Re: [Rd] gfortran optimization problems
Dave, yes, we have experienced different results in different platforms, the problem was partly due to the different versions of gfortran installed across platforms, are you using the same version for all the platforms? Clearly, gfortran is undergoing rapid changes so that the results might vary a lot across versions. You might want to contact the developers should you find out a compiler bug. Kind regards Simone On Sat, Oct 25, 2008 at 12:53 AM, Dave Roberts <[EMAIL PROTECTED]> wrote: > Colleagues, > > I have a routine in package labdsv that calls a FORTRAN subroutine. > Recently, I was informed that it sometimes gives different results on a PC > and Mac, and that the PC version is clearly wrong. I tested it on linux > (because I don't have a PC), and I get the same (incorrect) behavior as the > PC. > > Simply by inserting debug WRITE statements in the FORTRAN I would get > different, and correct, results, so I assumed it was an optimization error. > > So, > > 1) R CMD SHLIB duleg.fdoes not work, and produces bogus code > > however, > > 2) gfortran -c alt_duleg.f > gcc -O -std=gnu99 -shared -L/usr/local/lib -o alt_duleg.so >alt_duleg.o -lgfortran -lm -lgcc_s > > does work (note the -O low optimization). Otherwise the gcc command is > identical to the one produced by R CMD SHLIB. > > Has anyone else seen this? Is there a simple way to have my package > enforce no optimization so others don't struggle with this? As far as I > know the same code worked under g77. > > Thanks, Dave Roberts > -- > > David W. Roberts office 406-994-4548 > Professor and Head FAX 406-994-3190 > Department of Ecology email [EMAIL PROTECTED] > Montana State University > Bozeman, MT 59717-3460 > > __ > R-devel@r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-devel > -- __ Simone Giannerini Dipartimento di Scienze Statistiche "Paolo Fortunati" Universita' di Bologna Via delle belle arti 41 - 40126 Bologna, ITALY Tel: +39 051 2098262 Fax: +39 051 232153 http://www2.stat.unibo.it/giannerini/ __ [[alternative HTML version deleted]] __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] pairwise.wilcox.test for paired samples
Dear R Core, pairwise.wilcox.test does not handle "paired = TRUE" correctly; e.g. set.seed(13) x <- rnorm(20) g <- c(rep(1, 10), rep(2, 10)) wilcox.test(x ~ g)$p.value # 0.075 pairwise.wilcox.test(x, g)$p.value # 0.075, o.k wilcox.test(x ~ g, paired = TRUE)$p.value # 0.105 pairwise.wilcox.test(x, g, paired = TRUE)$p.value # 0.075, wrong The line wilcox.test(xi, xj, ...)$p.value should read wilcox.test(xi, xj, paired = paired, ...)$p.value Best regards, Matthias my sessionInfo: R version 2.8.0 Patched (2008-10-23 r46779) i686-pc-linux-gnu locale: LC_CTYPE=de_DE.UTF-8;LC_NUMERIC=C;LC_TIME=de_DE.UTF-8;LC_COLLATE=de_DE.UTF-8;LC_MONETARY=C;LC_MESSAGES=de_DE.UTF-8;LC_PAPER=de_DE.UTF-8;LC_NAME=C;LC_ADDRESS=C;LC_TELEPHONE=C;LC_MEASUREMENT=de_DE.UTF-8;LC_IDENTIFICATION=C attached base packages: [1] stats graphics grDevices utils datasets methods base -- Dr. Matthias Kohl www.stamats.de __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] R echoes integer literals incorrectly, under certain conditions (PR#13204)
Full_Name: Stephen Milborrow
Version: R 2.8.0
OS: Windows XP
Submission from: (NULL) (81.159.65.46)
Under certain conditions, R 2.8.0 incorrectly echoes integer literals without
the "L" suffix. R 2.7.2 is ok.
To reproduce, create two files as follows:
a.R containing: source("b.R", echo=TRUE)
b.R containing: x <- 1L
Now execute: Rterm --vanilla --silent sessionInfo()
R version 2.8.0 (2008-10-20)
i386-pc-mingw32
locale:
LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
States.1252;LC_MONETARY=English_United
States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
www.milbo.users.sonic.net
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] 0 ^ NaN == Inf, why?
