An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-sig-mac/attachments/20091130/a6b6bd52/attachment.pl>
Built on Mac, Installing on Windows - should this work?
8 messages · Bryan Hanson, Brian Ripley, Duncan Murdoch +1 more
On 30/11/2009 11:10 AM, Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed. Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed.
The source tarball (the .tar.gz file) would be fine on Windows, but most Windows users don't have the tools to install it. Since you don't have any compiled code it's probably not hard to put together a zip file equivalent to what R CMD INSTALL --binary would produce on Windows, but it's probably easier to use Uwe Ligge's package building service, available at http://129.217.207.166. Duncan Murdoch
By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]] ------------------------------------------------------------------------
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
On Nov 30, 2009, at 11:10 , Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed.
You need the *binary* package for windows. If you're lucky, all you have to do is to install the package on your Mac and zip it up (the *installed* package, not the source!). So for example mkdir /tmp/pkg R CMD INSTALL -l /tmp/pkg ChemoSpec cd /tmp/pkg zip -r ChemoSpec_1.0.zip ChemoSpec If you get unlucky, you have to use a Windows machine (or more conveniently Windows in a VirtualBox ;)) to build the binary package. (There are ways to cross-compile, but if you have no native code you should not need to go there ...). Cheers, Simon
Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed. By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]]
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
On Mon, 30 Nov 2009, Duncan Murdoch wrote:
On 30/11/2009 11:10 AM, Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed. Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed.
The source tarball (the .tar.gz file) would be fine on Windows, but most Windows users don't have the tools to install it. Since you don't have any compiled code it's probably not hard to put together a zip file equivalent to what R CMD INSTALL --binary would produce on Windows, but it's probably easier to use Uwe Ligge's package building service, available at http://129.217.207.166.
I agree with the conclusion (not least as it checks the package works on Windows), but in current R (2.10.0) you actually don't need any extra tools to install such a source package on Windows. It's all in the manuals ....
Duncan Murdoch
By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]] ------------------------------------------------------------------------
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
Brian D. Ripley, ripley at 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
Hello Folks, thanks for the suggestions. I did what Simon suggested. And I did what Ducan suggested, namely to use Uwe's building service. I think both of these produced the same set of files (though I don't know how to check rigorously). So I moved these two archives, along with my source archive, over to the Windows machine. I tried installing all of them via the GUI dialog boxes, and when that didn't work, I tried doing it from the command line. I'm pretty certain I'm now no longer in the Mac realm :-( and I'm pretty certain I'm missing something obvious. It looks like it can't find the file, but the paths and names look right to me. If one unstuffs these manually they look like they have the files. Any suggestions would be appreciated. The transcript from what I did is immediately below. Thanks, Bryan ************* Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University, Greencastle IN USA
getwd()
[1] "C:/Documents and Settings/Chemistry 1/Desktop"
list.files()
[1] "ChemoSpec_1.31-SU.zip" "ChemoSpec_1.31-UL.zip" "ChemoSpec_1.31.tar.gz" "outcome.txt" # the next few steps are attempts at manual installations:
install.packages(pkgs = "ChemoSpec_1.31-SU.zip", repos = NULL)
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, type =
"source") * installing *source* package 'ChemoSpec' ... ** R ** data ** inst ** preparing package for lazy loading Warning in library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'chemometrics' Error : package 'chemometrics' could not be loaded ERROR: lazy loading failed for package 'ChemoSpec' * removing 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' * restoring previous 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' Warning message: In install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, : installation of package 'ChemoSpec_1.31.tar.gz' had non-zero exit status # the next few are the attempts with the GUI dialog boxes
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-UL/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning messages: 1: In unzip(zipname, exdir = dest) : error 1 in extracting from zip file 2: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31.tar.gz/DESCRIPTION', probable reason 'No such file or directory'
sessionInfo()
R version 2.10.0 (2009-10-26) i386-pc-mingw32 locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C [5] LC_TIME=English_United States.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached): [1] tools_2.10.0
On 11/30/09 11:48 AM, "Prof Brian Ripley" <ripley at stats.ox.ac.uk> wrote:
On Mon, 30 Nov 2009, Duncan Murdoch wrote:
On 30/11/2009 11:10 AM, Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed. Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed.
The source tarball (the .tar.gz file) would be fine on Windows, but most Windows users don't have the tools to install it. Since you don't have any compiled code it's probably not hard to put together a zip file equivalent to what R CMD INSTALL --binary would produce on Windows, but it's probably easier to use Uwe Ligge's package building service, available at http://129.217.207.166.
