[Rd] Problem with Rcmd check and S4 methods

[John Chambers]

Duncan Murdoch wrote:
> 
> 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?

It's a little tricky, and in any case the better solution is to ensure
that the semantics are the same with or without the saved image.

If it turns out that the time penalty for library(foo) with no saved
image is very large when foo contains class/method code, we could give a
warning at INSTALL time if a saved image is not created.  Or perhaps
make saved images the default?

John

> 
> >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

-- 
John M. Chambers                  jmc at bell-labs.com
Bell Labs, Lucent Technologies    office: (908)582-2681
700 Mountain Avenue, Room 2C-282  fax:    (908)582-3340
Murray Hill, NJ  07974            web: http://www.cs.bell-labs.com/~jmc