In R, 0 ^ NaN yields Inf. I would have expected NaN or perhaps 0. Is this behaviour intended? >sessionInfo() R version 2.8.0 (2008-10-20) i386-pc-mingw32 locale: LC_COLLATE=English_United States.1252;LC_CTYPE=English_United States.1252;LC_MONETARY=English_United States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base Steve Milborrow www.milbo.users.sonic.net __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] R echoes integer literals incorrectly, under certain conditions (PR#13204)
On 25/10/2008 10:45 AM, [EMAIL PROTECTED] wrote:
Full_Name: Stephen Milborrow
Version: R 2.8.0
OS: Windows XP
Submission from: (NULL) (81.159.65.46)
Under certain conditions, R 2.8.0 incorrectly echoes integer literals without
the "L" suffix. R 2.7.2 is ok.
To reproduce, create two files as follows:
a.R containing: source("b.R", echo=TRUE)
b.R containing: x <- 1L
Now execute: Rterm --vanilla --silent
It works fine for me. Could you try changing b.R to contain a comment, e.g.
x <- # a comment
1L
and let us know what you see in the two cases? I'm guessing you won't
see the comment being echoed when you source a.R because for some reason
your system is deparsing instead of going back to the source to read it,
but I don't know what'll happen with b.R.
Duncan Murdoch
sessionInfo()
R version 2.8.0 (2008-10-20)
i386-pc-mingw32
locale:
LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
States.1252;LC_MONETARY=English_United
States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
attached base packages:
[1] stats graphics grDevices utils datasets methods base
www.milbo.users.sonic.net
__
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 echoes integer literals incorrectly, under certain conditions (PR#13205)
On 25/10/2008 10:45 AM, [EMAIL PROTECTED] wrote:
> Full_Name: Stephen Milborrow
> Version: R 2.8.0
> OS: Windows XP
> Submission from: (NULL) (81.159.65.46)
>
>
> Under certain conditions, R 2.8.0 incorrectly echoes integer literals without
> the "L" suffix. R 2.7.2 is ok.
>
> To reproduce, create two files as follows:
> a.R containing: source("b.R", echo=TRUE)
> b.R containing: x <- 1L
>
> Now execute: Rterm --vanilla --silent This echoes:x <- 1
> It should echo: x <- 1L
>
> This behaviour does not occur if you source("b.R") directly from the R command
> prompt, or if you use R 2.7.2.
It works fine for me. Could you try changing b.R to contain a comment, e.g.
x <- # a comment
1L
and let us know what you see in the two cases? I'm guessing you won't
see the comment being echoed when you source a.R because for some reason
your system is deparsing instead of going back to the source to read it,
but I don't know what'll happen with b.R.
Duncan Murdoch
>
>> sessionInfo()
> R version 2.8.0 (2008-10-20)
> i386-pc-mingw32
>
> locale:
> LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> States.1252;LC_MONETARY=English_United
> States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
>
> attached base packages:
> [1] stats graphics grDevices utils datasets methods base
>
> www.milbo.users.sonic.net
>
> __
> 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] pairwise.wilcox.test for paired samples
Matthias Kohl wrote: Dear R Core, pairwise.wilcox.test does not handle "paired = TRUE" correctly; e.g. The line wilcox.test(xi, xj, ...)$p.value should read wilcox.test(xi, xj, paired = paired, ...)$p.value Yes, fixed now. Thanks. -pd -- 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] 0 ^ NaN == Inf, why?
Stephen Milborrow wrote:
> In R, 0 ^ NaN yields Inf. I would have expected NaN or perhaps 0. Is this
> behaviour intended?
>
Well, it certainly follows from the implementation. In the R_pow C routine,
double R_pow(double x, double y) /* = x ^ y */
{
if(x == 1. || y == 0.)
return(1.);
if(x == 0.) {
if(y > 0.) return(0.);
/* y < 0 */ return(R_PosInf);
}
It does seem, however, that NaN is the logical result here, which I
think results from changing the implementation to:
if(x == 0.) {
if(y > 0.) return(0.);
else if(y < 0) return(R_PosInf);
else return(y); /* NA or NaN, we assert */
}
Other things being equal, `^` should follow the C pow() function for
numeric arguments. After writing pow() as an R function that calls this
C function:
> pow(0,NaN)
[1] NaN
> pow(0,NA)
[1] NA
> pow(0,0)
[1] 1
The second example presumably falls out because the C function returns
its second argument if that is a NaN (which a numeric NA is, in C but
not in R). The modified code in R_pow mimics that behavior.
