Bug when using a DynamicQuantities.QuantityArray based on StepRangeLen as axis

[BambOoxX]

I stumbled upon this behavior while discovering AxisKeys capabilities when trying to create a KeyedArray using a DynamicQuantities.QuantityArray as time axis.

Here is a MWE

#] activate --temp
#] add DynamicQuantities
#] add AxisKeys

using DynamicQuantities
using AxisKeys


t = QuantityArray(range(1,step=0.1,length=10),u"s")
KeyedArray(rand(10,3); time = t, channels = string.(1:3))

# Note that collecting the range works fine
# KeyedArray(rand(10,3); time = collect(t), channels = string.(1:3))

This fails with

2-dimensional KeyedArray(NamedDimsArray(...)) with keys:
↓   time ∈ 10-element QuantityArray{Float64,...}
→   channels ∈ 3-element Vector{String}
And data, 10×3 Matrix{Float64}:
Error showing value of type KeyedArray{Float64, 2, NamedDimsArray{(:time, :channels), Float64, 2, Matrix{Float64}}, Tuple{QuantityArray{Float64, 1, Dimensions{FRInt32}, Quantity{Float64, Dimensions{FRInt32}}, StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}}, Vector{String}}}:

SYSTEM (REPL): showing an error caused an error
ERROR: 1-element ExceptionStack:
MethodError: no method matching StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}(::UndefInitializer, ::Tuple{Int64})
The type `StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}` exists, but no method is defined for this combination of argument types when trying to construct it.

Closest candidates are:
  StepRangeLen{T, R, S, L}(::R, ::S, ::Integer, ::Integer) where {T, R, S, L}
   @ Base range.jl:509
  StepRangeLen{T, R, S, L}(::R, ::S, ::Integer) where {T, R, S, L}
   @ Base range.jl:509
  StepRangeLen{T, R, S, L}(::StepRangeLen{T, R, S, L}) where {T<:AbstractFloat, R<:Base.TwicePrecision, S<:Base.TwicePrecision, L}
   @ Base twiceprecision.jl:539
  ...

Stacktrace:
  [1] similar(::Type{StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}}, dims::Tuple{Int64})
    @ Base .\abstractarray.jl:867
  [2] similar(::Type{StepRangeLen{…}}, shape::Tuple{Base.OneTo{…}})
    @ Base .\abstractarray.jl:866
  [3] similar(bc::Base.Broadcast.Broadcasted{…}, ::Type{…})
    @ DynamicQuantities D:\bamboo\.julia\packages\DynamicQuantities\AQfiw\src\arrays.jl:326
  [4] copy
    @ .\broadcast.jl:919 [inlined]
  [5] materialize
    @ .\broadcast.jl:894 [inlined]
  [6] keyed_print_matrix(io::IOContext{…}, A::KeyedArray{…}, reduce_size::Bool)
    @ AxisKeys D:\bamboo\.julia\packages\AxisKeys\a6h2b\src\show.jl:100
  [7] print_matrix(io::IOContext{REPL.LimitIO{…}}, A::KeyedArray{Float64, 2, NamedDimsArray{…}, Tuple{…}})
    @ AxisKeys D:\bamboo\.julia\packages\AxisKeys\a6h2b\src\show.jl:76
  [8] print_array(io::IOContext{REPL.LimitIO{…}}, X::KeyedArray{Float64, 2, NamedDimsArray{…}, Tuple{…}})
    @ Base .\arrayshow.jl:358
  [9] show(io::IOContext{…}, ::MIME{…}, X::KeyedArray{…})
    @ Base .\arrayshow.jl:363
 [10] show_repl(io::IO, mime::MIME{Symbol("text/plain")}, x::Any)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:548
 [11] show_limited(io::IO, mime::MIME, x::Any)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:517
 [12] (::REPL.var"#display##0#display##1"{REPL.REPLDisplay{…}, MIME{…}, Base.RefValue{…}})(io::Any)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:540
 [13] with_repl_linfo(f::Any, repl::REPL.LineEditREPL)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:833
 [14] display(d::REPL.REPLDisplay, mime::MIME{Symbol("text/plain")}, x::Any)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:526
 [15] display
    @ D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:546 [inlined]
 [16] display(x::Any)
    @ Base.Multimedia .\multimedia.jl:340
 [17] (::REPL.var"#print_response##2#print_response##3")()
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:590
 [18] repl_backend_loop(backend::REPL.REPLBackend, get_module::Function)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:445
 [19] start_repl_backend(backend::REPL.REPLBackend, consumer::Any; get_module::Function)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:427
 [20] start_repl_backend
    @ D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:424 [inlined]
 [21] run_repl(repl::REPL.AbstractREPL, consumer::Any; backend_on_current_task::Bool, backend::Any)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:653
 [22] run_repl(repl::REPL.AbstractREPL, consumer::Any)
    @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:639
 [23] run_std_repl(REPL::Module, quiet::Bool, banner::Symbol, history_file::Bool)
    @ Base .\client.jl:478
 [24] run_main_repl(interactive::Bool, quiet::Bool, banner::Symbol, history_file::Bool)
    @ Base .\client.jl:499
 [25] repl_main
    @ .\client.jl:586 [inlined]
 [26] _start()
    @ Base .\client.jl:561
Stacktrace:
 [1] print_response(errio::IO, response::Any, backend::Union{…}, show_value::Bool, have_color::Bool, specialdisplay::Union{…})
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:599
 [2] (::REPL.var"#print_response##0#print_response##1"{REPL.LineEditREPL, Pair{Any, Bool}, Bool, Bool})(io::Any)
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:558
 [3] with_repl_linfo(f::Any, repl::REPL.LineEditREPL)
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:833
 [4] print_response(repl::REPL.AbstractREPL, response::Any, show_value::Bool, have_color::Bool)
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:556
 [5] (::REPL.var"#do_respond#73"{…})(s::REPL.LineEdit.MIState, buf::Any, ok::Bool)
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:1192
 [6] run_interface(terminal::Base.Terminals.TextTerminal, m::REPL.LineEdit.ModalInterface, s::REPL.LineEdit.MIState)
   @ REPL.LineEdit D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\LineEdit.jl:2854
 [7] run_frontend(repl::REPL.LineEditREPL, backend::REPL.REPLBackendRef)
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:1651
 [8] (::REPL.var"#61#62"{REPL.LineEditREPL, REPL.REPLBackendRef})()
   @ REPL D:\bamboo\.julia\juliaup\julia-1.12.5+0.x64.w64.mingw32\share\julia\stdlib\v1.12\REPL\src\REPL.jl:650
Some type information was truncated. Use `show(err)` to see complete types.

