./essentials.jl:716 [inlined]
│     [17] invokelatest
│        @ ./essentials.jl:714 [inlined]
│     [18] run_interface(terminal::REPL.Terminals.TextTerminal, m::REPL.LineEdit.ModalInterface, s::REPL.LineEdit.MIState)
│        @ REPL.LineEdit /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/LineEdit.jl:2493
│     [19] run_frontend(repl::REPL.LineEditREPL, backend::REPL.REPLBackendRef)
│        @ REPL /Applications/Julia-1.7.app/Contents/Resources/julia/share/julia/stdlib/v1.7/REPL/src/REPL.jl:1232
│     [20] (::REPL.var"#49#54"{REPL.LineEditREPL, REPL.REPLBackendRef})()
│        @ REPL ./task.jl:429
└ @ Base errorshow.jl:320

Stacktrace:
  [1] assess(gen_fn::GenSMCGF.SMCGF{Any, Gen.DynamicDSLTrace, Gen.DynamicDSLFunction{Any}, Nothing, Gen.DynamicDSLFunction{Any}, Nothing, Nothing, typeof(Main.GenInferenceTunerTour.DiscreteHMM.k1), Main.GenInferenceTunerTour.DiscreteHMM.var"###reverse_kernel#304"}, args::Tuple{Vector{Gen.ChoiceMap}, Vector{Tuple{Int64, Main.GenInferenceTunerTour.DiscreteHMM.DiscreteHMMConfiguration}}, Vector{Tuple{Gen.IntDiff, Gen.NoChange}}, Vector{Tuple}, Vector{Tuple}, Int64, Int64}, choices::Gen.DynamicChoiceMap)

Two possibilities:

1. This is just an uninformative error message for a failure to match method dispatch.
2. There's an escape/hygiene issue when expanding the kernel DSL into a module, and then calling code from that module.

[femtomc]

This is number (1)

Yikes. I'm keeping this issue open for a future hackathon -- it would be great if we could alleviate this insane stacktrace for method dispatch errors.

[alex-lew]

@femtomc I don't totally follow — is the Kernel DSL still broken in some way? If not, can we rename this issue, and add some description of what problem it's pointing out? (i.e., better error messages in general, or for a specific kind of problem?) Thanks!

[femtomc]

@alex-lew Thanks for keeping track. Yes, it's not broken -- but I found that if a MethodError is thrown, this LoadError is thrown, which doesn't match the "normal" MethodError. I suspect this might be confusing -- and indeed, it confused me when I first saw it.