Along the line, notice that both R_pow and pow give 0^0 as 1. (Just at
a guess, C might give 0^-0 as Inf, but I don't know how to test that in R.)
Other opinions?
John
>
>> sessionInfo()
>>
> R version 2.8.0 (2008-10-20)
> i386-pc-mingw32
>
> locale:
> LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> States.1252;LC_MONETARY=English_United
> States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252
>
> attached base packages:
> [1] stats graphics grDevices utils datasets methods base
>
> Steve Milborrow
> www.milbo.users.sonic.net
>
> __
> R-devel@r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-devel
>
>
[[alternative HTML version deleted]]
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] 0 ^ NaN == Inf, why?
A small PS: John Chambers wrote: Along the line, notice that both R_pow and pow give 0^0 as 1. (Just at a guess, C might give 0^-0 as Inf, but I don't know how to test that in R.) I tried a little harder, and apparently the guess is wrong. It seems that pow(0, -0) is 1 in C. Would seem better to either have pow(0,0) and pow(0,-0) both be NaN or else 1 and Inf, but ... John __ 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
[Rd] Dynamic linking to binary code from other packages??
Dear R-devel, I am writing a package that needs some C++ functions from an external SDK (Affymetrix fusion). The same functions are already compiled into another package (affxparser). Can I dynamically link to the compiled code in affxparser, rather than re-compiling these separately into my package? Not a lot of experience on this subject! Thanks, Charles __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
Re: [Rd] Dynamic linking to binary code from other packages??
On 25 October 2008 at 16:05, Charles Danko wrote: | Dear R-devel, | | I am writing a package that needs some C++ functions from an external | SDK (Affymetrix fusion). The same functions are already compiled into | another package (affxparser). | | Can I dynamically link to the compiled code in affxparser, rather than | re-compiling these separately into my package? Yes. Working examples are lme4 using Matrix. This is documented in the 'R Extensions' manual. | Not a lot of experience on this subject! I don't think it is the easiest way around. Maybe start by 'just' recompiling and relinking. Dirk -- Three out of two people have difficulties with fractions. __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] Compilation error when X11 is off (PR#13206)
Full_Name: Yegor Bryukhov Version: 2.8.0 OS: Linux Submission from: (NULL) (69.22.221.161) in file src/platform.c line 1657 - a semi-colon is missing for the case when X11 is turned off __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] R CMD Rd2dvi (PR#13200)
Hello, I can build packages with R-2.8.0 but I cannot run the code: R CMD Rd2dvi --pdf --no-clean =20 Error message: .: Can't open C:/Apps/R/R280/share/sh/dcf.sh: No such file or directory The system cannot find the path specified. =20 The above code works fine in R-2.7.2. =20 Comparing the bash script Rd2dvi.sh between R-2.7.2 and R-2.8.0 makes me think that the script is incorrectly determining my system to be UNIX rather than WINDOWS. I tried the option --os=3Dwindows without any success. =20 Good luck.