Results of Resubmission 2: An error in cmcre!

[swihart]

So I have passed checks on the 3 platforms. Cran's check failed.

* checking examples ... ERROR
Running examples in ‘repeated-Ex.R’ failed
The error most likely occurred in:

> base::assign(".ptime", proc.time(), pos = "CheckExEnv")
> ### Name: cmcre
> ### Title: Continuous-time Two-state Markov Processes with Random Effect
> ### Aliases: cmcre
> ### Keywords: models
>
> ### ** Examples
>
>
> # 12 subjects observed at intervals of 7 days
> y <- matrix(c(1,7,1,2,3,5,
+       2,7,10,2,2,0,
+       3,7,7,0,1,1,
+       4,7,2,1,0,7,
+       5,7,1,1,1,11,
+       6,7,5,4,4,1,
+       7,7,1,1,1,8,
+       8,7,2,3,4,2,
+       9,7,9,0,0,0,
+       10,7,0,1,2,8,
+       11,7,8,2,2,1,
+       12,7,9,2,2,1),ncol=6, byrow=TRUE)
> # ordinary Markov process
> cmcre(y, par=c(-0.2,-1))
Error in cmcre(y, par = c(-0.2, -1)) :
  array over-run in .C("LoadData") in integer argument 4
Calls: cmcre -> .C
Execution halted

Look at lines 207-214 in cmcre.r

x <- .C("LoadData",
	response=as.double(t(cbind(response[,1],1,response[,2:6],covariate))),
	nrow=as.integer(dim(response)[1]),
	nSize=as.integer(dim(response)[2]+1+cov),
	NumSubjects=integer(1),
	Error=integer(1),
	PACKAGE="repeated")
if(x$Error>0)stop("error in storing data")

I think we are dealing with the line nSize=as.integer(dim(response)[2]+1+cov), and one thing I see is that cov is logical...is that causing a problem?

No, no. The error says array and the nSize is an integer scalar. My money is on
plNumSubjects being argument 4 in lines 34-37 of gaps.c:

LoadData( double *response, int *nrow, int *nSize,
          long * plNumSubjects, 
          long *plReturn )
{

In cmcre.r, plNumSubjects corresponds to NumSubjects=integer(1),. interger(1) is a way of making an array that is length 1. Somewhere in the c it might be made longer... and this stuff sometimes throws an error with compiling nad sometimes not (is my impression from reading stuff).

[swihart]

kurt said he was using:

Debian testing with the GCC 6 compilers.

I vote to remove cmcre for now. LoadData is only used in it. See if we can repeated on cran. I will add it to .Rbuildignore.

> devtools::use_build_ignore("./R/cmcre.r")
> devtools::use_build_ignore("./man/cmcre.Rd")

... and .Rbuildignore didn't work. So I exported them out of the /R folder and out of /man folder into package directory and then edited .Rbuildignore and now they are ignored. undo these when you want to bring cmcre back into the fold