Re: [Rd] zero variance in part of a glm (PR#11355)
[EMAIL PROTECTED] wrote:
In this real example (below), all four of the replicates in one
treatment combination had zero failures, and this produced a very high
standard error in the summary.lm.
Why are you reporting trouble with your own understanding as a bug in R?
I don't see anything out of the ordinary here. In the first case you
have divergence, not because of a zero variance, but because you are
trying to estimate an infinite value for the linear predictor (l =
logit(1) for the combination where everyone dies). This will also happen
if you fit an ordinary logistic regression, whenever you have 0s in the
y*antibio*toxin marginal table.
=20
Just adding one failure to one of the replicates produced a well-behaved
standard error.
=20
I don't know if this is a bug, but it is certainly hard for users to
understand.
=20
I would value your comments=20
=20
Thanks
=20
Best wishes,
=20
Mick
=20
Prof M.J. Crawley
=20
Imperial College London
Silwood Park
Ascot
Berks
SL5 7PY
UK
=20
Phone (0) 207 5942 216
Fax (0) 207 5942 339
=20
The data are from a bioassay in which a factorial experiment with two
factors (antibio and toxin) each with three levels was replicated four
times. The response is "mi" and "n-mi"
=20
Note that lines 17 to 20 in the dataframe have no failures (24 dead out
of 24 individuals)
data<-read.table("c:\\temp\\ab1.txt",header=3DT)
attach(data)
names(data)
=20
[1] "antibio" "toxin" "rep" "alive" "af" "dead""mi"
[8] "n"=20=20=20=20
=20
data
=20
=20
=20
antibio toxin rep alive af dead mi n
1 camp control 124 00 0 24
2 camp control 223 01 1 24
3 camp control 323 01 1 24
4 camp control 421 03 3 24
5 camp Cry1Ab 121 43 7 24
6 camp Cry1Ab 220 34 7 24
7 camp Cry1Ab 320 44 8 24
8 camp Cry1Ab 418 76 13 24
9 camp Vip3A 1 7 3 17 20 24
10camp Vip3A 210 6 14 20 24
11camp Vip3A 311 5 13 18 24
12camp Vip3A 410 2 14 16 24
13 bcock control 121 03 3 24
14 bcock control 224 00 0 24
15 bcock control 324 30 3 24
16 bcock control 423 11 2 24
17 bcock Cry1Ab 1 4 4 20 24 24
18 bcock Cry1Ab 2 4 4 20 24 24
19 bcock Cry1Ab 3 1 1 23 24 24
20 bcock Cry1Ab 4 2 2 22 24 24
21 bcock Vip3A 111 4 13 17 24
22 bcock Vip3A 215 59 14 24
23 bcock Vip3A 3 5 3 19 22 24
24 bcock Vip3A 410 6 14 20 24
25 ana control 123 01 1 24
26 ana control 223 01 1 24
27 ana control 322 02 2 24
28 ana control 423 01 1 24
29 ana Cry1Ab 118 16 7 24
30 ana Cry1Ab 217 47 11 24
31 ana Cry1Ab 313 4 11 15 24
32 ana Cry1Ab 414 3 10 13 24
33 ana Vip3A 121 123 15 24
34 ana Vip3A 212 6 12 18 24
35 ana Vip3A 3 9 5 15 20 24
36 ana Vip3A 4 9 1 15 16 24
=20
y<-cbind(mi,n-mi)
model<-glm(y~antibio*toxin,quasibinomial)
summary(model)
=20
Call:
glm(formula =3D y ~ antibio * toxin, family =3D quasibinomial)
=20
Deviance Residuals:=20
Min 1Q Median 3Q Max=20=20
-2.0437 -0.5645 -0.1022 0.6921 1.9996=20=20
=20
Coefficients:
Estimate Std. Error t value Pr(>|t|)=20=20=20=
=20
(Intercept) -2.901e+00 5.020e-01 -5.780 3.79e-06 ***
antibiobcock 5.035e-01 6.441e-010.7820.441=20=20=20=
=20
antibiocamp 2.100e-15 7.099e-01 2.96e-151.000=20=20=20=
=20
toxinCry1Ab 2.818e+00 5.494e-015.129 2.15e-05 ***
toxinVip3A3.840e+00 5.600e-016.857 2.29e-07 ***
antibiobcock:toxinCry1Ab 2.050e+01 2.365e+030.0090.993=20=20=20=
=20
antibiocamp:toxinCry1Ab -4.721e-01 7.795e-01 -0.6060.550=20=20=20=
=20
antibiobcock:toxinVip3A -2.868e-01 7.381e-01 -0.3890.701=20=20=20=
=20
antibiocamp:toxinVip3A2.748e-01 7.975e-010.3450.733=20=20=20=
=20
---
Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1=20
=20
(Dispersion parameter for quasibinomial family taken to be 1.19442)
=20
Null deviance: 515.115 on 35 degrees of freedom
Residual deviance: 35.047 on 27 degrees of freedom
AIC: NA
=20
Number of Fisher Scoring iterations: 17
=20
# note the standard error of the firast interaction term =3D 2365)
=20
# add a single failure to one replicate
=20
y2<-y
y2[17,]<-c(23,1)
model2<-glm(y2~antibio*toxin,quasibinomial)
summary(model2)
=20
Call:
glm(formula =3D y2 ~ antibio * toxin, family =3D quasibinomial)
=20
Deviance Residuals:=20
Min 1Q Median 3Q Max=20=20
-2.044 -0.627 -0.221 0.709 2.000=20=20
=20
Coefficients:
Estimate Std. Error t value Pr(>|t|)=20=20=20=
=20
(Intercept) -2.90
Re: [Rd] error with R-patched (2008-05-02 r45591)
Prof Brian Ripley wrote:
From the NEWS file for R-patchd:
BUG FIXES
oThe parser sometimes accepted invalid quoted strings in a UTF-8
locale.
This is likely to be the issue, but we still don't have a reproducible
example (see the posting guide).
As Brian probably already knows, the issue is likely to be that your
source file is in a non-UTF-8 encoding. (Emacs has become a little too
smooth in handling such files almost invisibly. The magic incantation is
C-x RET f utf-8
before saving.)
It can be reproduces as follows:
[EMAIL PROTECTED] ~]$ cat > xxx.R
"æøå"
[EMAIL PROTECTED] ~]$ iconv -t latin1 xxx.R > .R
and then R-patched gives
> source("~/.R")
Error in source("~/.R") :
invalid multibyte character in mbcs_get_next at line 1
whereas R-2.6.2 gives
> source(".R")
> .Last.value
$value
[1] "\xe6\xf8\xe5"
$visible
[1] TRUE
On Sat, 3 May 2008, Mark Kimpel wrote:
I have found the cause of the problem I am having with R-patched. In my
.Rprofile I have a function that loads a bunch of my own functions, one
of which contains the character "®". No other version of R I have used
since writing this function ~2 yrs. ago has had a problem reading in
this
character, but apparently R-patched does (invalid multibyte character in
mbcs_get_next at line 17).
That is a parser error message.
