[R] Error generated by nlme::gnls

[Rolf Turner]

On Sat, 23 Jul 2022 21:00:25 -0400
Ben Tupper <btupper using bigelow.org> wrote:

> Could this be related to a new if() behavior introduced in v4.2.0 ?
> See the "SIGNIFICANT USER-VISIBLE CHANGES" for v4.2.0 in the NEWS
> 
> https://cloud.r-project.org/doc/manuals/r-release/NEWS.html

No.  What's going on is much weirder than that, and looks to me like
a bug has been introduced in formula() or in something related.

My impression is that if the right hand side of a formula gets "too
long", then it gets split into parts --- which messes everything up.

Consider:

lhs1 <- paste(paste0("x",1:12),collapse=" + ")
f1   <- as.formula(paste("x ~ -1 +",lhs1))
print(length(deparse(f1[[3]])))

You get the value 2.  Note that deparse(f1[[3]]) gives

[1] "-1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11 + "
[2] "    x12"       

If we shorten the right hand side of the formula just a wee bit

lhs2 <- paste(paste0("x",1:11),collapse=" + ")
f2   <- as.formula(paste("x ~ -1 +",lhs2))

then deparse(f2[[3]]) is of length 1, as it should be:

> > deparse(f2[[3]])
> [1] "-1 + x1 + x2 + x3 + x4 + x5 + x6 + x7 + x8 + x9 + x10 + x11"

I could of course be missing something, but it really looks to me as if
something has gone up to Puttee here.

Some input from someone in R-Core would be valuable here.

cheers,

Rolf Turner

-- 
Honorary Research Fellow
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276