R help - Mar 2009 - cmprsk- another survival-depedent package causes R crash

Dear Prof Gray and everyone,

As our package developers discussed about incompatibility between Design and
survival packages,  I faced another problem with cmprsk- a survival dependent
packacge.
The problem is exactly similar to what happened to the Design package that when
I just started running cuminc function, R was suddenly closed.
These incidents suggest that maybe many other survival dependent packages being
involved the problem
Could you please consider the matter

My R version: 2.8.1
Window XP Service pack 3

Regards
Nguyen D Nguyen
Garvan Institute, Australia

	[[alternative HTML version deleted]]

Nguyen Dinh Nguyen wrote:
> Dear Prof Gray and everyone,
> 
> As our package developers discussed about incompatibility between Design
and survival packages,  I faced another problem with cmprsk- a survival
dependent packacge.
> The problem is exactly similar to what happened to the Design package that
when I just started running cuminc function, R was suddenly closed.
> These incidents suggest that maybe many other survival dependent packages
being involved the problem
> Could you please consider the matter
> 
> My R version: 2.8.1
> Window XP Service pack 3
> 
> Regards
> Nguyen D Nguyen
> Garvan Institute, Australia
> 
Nguyen,

As much as this costs some effort from us, it is the risk we take when 
calling another package's code at other than the highest level.  Other 
package developers are especially free to change calls to C and Fortran 
code, and we hook into some of survival's C code which has changed 
(slightly).

Frank
-- 
Frank E Harrell Jr   Professor and Chair           School of Medicine
                      Department of Biostatistics   Vanderbilt University

Yes, there are other packages with incompatibilities with the new version of
'survival'.  The package maintainers for all the packages that fail R
CMD check have all been notified and given suggestions for how to update. You
can see which packages fail CMD check by looking at the CRAN check results
linked from the package's CRAN page.

If you find that the change has caused problems for a package but hasn't
caused a CMD check failure, then it would be helpful to report it, since we
might not know.  Otherwise I think you can assume that an update is already on
someone's to-do list.

     -thomas


On Mon, 30 Mar 2009, Nguyen Dinh Nguyen wrote:
> Dear Prof Gray and everyone,
>
> As our package developers discussed about incompatibility between Design
and survival packages,  I faced another problem with cmprsk- a survival
dependent packacge.
> The problem is exactly similar to what happened to the Design package that
when I just started running cuminc function, R was suddenly closed.
> These incidents suggest that maybe many other survival dependent packages
being involved the problem
> Could you please consider the matter
>
> My R version: 2.8.1
> Window XP Service pack 3
>
> Regards
> Nguyen D Nguyen
> Garvan Institute, Australia
>
> 	[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
Thomas Lumley			Assoc. Professor, Biostatistics
tlumley at u.washington.edu	University of Washington, Seattle

> As our package developers discussed about incompatibility between Design
and
survival packages,  I faced another problem with cmprsk- a survival dependent 
packacge.
> The problem is exactly similar to what happened to the Design package that 
when I just started running cuminc function, R was suddenly
closed.
> These incidents suggest that maybe many other survival dependent packages 
being involved the problem

  I don't see how this is related to survival.  I just checked the source
code
to the cmprsk function, and it has no dependencies on my library.  As I would 
expect, the cmprks function works as expected on our machines.
  Could you send a reproducable example?
  
  	Terry Therneau