[ESS] Problem starting specific R version

[Henric Winell]

Hi Alex,

On 2018-07-01 14:03, Alex Branham wrote:

> Thanks for the report.

Thanks for the swift reply and interest in this matter.

> Can you let me know the exact name of those two programs, including the 
> full path (e.g. C:\\Program Files\R\R-3.5\R.exe or whatever)? And the 

Under 64-bit Windows, which is a bi-arch system, the executables can be 
found in the bin\i386 and bin\x64 directories under the version-specific 
R installation directory.  The correct paths are recorded in the 
ess-rterm-version-paths variable:

ess-rterm-version-paths is a variable defined in ‘ess-custom.el’.
Its value is
("C:/Program Files/R/R-3.6/bin/i386/Rterm.exe" "C:/Program 
Files/R/R-3.5/bin/i386/Rterm.exe" "C:/Program 
Files/R/R-3.4/bin/i386/Rterm.exe" "C:/Program 
Files/R/R-3.3/bin/i386/Rterm.exe" "C:/Program 
Files/R/R-3.6/bin/x64/Rterm.exe" "C:/Program 
Files/R/R-3.5/bin/x64/Rterm.exe" "C:/Program 
Files/R/R-3.4/bin/x64/Rterm.exe" "C:/Program 
Files/R/R-3.3/bin/x64/Rterm.exe")


Issuing 'M-x R-[TAB]' results in

Click on a completion to select it.
In this buffer, type RET to select the completion near point.

Possible completions are:
R-3.3-32bit 	R-3.3-64bit 	R-3.4-32bit
R-3.4-64bit 	R-3.5-32bit 	R-3.5-64bit
R-3.6-32bit 	R-3.6-64bit 	R-fix-T-F
R-initialize-on-start 	R-mode 	R-newest
R-transcript-mode

where the first couple of entries should correspond to the executables 
above.

I've set inferior-ess-r-program-name to "C:/Program 
Files/R/R-3.6/bin/x64/Rterm.exe" so that 'M-x R' starts the 64-bit devel 
version of R.

> value of exec-path in emacs (C-h v exec-path)

exec-path is a variable defined in ‘C source code’.
Its value is
("c:/Program Files/R/Rtools/bin" "C:/Program 
Files/R/Rtools/mingw_64/bin" "C:/Program Files/R/R-3.6/bin" 
"C:/MiKTeX/miktex/bin/x64/" "C:/Program Files/GNU Emacs/bin" "C:/Program 
Files/doxygen/bin" "C:/Program Files (x86)/Graphviz/bin" 
"C:/ProgramData/Oracle/Java/javapath" "C:/Program Files (x86)/Common 
Files/Oracle/Java/javapath" "C:/Windows/system32" "C:/Windows" 
"C:/Windows/System32/Wbem" "C:/Windows/System32/WindowsPowerShell/v1.0/" 
"C:/Program Files (x86)/Skype/Phone/" "C:/Program Files (x86)/KiTTy" 
"C:/Program Files/qpdf/bin" "C:/nuweb" "C:/Program Files 
(x86)/Intel/Intel(R) Management Engine Components/DAL" "C:/Program 
Files/Intel/Intel(R) Management Engine Components/DAL" "C:/Program Files 
(x86)/Intel/Intel(R) Management Engine Components/iCLS/" "C:/Program 
Files/Intel/Intel(R) Management Engine Components/iCLS/" "C:/Program 
Files (x86)/Intel/Intel(R) Management Engine Components/IPT" "C:/Program 
Files/Intel/Intel(R) Management Engine Components/IPT" "C:/Program 
Files/Git/cmd" "C:/Program Files (x86)/Pandoc/" "C:/Program 
Files/TortoiseSVN/bin" "C:/Program Files/Intel/WiFi/bin/" "C:/Program 
Files/Common Files/Intel/WirelessCommon/" 
"C:/Users/henwin/AppData/Local/Microsoft/WindowsApps" 
"C:/MiKTeX/miktex/bin/x64/" "C:/Program Files/Intel/WiFi/bin/" 
"C:/Program Files/Common Files/Intel/WirelessCommon/" "c:/Program 
Files/GNU Emacs/libexec/emacs/25.3/x86_64-w64-mingw32")


Kind regards,
Henric



> 
> Thanks,
> Alex
> 
> 
> On Sun, Jul 1, 2018, 6:53 AM Henric Winell via ESS-help 
> <ess-help using r-project.org <mailto:ess-help using r-project.org>> wrote:
> 
>     Hi,
> 
>     I'd like to report the following, which I suspect is a bug:
> 
>     On my Windows 10 system using Emacs 25.3.1 I've got several R versions.
>     For example, I've got R-3.5.0 and R-devel installed under C:\Program
>     Files\R\R-3.5 and C:\Program Files\R\R-3.6, respectively.
> 
>     Up until quite recently, I could start a specific version, e.g., the
>     64-bit version of R-3.5.0 by issuing 'M-x R-3.5-64bit'.  This is now
>     broken and results in "Searching for program: No such file or
>     directory,
>     R-3.5-64bit".
> 
>     I've got a couple of "older" ESS versions lying around and found
>     that it
>        still works is ess-20180514.721, but is broken in ess-20180522.213
>     (and still broken in ess-20180701.100).  I checked the commit history
>     and suspect that the 'Simplify how R-X.Y.Z style functions are created'
>     change that got merged in on May 21 may have something to do with this
>     problem.
> 
> 
>     Kind regards,
>     Henric Winell
> 
>     ______________________________________________
>     ESS-help using r-project.org <mailto:ESS-help using r-project.org> mailing list
>     https://stat.ethz.ch/mailman/listinfo/ess-help
>