Contradiction in plot_topoplot signature

[vladdez]

Why I see this error

julia>     plot_topoplot(dat[:, 50, 1])
ERROR: AssertionError: No positions found, did you forget to provide them via positions=XX, or labels=YY?

if according to signature positions are kwargs plot_topoplot(data; positions = nothing, labels = nothing, kwargs...)?
Some logical error here

[vladdez]

#219

[behinger]

A comment form the off: this behavior mimicks topoplots.jl behavior where positions is a positional argument as well.

If we change it here, we could also change it there. There reason it is not one in topo plots.jl, is that the second positional argument is "labels", not sure we support the labels still?

[vladdez]

1. can I implement that? (actually I already did 😅)
2. what do you mean by support? UM or Topoplots? Labels as i remember do not work in the last one since a long time.

[behinger]

So right now you should be able to do:

plot_topoplots(data;positions=pos)
or
plot_topoplot(data;labels=labels)

Whereas in TopoPlots.jl afaik the interface is crap:
eeg_topoplot(data::Vector{<: Real}, labels::Vector{<: AbstractString})

and you have to provide dummy labels if you don't have them (which I think should change - but communication there is a difficult to say the least).

Now we could say we want to have positions as a positional argument, but I dont yet see the reason why we would want that.

[vladdez]

So there are only two options:

1. we adjust to signature in TopoPlots.jl
2. Topoplots.jl will change their signature

Which is better then?
P.S. of course we can keep it as it is, but it is will be just a source of constant errors.

[behinger]

I thought about it again: I really think labels as keyword and positions as positional is the best way to go.

I will make a PR for TopoPlots.jl