Has something intentional or unintentional
occured to cause this? I suspect unintentional since this would seem to
be a loss of functionalilty unlikely to be introduced in a patch of a
release.
Anyway, that's the issue, R-patched doesn't like "®".
I get
"®"
[1] "®"
in R-patched in UTF-8.
Mark
On Sat, May 3, 2008 at 4:20 PM, Prof Brian Ripley
<[EMAIL PROTECTED]>
wrote:
On Sat, 3 May 2008, Mark Kimpel wrote:
Just installed the latest R-patched and get the
error seen at the end of the
output below. Don't know what to make of this but
thought I should report
it.
Nor do we. You have something incorrect in your startup files, but
we don't have those.
In general, when installing the latest version of R,
should one use
R-latest or R-patched? R-latest is stable, but one
would assume that
R-patched includes the latest bug fixes.
And you terminolgy is from where? (Not e.g. the R FAQ.)
Mark
mkimpel-m90 ~/R_HOME/R-patched/R-build/bin: ./R
R version 2.7.0 Patched (2008-05-02 r45591)
Copyright (C) 2008 The R Foundation for Statistical Computing
ISBN 3-900051-07-0
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.
Natural language support but running in an English locale
R is a collaborative project with many contributors.
Type 'contributors()' for more information and
'citation()' on how to cite R or R packages in publications.
Type 'demo()' for some demos, 'help()' for on-line help, or
'help.start()' for an HTML browser interface to help.
Type 'q()' to quit R.
Loading required package: utils
[1] "packages do not need updated"
Error in source(source.files[i]) :
invalid multibyte character in mbcs_get_next at line 17
[Previously saved workspace restored]
sessionInfo()
R version 2.7.0 Patched (2008-05-02 r45591)
x86_64-unknown-linux-gnu
locale:
LC_CTYPE=en_US.UTF-8;LC_NUMERIC=C;LC_TIME=en_US.UTF-8;LC_COLLATE=en_US.UTF-8;LC_
MONETARY=C;LC_MESSAGES=en_US.UTF-8;LC_PAPER=en_US.UTF-8;LC_NAME=C;LC_ADDRESS=C;L
C_TELEPHONE=C;LC_MEASUREMENT=en_US.UTF-8;LC_IDENTIFICATION=C
attached base packages:
[1] stats graphics grDevices datasets utils methods
base
--
Mark W. Kimpel MD ** Neuroinformatics ** Dept. of Psychiatry
Indiana University School of Medicine
15032 Hunter Court, Westfield, IN 46074
(317) 490-5129 Work, & Mobile & VoiceMail
(317) 663-0513 Home (no voice mail please)
**
[[alternative HTML version deleted]]
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
--
Brian D. Ripley, [EMAIL PROTECTED]
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, UKFax: +44 1865 272595
--
Mark W. Kimpel MD ** Neuroinformatics ** Dept. of Psychiatry
Indiana University School of Medicine
15032 Hunter Court, Westfield, IN 46074
(317) 490-5129 Work, & Mobile & VoiceMail
(317) 663-0513 Home (no voice mail please)
**
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
--
O
[Rd] Change in the Tcl/Tk loading in R 2.7.0 (under Unix/Mac OS X)?
Hello,
Up to R 2.6.2, I used to start Tcl *without Tk* (I need only Tcl for
some part of my work, like a socket server written in Tcl only, for
instance) with this code under Mac OS X (particularly on this system,
because I don't want to start X11 just to use Tcl code, which is
required for Tk!):
> Sys.unsetenv("DISPLAY")
> library(tcltk)
I got then the message "no DISPLAY variable so Tk is not available", but
could work with Tcl without problems.
Now, with R 2.7.0, I got the following and Tcl failed to load:
Loading Tcl/Tk interface ... Error in fun(...) : no display name and no
$DISPLAY environment variable
Error : .onLoad failed in 'loadNamespace' for 'tcltk'
Error: package/namespace load failed for 'tcltk'
I try to locate the message "no display name and no $DISPLAY environment
variable" in the code but I cannot find it. Could someone help me please?
> sessionInfo()
R version 2.7.0 Patched (2008-04-22 r45460)
i386-apple-darwin8.10.1
locale:
en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
Many thanks,
Philippe Grosjean
--
..<°}))><
) ) ) ) )
( ( ( ( (Prof. Philippe Grosjean
) ) ) ) )
( ( ( ( (Numerical Ecology of Aquatic Systems
) ) ) ) ) Mons-Hainaut University, Belgium
( ( ( ( (
..
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Change in the Tcl/Tk loading in R 2.7.0 (under Unix/Mac OS X)?
First, this does depend on what build of R you use. The 'HAVE_AQUA'
define matters, and is only defined on some Mac OS X builds (e.g. not the
default one, but the one used for the CRAN binary).
Second, there are several possible versions of Tcl/Tk and X11 on MacOS X,
and as I suspect one of your R versions was not a CRAN build, I am not
willing to make assumptions here. (E.g. I do run a self-compiled
command-line version of R using ActiveTcl's 8.5.2 code.)
Finally, I see no evidence that this is not specific to Mac OS X, and your
desired behaviour is still present on other Unix-alikes.
If you want the ability to start Tcl and not Tk on a CRAN build of R for
Mac OS X, I think you need to negotiate a way to do it as I believe it is
not intended to be available. But that's definitely an R-sig-mac topic.
On Sun, 4 May 2008, Philippe Grosjean wrote:
Hello,
Up to R 2.6.2, I used to start Tcl *without Tk* (I need only Tcl for some
part of my work, like a socket server written in Tcl only, for instance) with
this code under Mac OS X (particularly on this system, because I don't want
to start X11 just to use Tcl code, which is required for Tk!):
Sys.unsetenv("DISPLAY")
library(tcltk)
I got then the message "no DISPLAY variable so Tk is not available", but
could work with Tcl without problems.
That message is only given if (src/library/tcltk/src/tcltk.c in both 2.6.2
and 2.7.0)
#if !defined(Win32) && !defined(HAVE_AQUA)
so your 2.6.x build appears not to have been defining HAVE_AQUA, which a
CRAN build would.
Now, with R 2.7.0, I got the following and Tcl failed to load:
Loading Tcl/Tk interface ... Error in fun(...) : no display name and no
$DISPLAY environment variable
Error : .onLoad failed in 'loadNamespace' for 'tcltk'
Error: package/namespace load failed for 'tcltk'
I try to locate the message "no display name and no $DISPLAY environment
variable" in the code but I cannot find it. Could someone help me please?
I think it is coming from Tk_Init, at
code = Tk_Init(RTcl_interp); /* Load Tk into interpreter */
if (code != TCL_OK)
error(Tcl_GetStringResult(RTcl_interp));
sessionInfo()
R version 2.7.0 Patched (2008-04-22 r45460)
i386-apple-darwin8.10.1
locale:
en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
Many thanks,
Philippe Grosjean
--
..<°}))><
) ) ) ) )
( ( ( ( (Prof. Philippe Grosjean
) ) ) ) )
( ( ( ( (Numerical Ecology of Aquatic Systems
) ) ) ) ) Mons-Hainaut University, Belgium
( ( ( ( (
..