I agree with the conclusion (not least as it checks the package works on Windows), but in current R (2.10.0) you actually don't need any extra tools to install such a source package on Windows. It's all in the manuals ....
Duncan Murdoch
By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]] ------------------------------------------------------------------------
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
On 30/11/2009 2:30 PM, Bryan Hanson wrote:
Hello Folks, thanks for the suggestions. I did what Simon suggested. And I did what Ducan suggested, namely to use Uwe's building service. I think both of these produced the same set of files (though I don't know how to check rigorously). So I moved these two archives, along with my source archive, over to the Windows machine. I tried installing all of them via the GUI dialog boxes, and when that didn't work, I tried doing it from the command line. I'm pretty certain I'm now no longer in the Mac realm :-( and I'm pretty certain I'm missing something obvious. It looks like it can't find the file, but the paths and names look right to me. If one unstuffs these manually they look like they have the files. Any suggestions would be appreciated. The transcript from what I did is immediately below. Thanks, Bryan ************* Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University, Greencastle IN USA
getwd()
[1] "C:/Documents and Settings/Chemistry 1/Desktop"
list.files()
[1] "ChemoSpec_1.31-SU.zip" "ChemoSpec_1.31-UL.zip" "ChemoSpec_1.31.tar.gz" "outcome.txt" # the next few steps are attempts at manual installations:
install.packages(pkgs = "ChemoSpec_1.31-SU.zip", repos = NULL)
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
Looks like your DESCRIPTION file was missing in this try.
install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, type =
"source") * installing *source* package 'ChemoSpec' ... ** R ** data ** inst ** preparing package for lazy loading Warning in library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'chemometrics'
Looks as through ChemoSpec depends on chemometrics, and you need to install that first. By default Windows will install dependencies when it's getting something from CRAN, but not when you're installing a local package with repos=NULL. So you need to install chemometrics first, then this way should work. The first one appears to be failing because the .zip file is badly formed. Duncan Murdoch
Error : package 'chemometrics' could not be loaded ERROR: lazy loading failed for package 'ChemoSpec' * removing 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' * restoring previous 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' Warning message: In install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, : installation of package 'ChemoSpec_1.31.tar.gz' had non-zero exit status # the next few are the attempts with the GUI dialog boxes
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-UL/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning messages: 1: In unzip(zipname, exdir = dest) : error 1 in extracting from zip file 2: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31.tar.gz/DESCRIPTION', probable reason 'No such file or directory'
sessionInfo()
R version 2.10.0 (2009-10-26) i386-pc-mingw32 locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C [5] LC_TIME=English_United States.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached): [1] tools_2.10.0
On 11/30/09 11:48 AM, "Prof Brian Ripley" <ripley at stats.ox.ac.uk> wrote:
On Mon, 30 Nov 2009, Duncan Murdoch wrote:
On 30/11/2009 11:10 AM, Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed. Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed.
The source tarball (the .tar.gz file) would be fine on Windows, but most Windows users don't have the tools to install it. Since you don't have any compiled code it's probably not hard to put together a zip file equivalent to what R CMD INSTALL --binary would produce on Windows, but it's probably easier to use Uwe Ligge's package building service, available at http://129.217.207.166.
I agree with the conclusion (not least as it checks the package works on Windows), but in current R (2.10.0) you actually don't need any extra tools to install such a source package on Windows. It's all in the manuals ....
