Skip to content

Error loading ctest

3 messages · partha_bagchi@hgsi.com, Thomas Lumley, Brian Ripley

partha_bagchi@hgsi.com
# 
Has anyone noticed this behavior:
Error in dyn.load(x, as.logical(local), as.logical(now)) :
        unable to load shared library "J:
\STATS\RW1000/library/ctest/libs/ctest.dll":
  LoadLibrary failure
_
platform Windows
arch     x86
os       Win32
system   x86, Win32
status
major    1
minor    0.0
year     2000
month    February
day      29
language R
The first time ctest fails to load and loads on the second try??
Thanks for any insight.


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
Thomas Lumley
#
On Wed, 8 Mar 2000 partha_bagchi at hgsi.com wrote:

            

      
      
      
    

        
It works here (WinNT)

	-thomas

            

      
      
      
    

        
Thomas Lumley
Assistant Professor, Biostatistics
University of Washington, Seattle

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !)  To: r-help-request at stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

  
  


  


      

    

    

      
      

        


  

        

      Brian Ripley
    

    

      
      #
    

  


  

      
On Wed, 8 Mar 2000 partha_bagchi at hgsi.com wrote:

        

            

      
      
      
    

        
Are you sure it loads the second time?  If you load a package and it fails
in .First.lib, then the environment is already there, and when you go to
load it again it does not reload it. (Try this with a library that has a
start-up banner like KernSmooth or locfit.)  It just finds the name already
there and quits. My expectation would be that the compiled code is
unavailable after what you describe.

Perhaps we ought to alter this to make a failure in .First.lib unload the
package.

We've corresponded on this privately:

            

      
      
      
    

        
is suspicious. Is this what you actually get or is your mailer (or
something else en route) breaking that line?

As Thomas says, this works for lots of other installations. Can you
try it on a local disc (if J: is your Novell one)?