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
--
Brian D. Ripley, [EMAIL PROTECTED]
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, UKFax: +44 1865 272595__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Change in the Tcl/Tk loading in R 2.7.0 (under Unix/Mac OS X)?
Prof Brian Ripley wrote:
First, this does depend on what build of R you use. The 'HAVE_AQUA'
define matters, and is only defined on some Mac OS X builds (e.g. not
the default one, but the one used for the CRAN binary).
Second, there are several possible versions of Tcl/Tk and X11 on MacOS
X, and as I suspect one of your R versions was not a CRAN build, I am
not willing to make assumptions here. (E.g. I do run a self-compiled
command-line version of R using ActiveTcl's 8.5.2 code.)
Finally, I see no evidence that this is not specific to Mac OS X, and
your desired behaviour is still present on other Unix-alikes.
Indeed!
If you want the ability to start Tcl and not Tk on a CRAN build of R for
Mac OS X, I think you need to negotiate a way to do it as I believe it
is not intended to be available. But that's definitely an R-sig-mac topic.
OK, thanks. I repost this question to r-sig-mac.
Philippe
On Sun, 4 May 2008, Philippe Grosjean wrote:
Hello,
Up to R 2.6.2, I used to start Tcl *without Tk* (I need only Tcl for
some part of my work, like a socket server written in Tcl only, for
instance) with this code under Mac OS X (particularly on this system,
because I don't want to start X11 just to use Tcl code, which is
required for Tk!):
Sys.unsetenv("DISPLAY")
library(tcltk)
I got then the message "no DISPLAY variable so Tk is not available",
but could work with Tcl without problems.
That message is only given if (src/library/tcltk/src/tcltk.c in both
2.6.2 and 2.7.0)
#if !defined(Win32) && !defined(HAVE_AQUA)
so your 2.6.x build appears not to have been defining HAVE_AQUA, which a
CRAN build would.
Now, with R 2.7.0, I got the following and Tcl failed to load:
Loading Tcl/Tk interface ... Error in fun(...) : no display name and
no $DISPLAY environment variable
Error : .onLoad failed in 'loadNamespace' for 'tcltk'
Error: package/namespace load failed for 'tcltk'
I try to locate the message "no display name and no $DISPLAY
environment variable" in the code but I cannot find it. Could someone
help me please?
I think it is coming from Tk_Init, at
code = Tk_Init(RTcl_interp); /* Load Tk into interpreter */
if (code != TCL_OK)
error(Tcl_GetStringResult(RTcl_interp));
sessionInfo()
R version 2.7.0 Patched (2008-04-22 r45460)
i386-apple-darwin8.10.1
locale:
en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8
attached base packages:
[1] stats graphics grDevices utils datasets methods base
Many thanks,
Philippe Grosjean
--
..<°}))><
) ) ) ) )
( ( ( ( (Prof. Philippe Grosjean
) ) ) ) )
( ( ( ( (Numerical Ecology of Aquatic Systems
) ) ) ) ) Mons-Hainaut University, Belgium
( ( ( ( (
..
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] R CMD check: small tests suggestion
> > It would be nice if R CMD check ran any file in the tests directory > > that has "one of the extensions .R, .S, .q, .r, or .s" - i.e. it > > should match the files processed in the R directory. > > > > Whereas what it is documented in 'Writing R Extensions' is to use .R or > .Rin files. This leaves the other extensions available for other purposes > -- the difference is that there should be no other files in the R directory, > but often will be in the tests directory. > > I have some sympathy for .r (which we allow for demos) given the prevalence > of case-insensitive file systems, but why would it be 'nice' to have the > others? And what would the output file be called? Should .S be processed > to .Sout and matched to a .Sout.save file? This would seem to introduce > more confusion that it might remove. Those are good points - .r is the extension that I really want as it would have saved me much frustration trying to figure out why my tests weren't running. An alternative would be to report * checking tests ... No test files found rather than * checking tests ... OK in the case in which there are no files run in the tests directory. Hadley -- http://had.co.nz/ __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] A patch for extending pdf device to embed popup text and web links
Dear all, I am sending a patch for R-2.7.0 for extending pdf device to embed pop up text and web links. The patch is available at http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=R-2.7.0.patch&refer=pdf%20device%20patch You can see what the patch can do through a sample output, which is available at http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=sample2.pdf&refer=pdf%20device%20patch An script is at http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=sample2.r&refer=pdf%20device%20patch Ei-ji Nakama kindly build Windows binary, which can be obtained from http://prs.ism.ac.jp/~nakama/tadakadosan/cran/ An installer, R-2.7.0pat-win32.exe, will install programs in C:\Program Files\R\R-2.7.0pat so that you can use both of original and patched version simultaneously. Regards, Tadashi Kadowaki __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] A patch for extending pdf device to embed popup text and web links
Dear Tadashi, Thank you very much for sharing your work. I am sending a patch for R-2.7.0 for extending pdf device to embed pop up text and web links. The patch is available at http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=R-2.7.0.patch&refer=pdf%20device%20patch You can see what the patch can do through a sample output, which is available at http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=sample2.pdf&refer=pdf%20device%20patch An script is at http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=sample2.r&refer=pdf%20device%20patch Ei-ji Nakama kindly build Windows binary, which can be obtained from http://prs.ism.ac.jp/~nakama/tadakadosan/cran/ An installer, R-2.7.0pat-win32.exe, will install programs in C:\Program Files\R\R-2.7.0pat so that you can use both of original and patched version simultaneously. I think it would be most useful if you could bundle your patched version of the pdf device in a separate device package, similar to the RSVGTipsDevice. What do you think ? Kind regards, Tobias __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] S4 / S3 / Sweave problem
I'm not sure if this as a bug or something I am doing wrong. It occurs
in both 2.7.0 and 2.6.2.
require("methods")
setClassUnion("OptionalPOSIXct", c("POSIXct", "NULL"))
setClass("TSmetax",
representation(serIDs="character", ExtractionDate="OptionalPOSIXct" ))
setGeneric("TSmetax",
def= function(x, ...) standardGeneric("TSmetax"))
setMethod("TSmetax", signature(x="character"),
definition= function(x, ...){
new("TSmetax", serIDs=x, ExtractionDate=Sys.time())
} )
> z <- new("TSmetax", serIDs="whatever", ExtractionDate= NULL)
> print(z)
An object of class "TSmetax"
Slot "serIDs":
[1] "whatever"
Slot "ExtractionDate":
NULL
Now if I make this an attribute of an S3 class an print, I get
> zz <- ts(1:10)
> attr(zz, "Meta") <- z
> print(zz)
Time Series:
Start = 1
End = 10
Frequency = 1
[1] 1 2 3 4 5 6 7 8 9 10
attr(,"Meta")
attr(,"serIDs")
[1] "whatever"
attr(,"ExtractionDate")
`NULL`
attr(,"class")
[1] "TSmetax"
attr(,"class")attr(,"package")
[1] ".GlobalEnv"
>
The `NULL` shows up i
shows up in the tex file generated by a vignette as
NULL which causes tex (pdftex) to throw an error.
