[ESS-bugs] debug(unique.default) -- and "hell breaks loose" ;-)

[Martin Maechler]

The subject is jokingly,
but watch :

M-x R   freshly, then in the *R* buffer :
Note the browse level [14] (!!)  and then the huge call stack :

---------------------------------------------------------------------
> debug(unique.default)
> unique(diag(4))
     [,1] [,2] [,3] [,4]
[1,]    1    0    0    0
[2,]    0    1    0    0
[3,]    0    0    1    0
[4,]    0    0    0    1
Browse[14]> 
debug: if (is.factor(x)) {
    z <- .Internal(unique(x, incomparables, fromLast, min(length(x), 
        nlevels(x) + 1L)))
    return(factor(z, levels = seq_len(nlevels(x)), labels = levels(x), 
        ordered = is.ordered(x)))
}
Browse[14]> sys.calls()
[[1]]
.ess_funargs("unique")

[[2]]
Ops.numeric_version(.ess.Rversion, "2.14.1")

[[3]]
as.numeric_version(e2)

[[4]]
numeric_version(x)

[[5]]
.make_numeric_version(x, strict, .standard_regexps()$valid_numeric_version)

[[6]]
unique(c(classes, "numeric_version"))

[[7]]
unique.default(c(classes, "numeric_version"))

[[8]]
.ess_funargs("unique")

  [.................................]
  [.................................]


[[49]]
unique.default(c(classes, "numeric_version"))

Browse[14]> 

----------

Yes, it is because our R code ESSR also calls unique() ->
unique.default() ...

But it is very unfortunate, of course, if I really want to debug
to which unique() method R dispatches, and that is what I had
wanted to do.

Martin