I have currently got the 64-bit version of Qt5 installed on a 64-bit Ubuntu machine but I'd also like to be able to do 32-bit builds on the same machine. On Windows I can simply select additional kits in the MaintenanceTool but on Linux this doesn't seem to be possible. Or am I missing something?
Three steps:
Get the 32 bit version from Qt or build from source with ./configure -platform linux-g++-32.
Make sure it gets installed to a different path than the 64-bit version. Can be achieved with something like ./configure -platform linux-g++-32 -prefix /opt/qt32 or the path that feels comfortable to you.
Fix compiler settings along these lines for building the application
As per my information, Binaries (or offline/online installers) for installing Qt (>= 5.6) on 32-bit linux are not provided. For example, If I want to use Qt 5.9.0 on 32-bit linux (debian, specifically), the only option is to download Qt source and build it. I have successfully built Qt 5.9.0 on my 32-bit linux machine. But I have to do this on all my development PC's. Instead if I can build it once and create an installer somehow, that will save a lot of time. Is there a way to create offline installer (e.g. *.run files) for installing Qt 5.9.0 on 32-bit Linux (e.g. debian)?
As stated by #Velkan, Qt is available on 32 bit Debian from the official repos.
But, indeed, there is no 32-bit Linux installer from the Qt project.
Regarding copying a Qt install from a Debian PC to another, you do not need to create an installer. Just copy/tar/zip the directory in which Qt is installed. The only requirement is that you must keep the installation in the same absolute path as the Qt Core library has the path hardcoded in it and it is use for plugin detection.
NB1: It is possible to patch Qt Core library to change the hardcoded path and even make it relative. It is not officially documented but you can take a look in the sources of tools like windeployqt or the online installer
NB2: Do not use Qt 5.9.0, prefer 5.9.2. Both are forward and backward binary compatible, but the latter received bug fixes.
I have been using RcppArmadillo to build a package, but I've run into a small problem. I have a perfectly working package for 32-bit R, but I would like for it to run on 64-bit. When I started working on the package I was using 32-bit R (without realizing it), but when I shifted to 64-bit R I stopped being able to build my package. I get the error saying "LoadLibrary failure: %1 is not a valid Win32 application."
I realize that the issue is that the .dll file is 32-bit so 64-bit R can't work with it. What I don't understand is how to recreate the .dll so that it is 64-bit. In 64-bit R when I run devtools::find_rtools() I get TRUE, so I suspect that I have to add something to the makevars files to make this work.
So my question in one sentence is: how do you get a working 32bit package with RcppArmadillo to work with 64bit on Windows? Any guidance would be greatly appreciated.
All CRAN packages build on 32 and 64 bit Windows.
All CRAN submissions, including my different Rcpp* packages do (apart from those requiring Unix, or C++11, or ...).
So if I were you, and had (temporary) troubles setting up my main box, I would still use the excellent win-builder which will generate 32 and 64 bit binaries for you.
Aha, it was the makevars files.
I had to add $(shell "${R_HOME}/bin${R_ARCH_BIN}/Rscript.exe" -e "Rcpp:::LdFlags()")
I created a development environment with Fedora 18 and Qt 5
When I compile my app it will be 64 bit, and depend on Qt 5. After some research it seems that will be a problem since Centos 6.x is still dependong on Qt 4.6, and Centos 5.x is dependant on earlier Qt, and I'm guessin Ubuntu has its own package limitations.
Is there a "safe" version of Qt I can compile against to ensure it runs everywhere? If not, how can I expect customers to run my program?
If you want to target Linux, then I would recommend using Qt 4.8 (available on most of the Linux Distros), forget about Qt 5 for a year or so. Hardly any Linux distro other than Ubuntu 13.04 and above comes with Qt 5.
Secondly, as far as compiling is considered, if you wish to target Linux distros other than Ubuntu and Fedora than never use Ubuntu or Fedora for compilation. These are bleeding edge Linux distros which include new libraries without much testing. You will not only face the problem of old Qt versions in some Linux distros as you will face the bigger problem of glibc (C library). Make sure the Linux distro you use for compilation has a minimum possible glibc version for Qt 4.8, otherwise even if any Linux distro has Qt 4.8 installed, but has a lower version of glibc, you app still won't run. You can check out distrowatch.com to find out glibc and Qt versions for any Linux Distro
Example:
I compiled my app on Ubuntu 12.04, Qt 4.8.1 with glibc v 2.16 (perhaps). I got a bug report than application does not run on CrunchBang which also has Qt 4.8.1 available but since it is a Debian unstable Linux Distro it uses glibc v2.13. Since that day I always compile my application on CrunchBang (very lighweight, you can Virtual Box for this) and hardly some one complains now.