=20 Cheers, Rowan =20 Sys.getenv() ADMB_HOMEC:\Apps\ADMB=09 ALLUSERSPROFILE C:\Documents and Settings\All Users=09 APPDATA C:\Documents and Settings\haighr\Application Data=09 CommonProgramFiles C:\Program Files\Common Files=09 COMPUTERNAME PAC03470=09 ComSpec C:\WINDOWS\system32\cmd.exe=09 DEFLOGDIRC:\Documents and Settings\All Users\Application Data\McAfee\DesktopProtection=09 FP_NO_HOST_CHECK NO=09 GFORTRAN_STDERR_UNIT -1=09 GFORTRAN_STDOUT_UNIT -1=09 HOME E:\Temp\MyDocs=09 HOMEDRIVEC:=09 HOMEPATH \Documents and Settings\haighr=09 HOMESHARE\\pacpbsfp1\HaighR$=09 LOGONSERVER \\PACIOSADC1=09 NUMBER_OF_PROCESSORS 2=09 OS Windows_NT=09 PATH E:\Archive\Bat;C:\Apps\R\Rtools\bin;C:/Apps/Dev-Cpp/bin;C:\Apps\MiKTeX\m iktex\bin;C:\Apps\R\Rtools\perl\bin;C:\Apps\R\Rtools\MinGW\bin;C:\Apps\P ython25\Scripts;C:\Apps\Ora81\bin;C:\WINDOWS\system32;C:\WINDOWS;C:\WIND OWS\System32\Wbem;C:\Apps\UltraEdit\=09 PATHEXT .COM;.EXE;.BAT;.CMD;.VBS;.VBE;.JS;.JSE;.WSF;.WSH=09 PROCESSOR_ARCHITECTURE x86=09 PROCESSOR_IDENTIFIER x86 Family 6 Model 15 Stepping 6, GenuineIntel=09 PROCESSOR_LEVEL 6=09 PROCESSOR_REVISION 0f06=09 ProgramFiles C:\Program Files=09 R_HOME C:\Apps\R\R280=09 R_LIBS_USER E:\Temp\MyDocs/R/win-library/2.8=09 R_USER E:\Temp\MyDocs=09 SESSIONNAME Console=09 SystemDrive C:=09 SystemRoot C:\WINDOWS=09 TEMP C:\Temp\HaighR=09 TMP C:\Temp\HaighR=09 USERDNSDOMAINPAC.DFO-MPO.CA=09 USERDOMAIN PAC=09 USERNAME HaighR=09 USERPROFILE C:\Documents and Settings\haighr=09 VSEDEFLOGDIR C:\Documents and Settings\All Users\Application Data\McAfee\DesktopProtection=09 windir C:\WINDOWS=09 =20 Rowan Haigh Research Biologist, Groundfish Section Marine Ecosystems & Aquaculture Division Fisheries and Oceans Canada Pacific Biological Station, Nanaimo, BC V9T 6N7 Tel. +1 250-756-7123 fac simile +1 250-756-7053 Email [EMAIL PROTECTED] =20 [[alternative HTML version deleted]] __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] write.csv(..., col.names = FALSE) (PR#13202)
Full_Name: Stefan Albrecht
Version: 2.7.2 (and 2.8.0)
OS: Windows NT
Submission from: (NULL) (194.127.8.17)
Dear R Debug-Team,
in write.csv() it is not possible to set both
row.names = FALSE, col.names = FALSE
since the col.names = FALSE gets overwritten:
> write.csv
function (...)
{
Call <- match.call(expand.dots = TRUE)
for (argname in c("col.names", "sep", "dec", "qmethod")) if
(!is.null(Call[[argname]]))
warning(gettextf("attempt to set '%s' ignored", argname),
domain = NA)
rn <- eval.parent(Call$row.names)
Call$col.names <- if (is.logical(rn) && !rn)
TRUE
else NA
Call$sep <- ","
Call$dec <- "."
Call$qmethod <- "double"
Call[[1]] <- as.name("write.table")
eval.parent(Call)
}
>
This is not documented and should be changed, as it is an unintended reaction.
With many thanks and best regards,
Stefan
__
R-devel@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-devel
[Rd] Errors in R CMD check for R 2.8.0
useR's, I get these 2 errors when I run "R CMD check" on my package, and I cannot figure out how to get around them. Does anyone have any ideas? If it is of any help, I use Windows XP. ... * checking replacement functions OK * checking foreign function calls ... OK * checking R code for possible problems ... NOTE Error in inDL(x, as.logical(local), as.logical(now), ...): unable to load shared library 'c:/Progra~1/R/R-2.8.0/library/tcltk/libs/tcltk.dll': LoadLibrary failure: The specified procedure could not be found. Error: .onLoad failed in 'loadNamespace' for 'tcltk' * Checking .Rd files ... OK * checking data for non-ASCII characters ... OK * creating kzs-Ex.R ... OK * checking examples ... ERROR Running examples in 'kzs-Ex.R' failed. I have no idea how to solve the first problem with references the tcltk package and I do not know why my examples won't run. Any help is appreciated. dxc13 -- View this message in context: http://www.nabble.com/Errors-in-R-CMD-check-for-R-2.8.0-tp20150565p20150565.html Sent from the R devel mailing list archive at Nabble.com. __ R-devel@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-devel