(It also seems to mess up my mail tool, so I hope this goes through.)
Paul
La version française suit le texte anglais.
This email may contain privileged and/or confidential in...{{dropped:26}}
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] S4 / S3 / Sweave problem
I'm not sure what the issue being asked about here is. The peculiar
printing I don't see and may well be a locale issue. (Is this UTF-8 and
TeX has not been told so?)
The other issue is that when printing an object with attributes, there is
no consideration of S4 classes amongst the attributes. That was either an
undocumented design decision or an omission. This is different from
print(attributes(zz))
$tsp
[1] 1 10 1
$class
[1] "ts"
$Meta
An object of class “TSmetax”
Slot "serIDs":
[1] "whatever"
Slot "ExtractionDate":
NULL
since although attributes are internally a pairlist and printed by a
special C function, attributes() returns a list.
It would be fairly easy to include a branch for S4 objects, but there are
ambuiguities still over what they are (I wrote up a set of questions over
a year ago about this). But it would seem fairly safe to send them to
show() in the same circumstances that autoprinting does for apparently S4
objects.
On Sun, 4 May 2008, Paul Gilbert wrote:
I'm not sure if this as a bug or something I am doing wrong. It occurs in
both 2.7.0 and 2.6.2.
My guess is that it was never intended that S4 objects be used as
attributes, in the same way that one of the unanswered questions is if S4
objects should be allowed to have attributes (other than slots).
require("methods")
setClassUnion("OptionalPOSIXct", c("POSIXct", "NULL"))
setClass("TSmetax",
representation(serIDs="character", ExtractionDate="OptionalPOSIXct" ))
setGeneric("TSmetax",
def= function(x, ...) standardGeneric("TSmetax"))
setMethod("TSmetax", signature(x="character"),
definition= function(x, ...){
new("TSmetax", serIDs=x, ExtractionDate=Sys.time())
} )
z <- new("TSmetax", serIDs="whatever", ExtractionDate= NULL)
print(z)
An object of class "TSmetax"
Slot "serIDs":
[1] "whatever"
Slot "ExtractionDate":
NULL
Now if I make this an attribute of an S3 class an print, I get
zz <- ts(1:10)
attr(zz, "Meta") <- z
print(zz)
Time Series:
Start = 1
End = 10
Frequency = 1
[1] 1 2 3 4 5 6 7 8 9 10
attr(,"Meta")
attr(,"serIDs")
[1] "whatever"
attr(,"ExtractionDate")
`NULL`
attr(,"class")
[1] "TSmetax"
attr(,"class")attr(,"package")
[1] ".GlobalEnv"
The `NULL` shows up i
shows up in the tex file generated by a vignette as
NULL which causes tex (pdftex) to throw an error.
(It also seems to mess up my mail tool, so I hope this goes through.)
--
Brian D. Ripley, [EMAIL PROTECTED]
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, UKFax: +44 1865 272595__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] ~/.Rprofile executed twice by R.app under Mac OS X (PR#11380)
All is in the title. My config is: Version: platform = i386-apple-darwin8.10.1 arch = i386 os = darwin8.10.1 system = i386, darwin8.10.1 status = major = 2 minor = 7.0 year = 2008 month = 04 day = 22 svn rev = 45424 language = R version.string = R version 2.7.0 (2008-04-22) GUI: R-GUI 1.24 (5102) Locale: en_US.UTF-8/en_US.UTF-8/C/C/en_US.UTF-8/en_US.UTF-8 Search Path: .GlobalEnv, tools:RGUI, package:stats, package:graphics, package:grDevices, package:utils, package:datasets, package:methods, Autoloads, package:base -- ..<°}))>< ) ) ) ) ) ( ( ( ( (Prof. Philippe Grosjean ) ) ) ) ) ( ( ( ( (Numerical Ecology of Aquatic Systems ) ) ) ) ) Mons-Hainaut University, Belgium ( ( ( ( ( .. __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] help with segmentation fault
Hello all, I'm trying to have C called by R. I wrote C codes that worked perfectly fine with R-2.6.0 in windows system ( using R tools). I had to change to SuSE Linux (this system has 3.2 GHz Intel Xeon processors and 4 GB of RAM), the C codes were compiled okay but when it was called to R-2.6.2, I got error mesage : ***caught segfault*** address 0x1df5000, cause 'memory not mapped' segmentation fault According to C references, this kind of error occurs when you get something wrong with dereferencing pointers or go out of edge of arrays. But the very same codes were fine in windows and I did not find any related errors in my codes. How could this happen and how can I fix it? Does this have to do with different systems (linux vs windows) or C/R program? Thank you in advance. -- Kyeongmi Univ. Memphis __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] help with segmentation fault
Katie, 99% it's a bug in your C code caused by accessing memory (probably writing to) which was not allocated or freed already or out of boundary. The fact that it worked fine on Windows is likely to be a pure coincidence of different compiler/optimisation settings. However, the main problem is that segfault is such a generic error that without looking at the code one will not be able to tell you anything. Debug it, get ddd with gdb and go step-wise or introduce lots of pritouts to nail down the location of the problem. Oleg Kyeongmi Cheon wrote: Hello all, I'm trying to have C called by R. I wrote C codes that worked perfectly fine with R-2.6.0 in windows system ( using R tools). I had to change to SuSE Linux (this system has 3.2 GHz Intel Xeon processors and 4 GB of RAM), the C codes were compiled okay but when it was called to R-2.6.2, I got error mesage : ***caught segfault*** address 0x1df5000, cause 'memory not mapped' segmentation fault According to C references, this kind of error occurs when you get something wrong with dereferencing pointers or go out of edge of arrays. But the very same codes were fine in windows and I did not find any related errors in my codes. How could this happen and how can I fix it? Does this have to do with different systems (linux vs windows) or C/R program? Thank you in advance. -- Dr Oleg Sklyar * EBI-EMBL, Cambridge CB10 1SD, UK * +44-1223-494466 __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] help with segmentation fault
One thing that may help with debugging (and seeing if you get the same error on windows) is to run your code with R> gctorture(TRUE) R> your.function() That way everything is garbage collected all the time so to speak and this may trigger an error on your windows machine. I agree with Oleg - in my experience it is always a user error. You could also try to compile it with warnings set high(er). Kasper On May 4, 2008, at 2:19 PM, Oleg Sklyar wrote: Katie, 99% it's a bug in your C code caused by accessing memory (probably writing to) which was not allocated or freed already or out of boundary. The fact that it worked fine on Windows is likely to be a pure coincidence of different compiler/optimisation settings. However, the main problem is that segfault is such a generic error that without looking at the code one will not be able to tell you anything. Debug it, get ddd with gdb and go step-wise or introduce lots of pritouts to nail down the location of the problem. Oleg Kyeongmi Cheon wrote: Hello all, I'm trying to have C called by R. I wrote C codes that worked perfectly fine with R-2.6.0 in windows system ( using R tools). I had to change to SuSE Linux (this system has 3.2 GHz Intel Xeon processors and 4 GB of RAM), the C codes were compiled okay but when it was called to R-2.6.2, I got error mesage : ***caught segfault*** address 0x1df5000, cause 'memory not mapped' segmentation fault According to C references, this kind of error occurs when you get something wrong with dereferencing pointers or go out of edge of arrays. But the very same codes were fine in windows and I did not find any related errors in my codes. How could this happen and how can I fix it? Does this have to do with different systems (linux vs windows) or C/R program? Thank you in advance. -- Dr Oleg Sklyar * EBI-EMBL, Cambridge CB10 1SD, UK * +44-1223-494466 __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] A patch for extending pdf device to embed popup text and web links
Hi
Tobias Verbeke wrote:
> Dear Tadashi,
>
> Thank you very much for sharing your work.
