Problem with Rcmd check and S4 methods

In the little orientation data package I'm working on, the very first
example fails, even though the same code works when used in a regular
session.  In my experience, this usually indicates T or F has been
used instead of TRUE or FALSE, but that's not the case this time.

The error message is

 > x <- eulerzyx(1,0,0)4bb

 Error: Trying to get slot "validity" from an object of a basic class
("NULL") with no slots
 Execution halted

What I think is the relevant source looks like this:

setClass('orientation')
setIs('orientation','vector')

setClass('eulerzyx', representation(x = 'matrix'))
setIs('eulerzyx', 'orientation')

eulerzyx <- function(psi, theta, phi)
    new('eulerzyx', x = cbind(psi, theta, phi))

Can anyone tell me what that error message means? 

Is it something
I've done wrong, or a problem with the methods package or the checks?

Duncan Murdoch

Duncan Murdoch wrote:

 > x <- eulerzyx(1,0,0)4bb

 Error: Trying to get slot "validity" from an object of a basic class
("NULL") with no slots
 Execution halted

The code has a NULL object that was supposed to be a class definition
(and therefore to have a "validity" slot).

I was able to add your source & the one-line example to a small package
and get through R CMD check.

Can you do example(eulerzyx) after attaching the package?

Are you using a saved image (e.g., with an empty install.R file in the
package directory)?  (Though if not, the check would often fail at an
earlier stage.)

It would be nice to be able to get a dump from the R CMD check
evaluation--maybe someone can suggest how.

 x <- eulerzyx(1,0,0)

traceback()

On Fri, 13 Jun 2003 13:18:23 -0400, you wrote:

It would be nice to be able to get a dump from the R CMD check
evaluation--maybe someone can suggest how.

If I just cut all the preamble from the Rout file and paste it into
the console, I get the same error.  Here's a traceback(), I'm not sure
what you mean by a dump.

Duncan

 x <- eulerzyx(1,0,0)

Error: Trying to get slot "validity" from an object of a basic class
("NULL") with no slots

traceback()

5: validObject(.Object)
4: initialize(value, ...)
3: initialize(value, ...)
2: new("eulerzyx", x = cbind(psi, theta, phi))
1: eulerzyx(1, 0, 0)

x <- rotvector(matrix(c(1,0,0,0,1,0,0,0,1)))

I think the problem is from not using a saved image.  Using your tar
file, if I untar into orientlib/, I get the same error you report.  But
after creating the install.R file:

guinan[RPackages]$ touch orientlib/install.R

the error in this example goes away.  (There is a later error:

x <- rotvector(matrix(c(1,0,0,0,1,0,0,0,1)))

Error: ncol(m) == 9 is not TRUE

but I'm guessing this may be more local :-)

On Fri, 13 Jun 2003 15:04:44 -0400, you wrote:

I think the problem is from not using a saved image.  Using your tar
file, if I untar into orientlib/, I get the same error you report.  But
after creating the install.R file:

Thanks, that fixed it.  Perhaps the check script could try to
determine if this was necessary?

guinan[RPackages]$ touch orientlib/install.R

the error in this example goes away.  (There is a later error:

x <- rotvector(matrix(c(1,0,0,0,1,0,0,0,1)))

Error: ncol(m) == 9 is not TRUE

but I'm guessing this may be more local :-)

Yes :-).

Duncan