Question: Use of as.name

[mboyas-mitre]

openVA/openVA/R/VAmain.r

Lines 168 to 173 in d6e2af7

	args$data <- as.name("data")
	args$train <- as.name("data.train")
	args$cause <- as.name("causes.train")
	args$type <- convert.type
	fit <- do.call("insilico.train", pairlist(args)[[1]][-1])

What is the logic behind using as.name and then a do.call here? Why not just call insilico.train directly on the appropriate arguments? At first glance, this seems like a complicated way of achieving the end goal, but we wanted to double check about what is happening in these lines of code before experimenting with alternatives.

[richardli]

It was supposed to pass the additional arguments, i.e., the "..." in the input argument, to insilico.train, without specifying everything in the document here. And it also serves to map the different argument names. Any suggestions to make it easier?

I know the best option is perhaps to rename everything across packages so that they are consistent. Will see if at some point someone can do that...