Ben Bolstad experienced something odd while running R CMD check over
a package of ours using r-devel (yesterday's fresh).
He tracked down the problem to a vignette calling 'library(MASS)'.
Please kindly ignore if this is a known (and temporary) issue.
Laurent
----- Forwarded message from Ben Bolstad <bolstad@stat.berkeley.edu> -----
Return-Path: <bolstad@stat.berkeley.edu>
X-Original-To: laurent@cbs.dtu.dk
Delivered-To: laurent@cbs.dtu.dk
Received: from avocet.mail.pas.earthlink.net (avocet.mail.pas.earthlink.net [207.217.120.50])
by genome.cbs.dtu.dk (Postfix) with ESMTP id 826362633D9E
for <laurent@cbs.dtu.dk>; Tue, 8 Apr 2003 21:27:29 +0200 (MDT)
Received: from h-68-164-84-127.snvacaid.covad.net ([68.164.84.127] helo=[192.168.0.3])
by avocet.mail.pas.earthlink.net with esmtp (Exim 3.33 #1)
id 192yfb-0003Hu-00; Tue, 08 Apr 2003 12:22:11 -0700
Subject: build problem with affy
From: Ben Bolstad <bolstad@stat.berkeley.edu>
To: rafa@jhu.edu
Cc: Laurent Gautier <laurent@cbs.dtu.dk>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-11)
Date: 08 Apr 2003 12:22:10 -0700
Message-Id: <1049829731.3929.50.camel@bmbbox.dyndns.org>
Mime-Version: 1.0
Status: O
Content-Length: 1446
Lines: 34
I am getting this error when doing a R CMD build --force on affy
I am using an updated R-devel rsync'ed this morning and I am using the
latest MASS package (which was my first thought). I am not having
problems building other packages.
Any thoughts?
* creating vignettes ... ERROR
Error in namespaceExport(ns, exports) : undefined exports: abbey,
accdeaths, Aids2, Animals, anorexia, austres, bacteria, beav1, beav2,
biopsy, birthwt, Boston, cabbages, caith, Cars93, cats, cement, chem,
coop, cpus, crabs, Cushings, DDT, deaths, drivers, eagles, epil, farms,
fdeaths, fgl,
forbes, GAGurine, galaxies, gehan, genotype, geyser, gilgais, hills,
housing, immer, Insurance, leuk, lh, mammals, MASS.data.load, mcycle,
mdeaths, Melanoma, menarche, michelson, minn38, motors, muscle, newcomb,
nlschools, nottem, npk, npr1, oats, OME, painters, petrol, phones,
Pima.te, Pima.tr, Pima.tr2, quine, Rabbit, road, rock, rotifer, Rubber,
ships, shoes, shrimp, shuttle, Sitka, Sitka89, Skye, snails, SP500,
steam, stormer, survey, synth.te, synth.tr, topo, Traffic, UScereal,
UScrime, VA, waders, whiteside, wtloss
Error in library(MASS) : package/namespace load failed
Error in driver$runcode(drobj, chunk, chunkopts) :
Error in library(MASS) : package/namespace load failed
Error in buildVignettes(dir = ".") : Error in driver$runcode(drobj,
chunk, chunkopts) :
Error in library(MASS) : package/namespace load failed
Execution halted
----- End forwarded message -----
--------------------------------------------------------------
currently at the National Yang-Ming University in Taipei, Taiwan
--------------------------------------------------------------
Laurent Gautier CBS, Building 208, DTU
PhD. Student DK-2800 Lyngby,Denmark
tel: +45 45 25 24 89 http://www.cbs.dtu.dk/laurent
Ben Bolstad experienced something odd while running R CMD check over
a package of ours using r-devel (yesterday's fresh).
He tracked down the problem to a vignette calling 'library(MASS)'.
Please kindly ignore if this is a known (and temporary) issue.
It is not a known issue, but we have no idea how to reproduce this from
your message. My guess is that your build of MASS is faulty, as you seem
to be missing all the data files. Does library(MASS) on its own work?
Did you do `make check-all' on your build?
Why did you clog the list with this three times, including full email
headers which are of no interest to us?
[...]
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
I CCed Ben Bolstad, whom I mistakingly forget in my previous post.
Thanks,
L.
PS: The mail abuse was perpretrated by putting 'r-devel@stat.math.ethz.ch'
in the field 'To:' of my mailing software. Lack of sleep could have make me
write, inadvertantly, twice the same recipient. Unbearable
mail headers should be absent from this post.
On Wed, Apr 09, 2003 at 08:18:50AM +0100, Prof Brian Ripley wrote:
On Wed, 9 Apr 2003, Laurent Gautier wrote:
Ben Bolstad experienced something odd while running R CMD check over
a package of ours using r-devel (yesterday's fresh).
He tracked down the problem to a vignette calling 'library(MASS)'.
Please kindly ignore if this is a known (and temporary) issue.
It is not a known issue, but we have no idea how to reproduce this from
your message. My guess is that your build of MASS is faulty, as you seem
to be missing all the data files. Does library(MASS) on its own work?
Did you do `make check-all' on your build?
It is not a known issue, but we have no idea how to reproduce this from
your message. My guess is that your build of MASS is faulty, as you seem
to be missing all the data files. Does library(MASS) on its own work?
Did you do `make check-all' on your build?
The problem is with library(MASS).
library(MASS)
Error in namespaceExport(ns, exports) : undefined exports: abbey,
accdeaths, Aids2, Animals, anorexia, austres, bacteria, beav1, beav2,
biopsy, birthwt, Boston, cabbages, caith, Cars93, cats, cement, chem,
coop, cpus, crabs, Cushings, DDT,
deaths, drivers, eagles, epil, farms, fdeaths, fgl, forbes, GAGurine,
galaxies,
gehan, genotype, geyser, gilgais, hills, housing, immer, Insurance,
leuk, lh, mammals, MASS.data.load, mcycle, mdeaths, Melanoma, menarche,
michelson, minn38, motors, muscle, newcomb, nlschools, nottem, npk,
npr1, oats, OME, painters, petrol, phones, Pima.te, Pima.tr, Pima.tr2,
quine, Rabbit, road, rock, rotifer, Rubber, ships, shoes, shrimp,
shuttle, Sitka, Sitka89, Skye, snails, SP500, steam, stormer, survey,
synth.te, synth.tr, topo, Traffic, UScereal, UScrime, VA, waders,
whiteside, wtloss
Error in library(MASS) : package/namespace load failed
I have checked the MASS installation and all the data files are there
and i can manually load() them just fine.
make check-all on my current rsync'ed R-devel chokes at.
make[3]: Entering directory `/R/tests/Examples'
collecting examples for package 'survival' ...
file ../../library/survival/R-ex cannot be opened at
../../share/perl/massage-Examples.pl line 107.
survival-Ex.R is unchanged
running code in 'survival-Ex.R' ...make[3]: *** [survival-Ex.Rout] Error
1
make[3]: Leaving directory `/R/tests/Examples'
make[2]: *** [test-Examples-Recommended] Error 2
make[2]: Leaving directory `/R/tests/Examples'
make[1]: *** [test-Examples-Recommended] Error 2
make[1]: Leaving directory `/R/tests'
make: *** [check-all] Error 2
Thanks,
Ben
It is not a known issue, but we have no idea how to reproduce this from
your message. My guess is that your build of MASS is faulty, as you seem
to be missing all the data files. Does library(MASS) on its own work?
Did you do `make check-all' on your build?
The problem is with library(MASS).
No, the problem is with your installation: everyone else is getting all
these checks to pass. You did not check the installation fully ....
I have checked the MASS installation and all the data files are there
and i can manually load() them just fine.
But that is not how they work: they are promises to load and you are
missing the promises. How can you `have checked' something you don't
understand?
make check-all on my current rsync'ed R-devel chokes at.
Ahah: do you have correct versions of the recommended packages? That is,
run tools/rsync-recommended?
make[3]: Entering directory `/R/tests/Examples'
collecting examples for package 'survival' ...
file ../../library/survival/R-ex cannot be opened at
../../share/perl/massage-Examples.pl line 107.
survival-Ex.R is unchanged
That's another error in your installation not seen elsewhere.
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595