Duncan Murdoch
By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]] ------------------------------------------------------------------------
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
Bryan, there are several issues, but let's start with the most hopeful one:
install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, type = "source")
* installing *source* package 'ChemoSpec' ... ** R ** data ** inst ** preparing package for lazy loading Warning in library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'chemometrics' Error : package 'chemometrics' could not be loaded ERROR: lazy loading failed for package 'ChemoSpec' * removing 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' * restoring previous 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' As Brian pointed out you can now install source packages on Windows -- the only reason it fails is that you are missing dependent packages - "chemometrics" in this case. So install all dependencies first and you don't have to worry about binary packages - just use the source and all will be well. Then all the other issues are that you're trying to install the binary package the wrong way. In fact the easiest way to install is to simply unpack in the "library" directory. The problems you encountered come from the name mismatches - the name of the file does really matter and your "SU" / "UL" postfix that you added manually is breaking it. Please use the original names! Cheers, Simon
On Nov 30, 2009, at 14:30 , Bryan Hanson wrote:
Hello Folks, thanks for the suggestions. I did what Simon suggested. And I did what Ducan suggested, namely to use Uwe's building service. I think both of these produced the same set of files (though I don't know how to check rigorously). So I moved these two archives, along with my source archive, over to the Windows machine. I tried installing all of them via the GUI dialog boxes, and when that didn't work, I tried doing it from the command line. I'm pretty certain I'm now no longer in the Mac realm :-( and I'm pretty certain I'm missing something obvious. It looks like it can't find the file, but the paths and names look right to me. If one unstuffs these manually they look like they have the files. Any suggestions would be appreciated. The transcript from what I did is immediately below. Thanks, Bryan ************* Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University, Greencastle IN USA
getwd()
[1] "C:/Documents and Settings/Chemistry 1/Desktop"
list.files()
[1] "ChemoSpec_1.31-SU.zip" "ChemoSpec_1.31-UL.zip" "ChemoSpec_1.31.tar.gz" "outcome.txt" # the next few steps are attempts at manual installations:
install.packages(pkgs = "ChemoSpec_1.31-SU.zip", repos = NULL)
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, type =
"source") * installing *source* package 'ChemoSpec' ... ** R ** data ** inst ** preparing package for lazy loading Warning in library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'chemometrics' Error : package 'chemometrics' could not be loaded ERROR: lazy loading failed for package 'ChemoSpec' * removing 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' * restoring previous 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' Warning message: In install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, : installation of package 'ChemoSpec_1.31.tar.gz' had non-zero exit status # the next few are the attempts with the GUI dialog boxes
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-UL/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning messages: 1: In unzip(zipname, exdir = dest) : error 1 in extracting from zip file 2: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31.tar.gz/DESCRIPTION', probable reason 'No such file or directory'
sessionInfo()
R version 2.10.0 (2009-10-26) i386-pc-mingw32 locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C [5] LC_TIME=English_United States.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached): [1] tools_2.10.0
On 11/30/09 11:48 AM, "Prof Brian Ripley" <ripley at stats.ox.ac.uk> wrote:
On Mon, 30 Nov 2009, Duncan Murdoch wrote:
On 30/11/2009 11:10 AM, Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed. Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed.
The source tarball (the .tar.gz file) would be fine on Windows, but most Windows users don't have the tools to install it. Since you don't have any compiled code it's probably not hard to put together a zip file equivalent to what R CMD INSTALL --binary would produce on Windows, but it's probably easier to use Uwe Ligge's package building service, available at http://129.217.207.166.
I agree with the conclusion (not least as it checks the package works on Windows), but in current R (2.10.0) you actually don't need any extra tools to install such a source package on Windows. It's all in the manuals ....
Duncan Murdoch
By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]] ------------------------------------------------------------------------
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
You guys are great! I understand the nuances between platforms much better now (the Mac environment does a lot for you in the background, which I had not given much thought to). Success came with unpacking the .zip file into the library folder. Then I still had to install all the dependencies before my package would "activate", but I'm in business now. Thanks a bunch to each of you! Bryan
On 11/30/09 2:46 PM, "Simon Urbanek" <simon.urbanek at r-project.org> wrote:
Bryan, there are several issues, but let's start with the most hopeful one:
install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, type = "source")
* installing *source* package 'ChemoSpec' ... ** R ** data ** inst ** preparing package for lazy loading Warning in library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'chemometrics' Error : package 'chemometrics' could not be loaded ERROR: lazy loading failed for package 'ChemoSpec' * removing 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' * restoring previous 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' As Brian pointed out you can now install source packages on Windows -- the only reason it fails is that you are missing dependent packages - "chemometrics" in this case. So install all dependencies first and you don't have to worry about binary packages - just use the source and all will be well. Then all the other issues are that you're trying to install the binary package the wrong way. In fact the easiest way to install is to simply unpack in the "library" directory. The problems you encountered come from the name mismatches - the name of the file does really matter and your "SU" / "UL" postfix that you added manually is breaking it. Please use the original names! Cheers, Simon On Nov 30, 2009, at 14:30 , Bryan Hanson wrote:
