Skip to content

can you library(MASS) with R 1.9.0?

10 messages · A.J. Rossini, Tony Plate, Robert King +2 more

#
I'm trying to track down a rather critical problem.  This happens both
on Debian's current unstable as well a from anoncvs built this
morning, reinstalling MASS.  Is anyone else seeing this or am I just
broken and special?

609$ /home/Rdevel/bin/R

R : Copyright 2004, The R Foundation for Statistical Computing
Version 1.9.0 Under development (unstable) (2004-03-04), ISBN
3-900051-00-3

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type 'license()' or 'licence()' for distribution details.

R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R in publications.

Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for a HTML browser interface to help.
Type 'q()' to quit R.
Error in loadNamespace(i[[1]], c(lib.loc, .libPaths()), keep.source) :
        package 'mva' does not have a name space
Error in library(MASS) : package/namespace load failed
I want to know if it's just me.

best,
-tony
#
I had a similar problem with a build made from a fresh rsync of r-devel 
made on Mar 1.  When I ran "make test" it reported (after running many 
other tests) that it could not find the libraries MASS or survival.  I did 
not investigate further.

-- Tony Plate
At Thursday 10:32 AM 3/4/2004, A.J. Rossini wrote:

            
#
You need the version in src/contrib/1.9.0/Recommended on cran (and you do 
for several other recommended packages).

tools/rsync-recommended should get you that version: is that how you are 
doing it?

As a check, MASS comes from VR version 7.2-0 in R-devel.
On Thu, 4 Mar 2004, A.J. Rossini wrote:

            

  
    
#
Thank you!

I had not used tools/rsync-recommended -- I just coped the rsync command 
from the R FAQ (i.e., rsync -rC --delete rsync.R-project.org::r-devel R).

The tools/rsync-recommended command downloaded the various recommended 
packages.

However, to get this I had to get the latest version of the tools (out of 
<http://www.stats.ox.ac.uk/pub/Rtools/tools.zip>http://www.stats.ox.ac.uk/pub/Rtools/tools.zip), 
and after changing my path to use it I get the dreaded "You have multiple 
copies of cygwin1.dll on your system" error, even though I have set my path 
so that I have only one copy of cygwin1.dll in my path (as far as I can see):

$ type -a cygwin1.dll
cygwin1.dll is /cygdrive/c/R/tools/cygwin1.dll
$ ls
c:\R\tools\ls.exe: *** fhandler size version mismatch detected - 0x70/0x6C.
You have multiple copies of cygwin1.dll on your system.
Search for cygwin1.dll using the Windows Start->Find/Search facility
and delete all but the most recent version.  The most recent version *should*
reside in x:\cygwin\bin, where 'x' is the drive on which you have
installed the cygwin distribution.

# this is the path I'm had for the commands above -- this command run with 
a different PATH so that it can run
$ echo $RPATH | tr ':' '\012'
.
/cygdrive/c/R/tools
/cygdrive/c/MinGW/bin
//jacona/home/tplate/depot/Spluslibs/scripts
//jacona/home/tplate/scripts
//jacona/home/data/bin
/cygdrive/c/texmf/miktex/bin
/cygdrive/c/Perl/bin
/cygdrive/c/jdk1.3/bin
/cygdrive/c/tap/scripts
/cygdrive/c/gnu/gnuplot
/cygdrive/c/WINNT/system32
/cygdrive/c/WINNT
/cygdrive/c/Program Files/Insightful/splus61/cmd
/cygdrive/c/Program Files/Insightful/splus61net/server/cmd
/cygdrive/c/Program Files/QAI/MarketQA/Bin
/cygdrive/e/marketqa/Bin
/cygdrive/c/marketqa/Bin
/cygdrive/c/PVSW/Bin
/cygdrive/c/Program Files/Microsoft Visual Studio/VC98/Bin
/cygdrive/c/Program Files/Microsoft Visual Studio/Common/MSDev98/Bin
/cygdrive/c/WINNT/system32
/cygdrive/c/WINNT
/cygdrive/c/WINNT/system32/WBEM
/cygdrive/c/Program Files/Perforce
/cygdrive/c/Program Files/Html Help Workshop
/cygdrive/c/gstools/gs/gs7.03/bin
/cygdrive/c/R/rw1080/bin
.

(I'm running under Windows 2000).

Any suggestions on how to get things working again are much appreciated.

-- Tony Plate
At Thursday 02:33 PM 3/4/2004, Prof Brian Ripley wrote:

  
  
#
It is just possible you need to log out and log in again, or even reboot, 
since an earlier version of the DLL might be loaded into the OS.
On Thu, 4 Mar 2004, Tony Plate wrote:

            

  
    
4 days later
#
(I'm guessing the same problem) is also present in 7.1.14-2 of the debian
package r-cran-vr with r-base-core 1.8.1.cvs.20040229-2
it can be fixed with

/usr/lib/R/library/MASS:

tolstoy:/usr/lib/R/library/MASS# diff -u NAMESPACE~ NAMESPACE
--- NAMESPACE~  Wed Mar 10 12:42:26 2004
+++ NAMESPACE   Wed Mar 10 12:42:38 2004
@@ -53,7 +53,7 @@
 #        qda.data.frame, qda.default, qda.formula, qda.matrix,
 #        rlm.default, rlm.formula)

-importFrom(mva, biplot)
+importFrom(stats, biplot)

 S3method("[", fractions)
 S3method("[<-", fractions)
On Thu, 4 Mar 2004, A.J. Rossini wrote:
----
Robert King, Statistics, School of Mathematical & Physical Sciences,
University of Newcastle, Australia
Room V133  ph +61 2 4921 5548
Robert.King@newcastle.edu.au   http://maths.newcastle.edu.au/~rking/

We are all agreed that your theory is crazy.  The question which divides us is
whether it is crazy enough to have a chance of being correct.  My own feeling
is that it is not crazy enough.
		-- Niels Bohr
#
On Wed, Mar 10, 2004 at 12:55:40PM +1100, Robert King wrote:
Yes, I left a small window during which pre-1.9.0 releases of R itself where
matched with 1.8.0 releases of CPAN packages. But several packages from the
$CRAN/src/contrib/1.9.0 branch have been added, and r-cran_7.20-1.

Hope this helps,  Dirk
#
On Tue, Mar 09, 2004 at 08:02:10PM -0600, Dirk Eddelbuettel wrote:
Oops, make that   "... and r-cran-vr_7.20-1 is one of them."

Dirk

  
    
#
A package of 1.8.1 should not have a package stats.  If this is 1.9.0, you 
should be using VR_7.2-0 or later.

Someone is confused about version numbers.
On Wed, 10 Mar 2004, Robert King wrote:

            

  
    
#
I believe Dirk has (or will have very soon) fixed this version
problem.  

best,
-tony


Prof Brian Ripley <ripley@stats.ox.ac.uk> writes: