-----Original Message-----
From: Prof Brian Ripley [mailto:ripley@stats.ox.ac.uk]
Sent: Friday, December 28, 2001 2:30 AM
To: Peter Dalgaard BSA
Cc: Wright, Kevin; R-bugs@biostat.ku.dk
Subject: Re: [Rd] reshape error in 1.4 (PR#1231)
On 27 Dec 2001, Peter Dalgaard BSA wrote:
Kevin.Wright@pioneer.com writes:
Full_Name: Kevin Wright
Version: 1.4
OS: Windows 95
Submission from: (NULL) (170.54.59.160)
Note: This was the 1.4 build for Windows that Brian
The first example in the help for reshape doesn't work
paste, this is what happens:
data(Indometh,package="nls")
summary(Indometh)
Subject time conc
1:11 Min. :0.250 Min. :0.0500
4:11 1st Qu.:0.750 1st Qu.:0.1100
2:11 Median :2.000 Median :0.3400
5:11 Mean :2.886 Mean :0.5918
6:11 3rd Qu.:5.000 3rd Qu.:0.8325
3:11 Max. :8.000 Max. :2.7200
wide<-reshape(Indometh,v.names="conc",idvar="Subject",
+ timevar="time",direction="wide")
Error in tapply(a, data[, idvar], function(b)
length(unique(b)) == 1) :
arguments must have same length
This doesn't appear to be reproducible on non-Windows machines, and
I'd suspect that Brian checks quite carefully for this kind of thing
on Windows as well. You wouldn't happen to have something floating
around in your workspace that could get in the way? Otherwise try
turning on debug(reshape) and when it is defined also
debug(reshapeWide). Then tell us what is in a and data[, idvar] when
the problem appears.
Yes, example(reshape) works on that build on the machine on
which it was
built. Running all the examples is part of make check, so if
they fail
for you it is very likely to be a local problem.
--
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 272860 (secr)
Oxford OX1 3TG, UK Fax: +44 1865 272595