>
>> I am sending a patch for R-2.7.0 for extending pdf device to embed pop
>> up text and
>> web links. The patch is available at
>> http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=R-2.7.0.patch&refer=pdf%20device%20patch
>> You can see what the patch can do through a sample output, which is
>> available at
>> http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=sample2.pdf&refer=pdf%20device%20patch
>> An script is at
>> http://www.okada.jp.org/RWiki/?plugin=attach&pcmd=open&file=sample2.r&refer=pdf%20device%20patch
>>
>> Ei-ji Nakama kindly build Windows binary, which can be obtained from
>> http://prs.ism.ac.jp/~nakama/tadakadosan/cran/
>> An installer, R-2.7.0pat-win32.exe, will install programs in
>> C:\Program Files\R\R-2.7.0pat
>> so that you can use both of original and patched version simultaneously.
>
> I think it would be most useful if you could bundle your patched version
> of the pdf device in a separate device package, similar to the
> RSVGTipsDevice.
I would second that approach, at least in the meantime. This sort of
functionality is interesting and useful, but needs careful thought to
make it compatible with the existing graphics facilities in R.
One major consideration is that the graphics system tries to be mostly
device-independent, so that R graphics code can be run on any device.
This would need adjustments to the common graphics engine/device C code
rather than just to the C code underlying the PDF device.
There is also the issue of following standard R coding style and
semantics. Rather than your approach of ...
text(5, 2, "text test")
pdf.link.on.text("http://www.google.com";)
... where the second call relies on a global setting to relate to the
appropriate text, it would be better to have something like ...
linkedText(5, 2, "text test", href="http://www.google.com";)
... so that the association is explicit.
By developing your ideas as an add-on package, people can try out your
ideas without having to recompile R. As Tobias has pointed out, the
RSVGTipsDevice package might be a good place to start.
Paul
> What do you think ?
>
> Kind regards,
> Tobias
>
> __
> R-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
--
Dr Paul Murrell
Department of Statistics
The University of Auckland
Private Bag 92019
Auckland
New Zealand
64 9 3737599 x85392
[EMAIL PROTECTED]
http://www.stat.auckland.ac.nz/~paul/
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] S4 / S3 / Sweave problem
Prof Brian Ripley wrote:
I'm not sure what the issue being asked about here is. The peculiar
printing I don't see and may well be a locale issue. (Is this UTF-8 and
TeX has not been told so?)
It seems to happen in locales UTF-8, POSIX and C. I have not tried others.
The other issue is that when printing an object with attributes, there
is no consideration of S4 classes amongst the attributes. That was
either an undocumented design decision or an omission. This is
different from
print(attributes(zz))
$tsp
[1] 1 10 1
$class
[1] "ts"
$Meta
An object of class “TSmetax”
Slot "serIDs":
[1] "whatever"
Slot "ExtractionDate":
NULL
since although attributes are internally a pairlist and printed by a
special C function, attributes() returns a list.
It would be fairly easy to include a branch for S4 objects, but there
are ambuiguities still over what they are (I wrote up a set of questions
over a year ago about this). But it would seem fairly safe to send them
to show() in the same circumstances that autoprinting does for
apparently S4 objects.
Yes, I do have the impression that print() does not do exactly what one
might like when it finds an S4 attribute in an S3 object. I hope this is
not a design decision. It would make it very difficult to migrate toward
S4 if it is not possible to include S4 objects in S3 objects this way.
On Sun, 4 May 2008, Paul Gilbert wrote:
I'm not sure if this as a bug or something I am doing wrong. It occurs
in both 2.7.0 and 2.6.2.
My guess is that it was never intended that S4 objects be used as
attributes, in the same way that one of the unanswered questions is if S4
objects should be allowed to have attributes (other than slots).
require("methods")
setClassUnion("OptionalPOSIXct", c("POSIXct", "NULL"))
setClass("TSmetax",
representation(serIDs="character", ExtractionDate="OptionalPOSIXct" ))
setGeneric("TSmetax",
def= function(x, ...) standardGeneric("TSmetax"))
setMethod("TSmetax", signature(x="character"),
definition= function(x, ...){
new("TSmetax", serIDs=x, ExtractionDate=Sys.time())
} )
z <- new("TSmetax", serIDs="whatever", ExtractionDate= NULL)
print(z)
An object of class "TSmetax"
Slot "serIDs":
[1] "whatever"
Slot "ExtractionDate":
NULL
Now if I make this an attribute of an S3 class an print, I get
zz <- ts(1:10)
attr(zz, "Meta") <- z
print(zz)
Time Series:
Start = 1
End = 10
Frequency = 1
[1] 1 2 3 4 5 6 7 8 9 10
attr(,"Meta")
attr(,"serIDs")
[1] "whatever"
attr(,"ExtractionDate")
`NULL`
attr(,"class")
[1] "TSmetax"
attr(,"class")attr(,"package")
[1] ".GlobalEnv"
The `NULL` shows up i
shows up in the tex file generated by a vignette as
NULL which causes tex (pdftex) to throw an error.
(It also seems to mess up my mail tool, so I hope this goes through.)
La version française suit le texte anglais.
This email may contain privileged and/or confidential information, and the Bank
of
Canada does not waive any related rights. Any distribution, use, or copying of
this
email or the information it contains by other than the intended recipient is
unauthorized. If you received this email in error please delete it immediately
from
your system and notify the sender promptly by email that you have done so.
Le présent courriel peut contenir de l'information privilégiée ou
confidentielle.
La Banque du Canada ne renonce pas aux droits qui s'y rapportent. Toute
diffusion,
utilisation ou copie de ce courriel ou des renseignements qu'il contient par une
personne autre que le ou les destinataires désignés est interdite. Si vous
recevez
ce courriel par erreur, veuillez le supprimer immédiatement et envoyer sans
délai à
l'expéditeur un message électronique pour l'aviser que vous avez éliminé de
votre
ordinateur toute copie du courriel reçu.
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] troubles with R CMD check and examples under Ubuntu gutsy
Dear listers, I was used to package pgirmess under Windows with everything OK, but, for the first time, I had a trial this afternoon on Ubuntu 7.10 gutsy (I have a double boot computer and work more and more under unix) and R 2.7.0. Everything went OK except this: sudo R CMD check pgirmess . * checking examples ... ERROR Running examples in 'pgirmess-Ex.R' failed. The error most likely occurred in: > ### * PermTest > > flush(stderr()); flush(stdout()) > > ### Name: PermTest > ### Title: Permutation test for lm, lme and glm (binomial and Poisson) > ### objects > ### Aliases: PermTest PermTest.lm PermTest.lme PermTest.glm print.PermTest > ### Keywords: htest > > ### ** Examples > > library(MASS) > mylm<-lm(Postwt~Prewt,data=anorexia) > PermTest(mylm,B=250) Error: could not find function "PermTest" Execution halted If I run R CMD check --no-examples pgirmess, everything comes OK. Seems again that this check makes problem. I googled a bit on R-devel but did not find any understandable post on that... Any idea about what happens ? (please consider if I can write some lines and build a package, I am *not* an advanced fellow in matter of packaging... and just know something about running the commands 'R CMD check' and R CMD build...) Thanks in advance, Patrick __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] troubles with R CMD check and examples under Ubuntu gutsy
It seems likely that the contents of 'pgirmess' under Linux are incomplete. There's a recent version of pgirmess on CRAN, and that works under Linux. I suggest you try that (start with the tarball, perhaps). On Mon, 5 May 2008, Patrick Giraudoux wrote: Dear listers, I was used to package pgirmess under Windows with everything OK, but, for the first time, I had a trial this afternoon on Ubuntu 7.10 gutsy (I have a double boot computer and work more and more under unix) and R 2.7.0. Everything went OK except this: sudo R CMD check pgirmess Do you really want to be doing this under root? I see no reason to do so -- just make sure you have a writeable current directory for your account. . * checking examples ... ERROR Running examples in 'pgirmess-Ex.R' failed. The error most likely occurred in: ### * PermTest flush(stderr()); flush(stdout()) ### Name: PermTest ### Title: Permutation test for lm, lme and glm (binomial and Poisson) ### objects ### Aliases: PermTest PermTest.lm PermTest.lme PermTest.glm print.PermTest ### Keywords: htest ### ** Examples library(MASS) mylm<-lm(Postwt~Prewt,data=anorexia) PermTest(mylm,B=250) Error: could not find function "PermTest" Execution halted If I run R CMD check --no-examples pgirmess, everything comes OK. Seems again that this check makes problem. I googled a bit on R-devel but did not find any understandable post on that... Any idea about what happens ? (please consider if I can write some lines and build a package, I am *not* an advanced fellow in matter of packaging... and just know something about running the commands 'R CMD check' and R CMD build...) Thanks in advance, Patrick -- Brian D. Ripley, [EMAIL PROTECTED] 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, UKFax: +44 1865 272595 __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] error with R-patched (2008-05-02 r45591)
Thanks guys. For the benefit of anyone following up on this in the future, I
solved my problem in my offending file with:
iconv -f LATIN1 -t UTF-8 methods.output.func.R >> methods.output.func.R
Mark
On Sun, May 4, 2008 at 4:13 AM, Peter Dalgaard <[EMAIL PROTECTED]>
wrote:
> Prof Brian Ripley wrote:
>
> > From the NEWS file for R-patchd:
> > >
> >
> > BUG FIXES
> >
> >oThe parser sometimes accepted invalid quoted strings in a UTF-8
> >locale.
> >
> > This is likely to be the issue, but we still don't have a reproducible
> > example (see the posting guide).
> >
> > As Brian probably already knows, the issue is likely to be that your
> source file is in a non-UTF-8 encoding. (Emacs has become a little too
> smooth in handling such files almost invisibly. The magic incantation is
>
> C-x RET f utf-8
>
> before saving.)
>
> It can be reproduces as follows:
>
> [EMAIL PROTECTED] ~]$ cat > xxx.R
> "æøå"
> [EMAIL PROTECTED] ~]$ iconv -t latin1 xxx.R > .R
>
> and then R-patched gives
>
> > source("~/.R")
> Error in source("~/.R") :
> invalid multibyte character in mbcs_get_next at line 1
>
> whereas R-2.6.2 gives
>
> > source(".R")
> > .Last.value
> $value
> [1] "\xe6\xf8\xe5"
>
> $visible
> [1] TRUE
>
>
>
>
>
>
> > On Sat, 3 May 2008, Mark Kimpel wrote:
> >
> > I have found the cause of the problem I am having with R-patched. In my
> > > .Rprofile I have a function that loads a bunch of my own functions,
> > > one
> > > of which contains the character "(R)". No other version of R I have used
> > > since writing this function ~2 yrs. ago has had a problem reading in
> > > this
> > > character, but apparently R-patched does (invalid multibyte character
> > > in
> > > mbcs_get_next at line 17).
> > >
> >
> > That is a parser error message.
> >
> > Has something intentional or unintentional
> > > occured to cause this? I suspect unintentional since this would seem
> > > to
> > > be a loss of functionalilty unlikely to be introduced in a patch of a
> > > release.
> > >
> > > Anyway, that's the issue, R-patched doesn't like "(R)".
> > >
> >
> > I get
> >
> > "(R)"
> > >
> > [1] "(R)"
> >
> > in R-patched in UTF-8.
> >
> >
> > > Mark
> > >
> > >
> > > On Sat, May 3, 2008 at 4:20 PM, Prof Brian Ripley <
> > > [EMAIL PROTECTED]>
> > > wrote:
> > > On Sat, 3 May 2008, Mark Kimpel wrote:
> > >
> > >Just installed the latest R-patched and get the
> > >error seen at the end of the
> > >output below. Don't know what to make of this but
> > >thought I should report
> > >it.
> > >
> > >
> > > Nor do we. You have something incorrect in your startup files, but
> > > we don't have those.
> > >
> > > In general, when installing the latest version of R,
> > > should one use
> > > R-latest or R-patched? R-latest is stable, but one
> > > would assume that
> > > R-patched includes the latest bug fixes.
> > >
> > >
> > > And you terminolgy is from where? (Not e.g. the R FAQ.)
> > >
> > > Mark
> > >
> > > mkimpel-m90 ~/R_HOME/R-patched/R-build/bin: ./R
> > >
> > > R version 2.7.0 Patched (2008-05-02 r45591)
> > > Copyright (C) 2008 The R Foundation for Statistical Computing
> > > ISBN 3-900051-07-0
> > >
> > > 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.
> > >
> > > Natural language support but running in an English locale
> > >
> > > R is a collaborative project with many contributors.
> > > Type 'contributors()' for more information and
> > > 'citation()' on how to cite R or R packages in publications.
> > >
> > > Type 'demo()' for some demos, 'help()' for on-line help, or
> > > 'help.start()' for an HTML browser interface to help.
> > > Type 'q()' to quit R.