As far as CentOs is considered, you will need to recompile your code on CentOS.
I'm trying to install rJava on a computer with Win 7 64 bit. When I run
install.packages("rJava")
everything seems to be fine:
Installing package(s) into ‘C:/Users/djq/Documents/R/win-library/2.13’
(as ‘lib’ is unspecified)
trying URL 'http://software.rc.fas.harvard.edu/mirrors/R/bin/windows/contrib/2.13/rJava_0.9-1.zip'
Content type 'application/zip' length 654936 bytes (639 Kb)
opened URL
downloaded 639 Kb
package 'rJava' successfully unpacked and MD5 sums checked
The downloaded packages are in
C:\TEMP\RtmpW2eAUw\downloaded_packages
but when I tried to load the library library(rJava) I get the following error:
Error : .onLoad failed in loadNamespace() for 'rJava', details:
call: inDL(x, as.logical(local), as.logical(now), ...)
error: unable to load shared object 'C:/Users/me/Documents/R/win-library/2.13/rJava/libs/x64/rJava.dll':
LoadLibrary failure: %1 is not a valid Win32 application.
Error: package/namespace load failed for 'rJava'
I have installed the Java JRE (trying both 32 and 64 bit versions) and the JDK (using both 32 and 64 bit). My R version is R version 2.13.1 (2011-07-08) and I'm using RStudio.
If I try using 32 bit R, I can load rJava fine. Is the problem just that rJava does not run in R 64 bit? How can I tell if a package is just for 32 instead of 64?
edit: just posted R.version():
> R.Version()
$platform
[1] "x86_64-pc-mingw32"
$arch
[1] "x86_64"
$os
[1] "mingw32"
$system
[1] "x86_64, mingw32"
$status
[1] ""
$major
[1] "2"
$minor
[1] "13.1"
$year
[1] "2011"
$month
[1] "07"
$day
[1] "08"
$`svn rev`
[1] "56322"
$language
[1] "R"
$version.string
[1] "R version 2.13.1 (2011-07-08)"
Update (July 2018):
The latest CRAN version of rJava will find the jvm.dll automatically, without manually setting the PATH or JAVA_HOME. However note that:
To use rJava in 32-bit R, you need Java for Windows x86
To use rJava in 64-bit R, you need Java for Windows x64
To build or check R packages with multi-arch (the default) you need to install both Java For Windows x64 as well as Java for Windows x86. On Win 64, the former installs in C:\Program files\Java\ and the latter in C:\Program Files (x86)\Java\ so they do not conflict.
As of Java version 9, support for x86 (win32) has been discontinued. Hence the latest working multi-arch setup is to install both jdk-8u172-windows-i586.exe and jdk-8u172-windows-x64.exe and then the binary package from CRAN:
install.packages("rJava")
The binary package from CRAN should pick up on the jvm by itself. Experts only: to build rJava from source, you need the --merge-multiarch flag:
install.packages('rJava', type = 'source', INSTALL_opts='--merge-multiarch')
Old anwser:
(Note: many of folks in other answers/comments have said to remove JAVA_HOME, so consider that. I have not revisited this issue recently to know if all the steps below are still necessary.)
Here is some quick advice on how to get up and running with R + rJava on Windows 7 64bit. There are several possibilities, but most have fatal flaws. Here is what worked for me:
Add jvm.dll to your PATH
rJava, the R<->Java bridge, will need jvm.dll, but R will have trouble finding that DLL. It resides in a folder like
C:\Program Files\Java\jdk1.6.0_25\jre\bin\server
or
C:\Program Files\Java\jre6\jre\bin\client
Wherever yours is, add that directory to your windows PATH variable. (Windows -> "Path" -> "Edit environment variables to for your account" -> PATH -> edit the value.)
You may already have Java on your PATH. If so you should find the client/server directory in the same Java "home" dir as the one already on your PATH.
To be safe, make sure your architectures match.If you have Java in Program Files, it is 64-bit, so you ought to run R64. If you have Java in Program Files (x86), that's 32-bit, so you use plain 32-bit R.
