Wrong version of Java is loaded during installation - r

Question

I have a weird issue while trying to install packages which depend on Java/rJava on my 64-bit Windows 10:
> devtools::install_github("ropensci/tabulizer", INSTALL_opts = "--no-multiarch")
Downloading GitHub repo ropensci/tabulizer#master
from URL https://api.github.com/repos/ropensci/tabulizer/zipball/master
Installing tabulizer
Downloading GitHub repo ropensci/tabulizerjars#master
from URL https://api.github.com/repos/ropensci/tabulizerjars/zipball/master
Installing tabulizerjars
"C:/PROGRA~1/R/R-34~1.2/bin/x64/R" --no-site-file --no-environ --no-save --no-restore --quiet CMD INSTALL \
"C:/Users/binis/AppData/Local/Temp/Rtmpiw2khr/devtools4d7472ae39f9/ropensci-tabulizerjars-60b7cec" \
--library="C:/Users/binis/Documents/R/win-library/3.4" --install-tests
* installing *source* package 'tabulizerjars' ...
** R
** inst
** preparing package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded
*** arch - i386
Error: package or namespace load failed for 'tabulizerjars':
.onLoad failed in loadNamespace() for 'rJava', details:
call: inDL(x, as.logical(local), as.logical(now), ...)
error: unable to load shared object 'C:/Users/binis/Documents/R/win-library/3.4/rJava/libs/i386/rJava.dll':
LoadLibrary failure: %1 is not a valid Win32 application.
Error: loading failed
Execution halted
*** arch - x64
ERROR: loading failed for 'i386'
* removing 'C:/Users/binis/Documents/R/win-library/3.4/tabulizerjars'
Installation failed: Command failed (1)
"C:/PROGRA~1/R/R-34~1.2/bin/x64/R" --no-site-file --no-environ --no-save --no-restore --quiet CMD INSTALL \
"C:/Users/binis/AppData/Local/Temp/Rtmpiw2khr/devtools4d744e067cae/ropensci-tabulizer-3aad163" \
--library="C:/Users/binis/Documents/R/win-library/3.4" --install-tests
ERROR: dependency 'tabulizerjars' is not available for package 'tabulizer'
* removing 'C:/Users/binis/Documents/R/win-library/3.4/tabulizer'
Installation failed: Command failed (1)
Now I'm aware that this looks like I messed up with Java versions or path information. But that's not the case (as far as I'm aware):
> library(rJava)
> .jinit()
> .jcall("java/lang/System", "S", "getProperty", "java.version")
[1] "1.8.0_151"
> .jcall("java/lang/System", "S", "getProperty", "java.home")
[1] "C:\\Program Files\\Java\\jre1.8.0_151"
> .jcall("java/lang/System", "S", "getProperty", "sun.arch.data.model")
[1] "64"
> .jcall("java/lang/System", "S", "getProperty", "os.arch")
[1] "amd64"
This shows that os architecture and Java architecture match (R is 64-bit as well), that java.home is set correctly and also that rJava works nicely on my machine.
Now I'm not sure I understand the error message correctly but it seems that R is trying to load the 32-bit version of rJava during installation which fails. I tested this by changing java home to a 32-bit installation on my machine:
> options(java.home="C:/Program Files (x86)/Java/jre1.8.0_151") #32-bit version
> library(rJava)
Error: package or namespace load failed for ‘rJava’:
.onLoad failed in loadNamespace() for 'rJava', details:
call: inDL(x, as.logical(local), as.logical(now), ...)
error: unable to load shared object 'C:/Users/binis/Documents/R/win-library/3.4/rJava/libs/x64/rJava.dll':
LoadLibrary failure: %1 is not a valid Win32 application.
The error is the same which supports my assumption. However, I do not understand why R would try to use a different Java version during installtion or how I could stop that. I already tested if INSTALL_opts = c("--no-multiarch", "--no-test-load"), type="source" will do anything but the error remains the same...

Answer 1

After I spent some more time on google/stackoverflow, I found that the syntax of install_github is apparently slightly different from install.packages. Instead of ´INSTALL_opts´, options are invoked with ´args´. The follwing command, therefore, installs the example package without "complaints":
devtools::install_github("ropensci/tabulizer",
args = c("--no-multiarch", "--no-test-load"))
I leave this here in case others run into the same issue.