> > >
> > > Loading required package: utils
> > > [1] "packages do not need updated"
> > > Error in source(source.files[i]) :
> > > invalid multibyte character in mbcs_get_next at line 17
> > > [Previously saved workspace restored]
> > >
> > > sessionInfo()
> > >
> > > R version 2.7.0 Patched (2008-05-02 r45591)
> > > x86_64-unknown-linux-gnu
> > >
> > > locale:
> > > LC_CTYPE=en_US.UTF-8;LC_NUMERIC=C;LC_TIME=en_US.UTF-8;LC_COLLATE=en_US.UTF-8;LC_
> > >
> > > MONETARY=C;LC_MESSAGES=en_US.UTF-8;LC_PAPER=en_US.UTF-8;LC_NAME=C;LC_ADDRESS=C;L
> > >
> > > C_TELEPHONE=C;LC_MEASUREMENT=en_US.UTF-8;LC_IDENTIFICATION=C
> > >
> > > attached base packages:
> > > [1] stats graphics grDevices datasets utils methods
> > > base
> > >
> > >
> > > --
> > > Mark W. Kimpel MD ** Neuroinformatics ** Dept. of Psychiatry
> > > Indiana University School of Medicine
> > >
> > > 15032 Hunter Court, Westfield, IN 46074
> > >
> > > (317) 490-5129 Work, & Mobile & VoiceMail
> > > (317) 663-0513 Home (no voice mail please)
> > >
> > > **
> > >
> > > [[a
Re: [Rd] error with R-patched (2008-05-02 r45591)
Mark Kimpel wrote: Thanks guys. For the benefit of anyone following up on this in the future, I solved my problem in my offending file with: iconv -f LATIN1 -t UTF-8 methods.output.func.R >> methods.output.func.R Mark That line triggers a knee-jerk reflex with me (possibly unfairly since it apparently did the job). I have always learned that using the same file for input and output is asking for trouble: if something goes wrong on the way, you end with neither of them. I'd do iconv -f LATIN1 -t UTF-8 methods.output.func.R > mv methods.output.func.R -- O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] error with R-patched (2008-05-02 r45591)
You're correct, bad form on my part. In reality, I did it in two steps as you suggested, just condensed for purposes of a brief communication. mark On Mon, May 5, 2008 at 1:52 AM, Peter Dalgaard <[EMAIL PROTECTED]> wrote: > Mark Kimpel wrote: > > > Thanks guys. For the benefit of anyone following up on this in the > > future, I > > solved my problem in my offending file with: > > > > iconv -f LATIN1 -t UTF-8 methods.output.func.R >> methods.output.func.R > > > > Mark > > > > > That line triggers a knee-jerk reflex with me (possibly unfairly since it > apparently did the job). I have always learned that using the same file for > input and output is asking for trouble: if something goes wrong on the way, > you end with neither of them. I'd do > > iconv -f LATIN1 -t UTF-8 methods.output.func.R > > mv methods.output.func.R > > > > > -- > O__ Peter Dalgaard Øster Farimagsgade 5, Entr.B > c/ /'_ --- Dept. of Biostatistics PO Box 2099, 1014 Cph. K > (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918 > ~~ - ([EMAIL PROTECTED]) FAX: (+45) 35327907 > > > -- Mark W. Kimpel MD ** Neuroinformatics ** Dept. of Psychiatry Indiana University School of Medicine 15032 Hunter Court, Westfield, IN 46074 (317) 490-5129 Work, & Mobile & VoiceMail (317) 663-0513 Home (no voice mail please) ** [[alternative HTML version deleted]] __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] troubles with R CMD check and examples under Ubuntu gutsy
Prof Brian Ripley a écrit : It seems likely that the contents of 'pgirmess' under Linux are incomplete. There's a recent version of pgirmess on CRAN, and that works under Linux. I suggest you try that (start with the tarball, perhaps). Sorry prof Ripley, but this is the same version as the one on CRAN (I am the developper)... Actually Kurt Hornik tried it and it went through with his debian yesterday (after some troubles -solved- about encoding and the new R 6.7.0), Uwe Ligges sent me an automated message telling that it has also been built under Windows successfully (as it was under my own Windows version). I have also checked that PermTest was in the package. So I suppose that it has something to do with Ubuntu... but don't know how to debug this... On Mon, 5 May 2008, Patrick Giraudoux wrote: Dear listers, I was used to package pgirmess under Windows with everything OK, but, for the first time, I had a trial this afternoon on Ubuntu 7.10 gutsy (I have a double boot computer and work more and more under unix) and R 2.7.0. Everything went OK except this: sudo R CMD check pgirmess Do you really want to be doing this under root? I see no reason to do so -- just make sure you have a writeable current directory for your account. . * checking examples ... ERROR Running examples in 'pgirmess-Ex.R' failed. The error most likely occurred in: ### * PermTest flush(stderr()); flush(stdout()) ### Name: PermTest ### Title: Permutation test for lm, lme and glm (binomial and Poisson) ### objects ### Aliases: PermTest PermTest.lm PermTest.lme PermTest.glm print.PermTest ### Keywords: htest ### ** Examples library(MASS) mylm<-lm(Postwt~Prewt,data=anorexia) PermTest(mylm,B=250) Error: could not find function "PermTest" Execution halted If I run R CMD check --no-examples pgirmess, everything comes OK. Seems again that this check makes problem. I googled a bit on R-devel but did not find any understandable post on that... Any idea about what happens ? (please consider if I can write some lines and build a package, I am *not* an advanced fellow in matter of packaging... and just know something about running the commands 'R CMD check' and R CMD build...) Thanks in advance, Patrick __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] troubles with R CMD check and examples under Ubuntu gutsy
Berwin A Turlach a écrit : G'day Patrick On Mon, 05 May 2008 07:02:23 +0200 Patrick Giraudoux <[EMAIL PROTECTED]> wrote: I was used to package pgirmess under Windows with everything OK, but, for the first time, I had a trial this afternoon on Ubuntu 7.10 gutsy (I have a double boot computer and work more and more under unix) and R 2.7.0. FWIW, on my Ubuntu 8.04 heron machine, with pgirmess_1.3.6.tar.gz from CRAN, I can run `R CMD check pgirmess_1.3.6.tar.gz' without problems. Are you talking about that version, or a current development version? But I have compiled R on my boxes myself. How did you install R? From source or by using the prepackaged Ubuntu deb files? From the pre-packaged Ubuntu, updated to 6.7.0 recently ? Do you think it may come from this ? sudo R CMD check pgirmess Why do you run this command as root? I would avoid something like this. Is the directory in which pgirmess is in not writeable for your usual user-id? Perhaps the problem is that the root account is not correctly set up to use R (but I would find that hard to believe). No, sorry it was just my mistake. After the failure as standard user, I was wondering if packaging should be done as super-user... but things come similar. Patrick __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] troubles with R CMD check and examples under Ubuntu gutsy