Re-launch R from the Windows Menu
If R is running, quit.
From the Start Menu , Start R / RGUI, RStudio. This is very important, to make R pick up your PATH changes.
Install rJava 0.9.2.
Earlier versions do not work! Mirrors are not up-to-date, so go to the source at www.rforge.net: http://www.rforge.net/rJava/files/. Note the advice there
“Please use
`install.packages('rJava',,'http://www.rforge.net/')`
to install.”
That is almost correct. This actually works:
install.packages('rJava', .libPaths()[1], 'http://www.rforge.net/')
Watch the punctuation! The mysterious “.libPaths()[1],” just tells R to install the package in the primary library directory. For some reason, leaving the value blank doesn’t work, even though it should default.
Getting rJava to work depends heavily on your computers configuration:
You have to use the same 32bit or 64bit version for both: R and JDK/JRE. A mixture of this will never work (at least for me).
If you use 64bit version make sure, that you do not set JAVA_HOME as a enviorment variable. If this variable is set, rJava will not work for whatever reason (at least for me). You can check easily within R is JAVA_HOME is set with
Sys.getenv("JAVA_HOME")
If you need to have JAVA_HOME set (e.g. you need it for maven or something else), you could deactivate it within your R-session with the following code before loading rJava:
if (Sys.getenv("JAVA_HOME")!="")
Sys.setenv(JAVA_HOME="")
library(rJava)
This should do the trick in most cases. Furthermore this will fix issue Using the rJava package on Win7 64 bit with R, too. I borrowed the idea of unsetting the enviorment variable from R: rJava package install failing.
For me, setting JAVA_HOME did the trick (instead of unsetting, as in another answer given here). Either in Windows:
set JAVA_HOME="C:\Program Files\Java\jre7\"
Or inside R:
Sys.setenv(JAVA_HOME="C:\\Program Files\\Java\\jre7\\")
But what's probably the best solution (since rJava 0.9-4) is overriding within R the Windows JAVA_HOME setting altogether:
options(java.home="C:\\Program Files\\Java\\jre7\\")
library(rJava)
The last question has an easy answer:
> .Machine$sizeof.pointer
[1] 8
Meaning I am running R64. If I were running 32 bit R it would return 4. Just because you are running a 64 bit OS does not mean you will be running 64 bit R, and from the error message it appears you are not.
EDIT: If the package has binaries, then they are in separate directories. The specifics will depend on the OS. Notice that your LoadLibrary error occurred when it attempted to find the dll in ...rJava/libs/x64/... On my MacOS system the ...rJava/libs/...` folder has 3 subdirectories: i386, ppc, and x86_64. (The ppc files are obviously useless baggage.)
Sorry for necro.
I have too run into the same issue and found out that rJava expects JAVA_HOME to point to JRE. If you have JDK installed, most probably your JAVA_HOME points to JDK. My quick solution:
Sys.setenv(JAVA_HOME=paste(Sys.getenv("JAVA_HOME"), "jre", sep="\\"))
I had a related problem with rJava. It would load but a package that depends on it, would not load.
Users may waste a lot of time with jvm.dll and PATH and JAVA_HOME when the real fix is to force the installer to just forget about i386. Use option for install.packages. (this also works when drat library is used. (credit goes to Dason)
install.packages("SqlRender",INSTALL_opts="--no-multiarch")
Also, you can modify just your user path with a win command like this:
setx PATH "C:\Program Files\Java\jre1.8.0_102\bin\server;%PATH%"
I had some trouble determining the Java package that was installed when I ran into this problem, since the previous answers didn't exactly work for me. To sort it out, I typed:
Sys.setenv(JAVA_HOME="C:/Program Files/Java/
and then hit tab and the two suggested directories were "jre1.8.0_31/" and "jre7/"
Jre7 didn't solve my problem, but jre1.8.0_31/ did. Final answer was running (before library(rJava)):
Sys.setenv(JAVA_HOME="C:/Program Files/Java/jre1.8.0_31/")
I'm using 64-bit Windows 8.1 Hope this helps someone else.