Hello Folks, thanks for the suggestions. I did what Simon suggested. And I did what Ducan suggested, namely to use Uwe's building service. I think both of these produced the same set of files (though I don't know how to check rigorously). So I moved these two archives, along with my source archive, over to the Windows machine. I tried installing all of them via the GUI dialog boxes, and when that didn't work, I tried doing it from the command line. I'm pretty certain I'm now no longer in the Mac realm :-( and I'm pretty certain I'm missing something obvious. It looks like it can't find the file, but the paths and names look right to me. If one unstuffs these manually they look like they have the files. Any suggestions would be appreciated. The transcript from what I did is immediately below. Thanks, Bryan ************* Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University, Greencastle IN USA
getwd()
[1] "C:/Documents and Settings/Chemistry 1/Desktop"
list.files()
[1] "ChemoSpec_1.31-SU.zip" "ChemoSpec_1.31-UL.zip" "ChemoSpec_1.31.tar.gz" "outcome.txt" # the next few steps are attempts at manual installations:
install.packages(pkgs = "ChemoSpec_1.31-SU.zip", repos = NULL)
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, type =
"source") * installing *source* package 'ChemoSpec' ... ** R ** data ** inst ** preparing package for lazy loading Warning in library(pkg, character.only = TRUE, logical.return = TRUE, lib.loc = lib.loc) : there is no package called 'chemometrics' Error : package 'chemometrics' could not be loaded ERROR: lazy loading failed for package 'ChemoSpec' * removing 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' * restoring previous 'C:/PROGRA~1/R/R-210~1.0/library/ChemoSpec' Warning message: In install.packages(pkgs = "ChemoSpec_1.31.tar.gz", repos = NULL, : installation of package 'ChemoSpec_1.31.tar.gz' had non-zero exit status # the next few are the attempts with the GUI dialog boxes
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-SU/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning message: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31-UL/DESCRIPTION', probable reason 'No such file or directory'
utils:::menuInstallLocal()
Error in gzfile(file, "r") : cannot open the connection In addition: Warning messages: 1: In unzip(zipname, exdir = dest) : error 1 in extracting from zip file 2: In gzfile(file, "r") : cannot open compressed file 'ChemoSpec_1.31.tar.gz/DESCRIPTION', probable reason 'No such file or directory'
sessionInfo()
R version 2.10.0 (2009-10-26) i386-pc-mingw32 locale: [1] LC_COLLATE=English_United States.1252 LC_CTYPE=English_United States.1252 [3] LC_MONETARY=English_United States.1252 LC_NUMERIC=C [5] LC_TIME=English_United States.1252 attached base packages: [1] stats graphics grDevices utils datasets methods base loaded via a namespace (and not attached): [1] tools_2.10.0
On 11/30/09 11:48 AM, "Prof Brian Ripley" <ripley at stats.ox.ac.uk> wrote:
On Mon, 30 Nov 2009, Duncan Murdoch wrote:
On 30/11/2009 11:10 AM, Bryan Hanson wrote:
Hi R Mac Folks... A question about using Mac-built pkgs on the dreaded Windows machines. I have successfully checked, built and installed a package on my Mac. The package tarball is at github.com/bryanhanson/ChemoSpec if you care to try installing it. It was checked and built with defaults, so I did
R CMD check ChemoSpec R CMD build ChemoSpec
And everything was fine. I can install it and run it on my Mac no problem. I believe it is a ?source? package ? there is no compiled code, it?s all straightfoward R. Now, I want to test the installation on Windows (because students have to use it next semester). I?m having a variety of problems getting it upzipped and installed. Before I spend a lot of time trouble shooting the Windows aspect, and here?s my question, should this package as I built it and so forth work on Windows without modification? Or do I have to do the check/build with different options? Or do I have to do the entire process on a Windows machine (pray not!). I?ve looked around in the manuals and help pages a bit, there doesn?t seem to be any hint that a different process is needed.
The source tarball (the .tar.gz file) would be fine on Windows, but most Windows users don't have the tools to install it. Since you don't have any compiled code it's probably not hard to put together a zip file equivalent to what R CMD INSTALL --binary would produce on Windows, but it's probably easier to use Uwe Ligge's package building service, available at http://129.217.207.166.
I agree with the conclusion (not least as it checks the package works on Windows), but in current R (2.10.0) you actually don't need any extra tools to install such a source package on Windows. It's all in the manuals ....
Duncan Murdoch
By the way, I can manually get the entire package intact as text files to the proper Windows directory, but R claims it is not a valid installed package. Any of the RGui Windows means of installing the package give various problems. I?ll keep trouble shooting these but I wanted to double check that what I am trying to do should work. Thanks, Bryan ************** Bryan Hanson Acting Chair Professor of Chemistry & Biochemistry DePauw University 602 S. College Avenue Greencastle, IN 46135 PHONE 765-658-4602 FAX 765-658-6084 hanson at depauw.edu http://academic.depauw.edu/~hanson/deadpezsociety.html http://www.depauw.edu/acad/chemistry/ http://academic.depauw.edu/~hanson/UMP/index.html [[alternative HTML version deleted]] ------------------------------------------------------------------------
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac
_______________________________________________ R-SIG-Mac mailing list R-SIG-Mac at stat.math.ethz.ch https://stat.ethz.ch/mailman/listinfo/r-sig-mac