Implement get_facet_fml() via tinyformula()/tinyframe()

[zeileis]

The tinyplot.formula() method uses auxiliary functions tinyformula()/tinyframe() (see https://github.com/grantmcdermott/tinyplot/blob/main/R/tinyformula.R) for processing its formulas, including facets.

In sanitize_facet() (see https://github.com/grantmcdermott/tinyplot/blob/main/R/sanitize_facet.R) an auxiliary function get_facet_fml() (see https://github.com/grantmcdermott/tinyplot/blob/main/R/facet.R#L680-L726) is used for cases when facet is still a formula. However, get_facet_fml() is not implemented using tinyformula()/tinyframe(). It might be a good idea to harmonize/streamline this implementation. This first came up in #562

As a first step, I tried to come up with an example when get_facet_fml() is actually used successfully. Could you provide an example, Grant @grantmcdermott ?

My reading is that this can only happen if tinyplot.default() is used because tinyplot.formula() already parses facet formulas and does not pass these on. But in tinyplot.default() I always get an error if I do tinyplot(x, y, facet = a ~ b) or tinyplot(x, y, facet = a ~ b, data = d).

[grantmcdermott]

I'm not at my computer right now, but I think it might have been in the context of with(data, tinyplot(..., facet = a ~ b)), i.e. grid-arranged facets for "atomic" plots.

[zeileis]

Thanks, this does seem to work. I thought I had tried an explicit attach() before as well but apparently I overlooked something... So this can serve as an example:

with(penguins, tinyplot(bill_dep ~ bill_len, facet = island ~ species))

[zeileis]

Ah, I overlooked that I must only use the default interface (which then calles get_facet_fml) and not the formula interface as above (which resolves the facet formula already). These both fail:

with(penguins, tinyplot(bill_len, bill_dep, facet = island ~ species))
## Error in model.frame.default(formula = island ~ species, data = data) : 
##   'data' must be a data.frame, environment, or list

tinyplot(penguins$bill_len, penguins$bill_dep, facet = island ~ species, data = penguins)
## Error in model.frame.default(formula = island ~ species, data = data) : 
##   'data' must be a data.frame, environment, or list