Update:
Check your version to determine what X should be (mine has changed several times since this post):
Sys.setenv(JAVA_HOME="C:/Program Files/Java/jre1.8.0_x/")
Download Java from https://java.com/en/download/windows-64bit.jsp for 64-bit windows\Install it
Download Java development kit from https://www.oracle.com/technetwork/java/javase/downloads/jdk8-downloads-2133151.html for 64-bit windows\install it
Then right click on “This PC” icon in desktop\Properties\Advanced system settings\Advanced\Environment Variables\Under System variables select Path\Click Edit\Click on New\Copy and paste paths “C:\Program Files\Java\jdk1.8.0_201\bin” and “C:\Program Files\Java\jre1.8.0_201\bin” (without quote) \OK\OK\OK
Note: jdk1.8.0_201 and jre1.8.0_201 will be changed depending on the version of Java development kit and Java
In Environment Variables window go to User variables for User\Click on New\Put Variable name as “JAVA_HOME” and Variable value as “C:\Program Files\Java\jdk1.8.0_201\bin”\Press OK
To check the installation, open CMD\Type javac\Press Enter and Type java\press enter It will show
In RStudio run
Sys.setenv(JAVA_HOME="C:\\Program Files\\Java\\jdk1.8.0_201")
Note: jdk1.8.0_201 will be changed depending on the version of Java development kit
Now you can install and load rJava package without any problem.
I need to have a 32 bit JRE available for my browser, but 64 bit JRE for R and rJava. The 32 bit JRE is installed in the default location. After some experimentation, I found that I only needed one of misterbee's suggestions to get rJava (version 0.9-6) working for me. All I did was add the path to my 64 bit java installation:
C:\apps\Java\jre\bin\server\jvm.dll
to the top of my path environment variable (your path will likely be different) and remove my JAVA_HOME as user2161065 suggested. I put this just ahead of the entry
C:\ProgramData\Oracle\Java\javapath
which the Oracle installer inserts at the top of the path and points to some symlinks to the 32 bit JRE. By adding the entry to 64 bit jvm.dll, looks like rJava could find what it needs.
I think this is an update. I was unable to install rJava (on Windows) until I installed the JDK, as per
Javac is not found
and
javac not working in windows command prompt. The message I was getting was
'javac' is not recognized as an internal or external command, operable program or batch file.
The JDK includes the JRE, and according to
https://cran.r-project.org/web/packages/rJava/index.html
the current version (0.9-7 published 2015-Jul-29)
of rJava
SystemRequirements: Java JDK 1.2 or higher (for JRI/REngine JDK 1.4 or higher), GNU make
So there you are: if rJava won't install because it can't find javac, and you have the JRE installed, then try the JDK. Also, make sure that JAVA_HOME points to the JDK and not the JRE.
I solved the issue by uninstalling apparently redundant Java software from my windows 7 x64 machine. I achieved this by first uninstalling all Java applications and then installing a fresh Java version. (Later I pointed R 3.4.3 x86_64-w64-mingw32 to the Java path, just to mention though I don't think this was the real issue.) Today only Java 8 Update 161 (64-bit) 8.0.1610.12 was left then. After this, install.packages("rJava"); library(rJava) did work perfectly.
This is a follow-up to Update (July 2018). I am on 64 bit Windows 10 but am set up to build r packages from source for both 32 and 64 bit with Rtools. My 64 bit jdk is jdk-11.0.2. When I can, I do everything in RStudio. As of March 2019, rjava is tested with <=jdk11, see github issue #157.
Install jdks to their default location per Update (July 2018) by #Jeroen.
In R studio, set JAVA_HOME to the 64 bit jdk
Sys.setenv(JAVA_HOME="C:/Program Files/Java/jdk-11.0.2")
Optionally check your environmental variable
Sys.getenv("JAVA_HOME")
Install the package per the github page recommendation
install.packages("rJava",,"http://rforge.net")
FYI, the rstudio scripting console doesn't like the double commas... but it works!
So many answers and yet, nothing worked for me - very disappointing and frustrating.
Every now and then I will run into this issue and waste time on Google.
So here is what DID work for me, I am on Windows 10, with everything fully updated.
I went here to download JDK and JRE 8u301, the latest version at this writing.
Oracle Latest Java Download Link
Note I had to create an account or log into my account - takes no time and is free, nothing to worry about.
I installed JRE and then JDK, using the exe just downloaded in the previous step.
In Rstudio (or in R), I ran Sys.setenv(JAVA_HOME='C:\\Program Files\\Java\\jdk1.8.0_301').
Running install.packages("rJava") led to the desired DONE (rJava). (of course, I assume you also have Rtools installed, goes without saying...)
Pretty simple and easy imho.
With the right steps, took me 3 min.