[BambOoxX]

Might be some edge case of #176

[mcabbott]

That's a weird bug!

The broadcast in this package's show code is here: https://github.com/mcabbott/AxisKeys.jl/blob/master/src/show.jl#L100
and can be made to fail on its own:

julia> using AxisKeys: ShowWith

julia> ShowWith.(1:3; color=:yellow)
3-element Vector{ShowWith{Int64, @NamedTuple{color::Symbol}}}:
 (1)
 (2)
 (3)

julia> ShowWith.(collect(t); color=:yellow)
10-element Vector{ShowWith{Quantity{Float64, Dimensions{FRInt32}}, @NamedTuple{color::Symbol}}}:
 (1.0 s)
 (1.1 s)
 (1.2 s)
 (1.3 s)
 (1.4 s)
 (1.5 s)
 (1.6 s)
 (1.7 s)
 (1.8 s)
 (1.9 s)

julia> ShowWith.(t; color=:yellow)
ERROR: MethodError: no method matching StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}(::UndefInitializer, ::Tuple{Int64})
The type `StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}` exists, but no method is defined for this combination of argument types when trying to construct it.

Closest candidates are:
  StepRangeLen{T, R, S, L}(::R, ::S, ::Integer, ::Integer) where {T, R, S, L}
   @ Base range.jl:502
  StepRangeLen{T, R, S, L}(::R, ::S, ::Integer) where {T, R, S, L}
   @ Base range.jl:502
  StepRangeLen{T, R, S, L}(::StepRangeLen{T, R, S, L}) where {T<:AbstractFloat, R<:Base.TwicePrecision, S<:Base.TwicePrecision, L}
   @ Base twiceprecision.jl:538
  ...

Stacktrace:
 [1] similar(::Type{StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}}, dims::Tuple{Int64})
   @ Base ./abstractarray.jl:868
 [2] similar(::Type{StepRangeLen{…}}, shape::Tuple{Base.OneTo{…}})
   @ Base ./abstractarray.jl:867
 [3] similar(bc::Base.Broadcast.Broadcasted{…}, ::Type{…})
   @ DynamicQuantities ~/.julia/packages/DynamicQuantities/AQfiw/src/arrays.jl:326
 [4] copy
   @ ./broadcast.jl:912 [inlined]
 [5] materialize(bc::Base.Broadcast.Broadcasted{…})
   @ Base.Broadcast ./broadcast.jl:872
 [6] top-level scope
   @ REPL[38]:1
Some type information was truncated. Use `show(err)` to see complete types.

In fact we can trigger the same error without this package at all, e.g.

julia> string.(t)
ERROR: MethodError: no method matching StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}(::UndefInitializer, ::Tuple{Int64})
The type `StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}` exists, but no method is defined for this combination of argument types when trying to construct it.

The DynamicQuantities.jl definition of similar is here:
https://github.com/JuliaPhysics/DynamicQuantities.jl/blob/main/src/arrays.jl#L321-L327

julia> typeof(t)
QuantityArray(::StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}, ::Quantity{Float64, Dimensions{FRInt32}})

julia> V = typeof(t).parameters[end]
StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}

julia> similar(V, axes(t))
ERROR: MethodError: no method matching StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}(::UndefInitializer, ::Tuple{Int64})
The type `StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}` exists, but no method is defined for this combination of argument types when trying to construct it.

Such methods of similar seem to only exist for Array, not for Transpose or UnitRange or... many other types:

julia> let x = rand(3) 
         similar(typeof(x), axes(x))
       end
3-element Vector{Float64}:
 2.180257672e-314
 6.907e-321
 2.8480945388892175e-306

julia> let x = rand(3)' 
         similar(typeof(x), axes(x))
       end
ERROR: MethodError: no method matching Adjoint{Float64, Vector{Float64}}(::UndefInitializer, ::Tuple{Int64, Int64})
The type `Adjoint{Float64, Vector{Float64}}` exists, but no method is defined for this combination of argument types when trying to construct it.

julia> let x = 1:3 
         similar(typeof(x), axes(x))
       end
ERROR: MethodError: Cannot `convert` an object of type UndefInitializer to an object of type Int64
The function `convert` exists, but no method is defined for this combination of argument types.

julia> let x = 1:0.1:3 
         similar(typeof(x), axes(x))
       end
ERROR: MethodError: no method matching StepRangeLen{Float64, Base.TwicePrecision{…}, Base.TwicePrecision{…}, Int64}(::UndefInitializer, ::Tuple{Int64})
The type `StepRangeLen{Float64, Base.TwicePrecision{Float64}, Base.TwicePrecision{Float64}, Int64}` exists, but no method is defined for this combination of argument types when trying to construct it.

So I think that the method of similar which DynamicQuantities.jl defines for broadcasting is probably wrong.

[BambOoxX]

Thanks for the thorough analysis, I opened an issue in DynamicQuantities