On Mon, 5 May 2008, Patrick Giraudoux wrote: Prof Brian Ripley a écrit : It seems likely that the contents of 'pgirmess' under Linux are incomplete. There's a recent version of pgirmess on CRAN, and that works under Linux. I suggest you try that (start with the tarball, perhaps). Sorry prof Ripley, but this is the same version as the one on CRAN (I am the developper)... Actually Kurt Hornik tried it and it went through with his debian yesterday (after some troubles -solved- about encoding and the new R Yes, I do know (I solved them). 6.7.0), Uwe Ligges sent me an automated message telling that it has also been built under Windows successfully (as it was under my own Windows version). I have also checked that PermTest was in the package. So I suppose that it has something to do with Ubuntu... but don't know how to debug this... I suggest you do as I suggest, for the most likely explanation is that this is not the same set of files. On Mon, 5 May 2008, Patrick Giraudoux wrote: Dear listers, I was used to package pgirmess under Windows with everything OK, but, for the first time, I had a trial this afternoon on Ubuntu 7.10 gutsy (I have a double boot computer and work more and more under unix) and R 2.7.0. Everything went OK except this: sudo R CMD check pgirmess Do you really want to be doing this under root? I see no reason to do so -- just make sure you have a writeable current directory for your account. . * checking examples ... ERROR Running examples in 'pgirmess-Ex.R' failed. The error most likely occurred in: ### * PermTest flush(stderr()); flush(stdout()) ### Name: PermTest ### Title: Permutation test for lm, lme and glm (binomial and Poisson) ### objects ### Aliases: PermTest PermTest.lm PermTest.lme PermTest.glm print.PermTest ### Keywords: htest ### ** Examples library(MASS) mylm<-lm(Postwt~Prewt,data=anorexia) PermTest(mylm,B=250) Error: could not find function "PermTest" Execution halted If I run R CMD check --no-examples pgirmess, everything comes OK. Seems again that this check makes problem. I googled a bit on R-devel but did not find any understandable post on that... Any idea about what happens ? (please consider if I can write some lines and build a package, I am *not* an advanced fellow in matter of packaging... and just know something about running the commands 'R CMD check' and R CMD build...) Thanks in advance, Patrick -- Brian D. Ripley, [EMAIL PROTECTED] 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, UKFax: +44 1865 272595__ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] troubles with R CMD check and examples under Ubuntu gutsy
Berwin A Turlach a écrit :
G'day Patrick,
On Mon, 05 May 2008 08:15:29 +0200
Patrick Giraudoux <[EMAIL PROTECTED]> wrote:
Berwin A Turlach a écrit :
But I have compiled R on my boxes myself. How did you install R?
>From source or by using the prepackaged Ubuntu deb files?
From the pre-packaged Ubuntu, updated to 6.7.0 recently ? Do you
think it may come from this ?
Could be possible, but I would find it hard to believe. But it raises
the question which pre-packaged packages have you all installed. Your
package pgirmess seems to have quite a few dependencies (which I had
to install first on my machine before I could run R CMD check). But
that could be a red herring too, since the problem seems to be that the
command PermTest is not found while the code in the examples are run.
There has been a threat on something similar in R-devel on August
2007... but I cannot understand the meaning.
Did you look at the file in pgirmess.Rcheck that I mentioned?
You'll find a copy of pgirmess-Ex.R, but I cannot get what is the
meaning of this file...
Also,
did you try to load the version of the library in pgrimess.Rcheck in a
R session and check if you find the PermTest function in that case?
Yes indeed. I found PermTest in the R folder and in the R-ex folder of
the pgirmess folder in the pgirmess.Rcheck folder.
With these problems, I am a bit of a trial-and-error person, so if I
cannot reproduce the error, I cannot find out what the problem is
Cheers,
Berwin
Thanks anyway for the hints. It may help.
### *
###
attach(NULL, name = "CheckExEnv")
assign("nameEx",
local({
s <- "__{must remake R-ex/*.R}__"
function(new) {
if(!missing(new)) s <<- new else s
}
}),
pos = "CheckExEnv")
## Add some hooks to label plot pages for base and grid graphics
assign("base_plot_hook",
function() {
pp <- par(c("mfg","mfcol","oma","mar"))
if(all(pp$mfg[1:2] == c(1, pp$mfcol[2]))) {
outer <- (oma4 <- pp$oma[4]) > 0; mar4 <- pp$mar[4]
mtext(sprintf("help(\"%s\")", nameEx()), side = 4,
line = if(outer)max(1, oma4 - 1) else min(1, mar4 - 1),
outer = outer, adj = 1, cex = .8, col = "orchid", las=3)
}
},
pos = "CheckExEnv")
assign("grid_plot_hook",
function() {
grid::pushViewport(grid::viewport(width=grid::unit(1, "npc") -
grid::unit(1, "lines"), x=0, just="left"))
grid::grid.text(sprintf("help(\"%s\")", nameEx()),
x=grid::unit(1, "npc") + grid::unit(0.5,
"lines"),
y=grid::unit(0.8, "npc"), rot=90,
gp=grid::gpar(col="orchid"))
},
pos = "CheckExEnv")
setHook("plot.new", get("base_plot_hook", pos = "CheckExEnv"))
setHook("persp",get("base_plot_hook", pos = "CheckExEnv"))
setHook("grid.newpage", get("grid_plot_hook", pos = "CheckExEnv"))
assign("cleanEx",
function(env = .GlobalEnv) {
rm(list = ls(envir = env, all.names = TRUE), envir = env)
RNGkind("default", "default")
set.seed(1)
options(warn = 1)
.CheckExEnv <- as.environment("CheckExEnv")
delayedAssign("T", stop("T used instead of TRUE"),
assign.env = .CheckExEnv)
delayedAssign("F", stop("F used instead of FALSE"),
assign.env = .CheckExEnv)
sch <- search()
newitems <- sch[! sch %in% .oldSearch]
for(item in rev(newitems))
eval(substitute(detach(item), list(item=item)))
missitems <- .oldSearch[! .oldSearch %in% sch]
if(length(missitems))
warning("items ", paste(missitems, collapse=", "),
" have been removed from the search path")
},
pos = "CheckExEnv")
assign("ptime", proc.time(), pos = "CheckExEnv")
## at least one package changes these via ps.options(), so do this
## before loading the package.
## Use postscript as incomplete files may be viewable, unlike PDF.
## Choose a size that is close to on-screen devices, fix paper
ps.options(width = 7, height = 7, paper = "a4", reset = TRUE)
grDevices::postscript("pgirmess-Ex.ps")
assign("par.postscript", graphics::par(no.readonly = TRUE), pos =
"CheckExEnv")
options(contrasts = c(unordered = "contr.treatment", ordered =
"contr.poly"))
options(warn = 1)
library('pgirmess')
assign(".oldSearch", search(), pos = 'CheckExEnv')
assign(".oldNS", loadedNamespaces(), pos = 'CheckExEnv')
cleanEx(); nameEx("PermTest")
### * PermTest
flush(stderr()); flush(stdout())
### Name: PermTest
### Title: Permutation test for lm, lme and glm (binomial and Poisson)
### objects
### Aliases: PermTest PermTest.lm PermTest.lme PermTest.glm print.PermTest
### Keywords: htest
### ** Examples
library(MASS)
mylm<-lm(Postwt~Prewt,data=anorexia)
PermTest(mylm,B=250)
## Dobson (1990) Page 93: Random
