rpackageversionmicrosoft-rdbplyr

R Open does not show/update to latest version of MRAN package

I want to update an R package using Microsoft R Open via Microsoft R Server. I can see in MRAN package repo that the package version is 1.2.0 but when I try an update in RStudio it says all packages up to date although respective package is still on version 1.1.0 (dbplyr in my case). Why is R not updating the package? RStudio points to C:\Program Files\Microsoft\ML Server\R_SERVER.

sessionInfo()
 #>R version 3.4.1 (2017-06-30)
 #>Platform: x86_64-w64-mingw32/x64 (64-bit)
 #>Running under: Windows 7 x64 (build 7601) Service Pack 1

 #Matrix products: default

 #>locale:
 #>[1] LC_COLLATE=English_United States.1252  LC_CTYPE=English_United States.1252    LC_MONETARY=English_United States.1252
 #>[4] LC_NUMERIC=C                           LC_TIME=English_United States.1252    

 #>attached base packages:
 #>[1] stats     graphics  grDevices utils     datasets  methods   base     

 #>other attached packages:
 #>[1] RevoUtilsMath_10.0.0 RevoUtils_10.0.5     RevoMods_11.0.0      MicrosoftML_1.5.0    mrsdeploy_1.1.2      RevoScaleR_9.2.1    
 #>[7] lattice_0.20-35      rpart_4.1-11        

 #>loaded via a namespace (and not attached):
  #>[1] codetools_0.2-15       CompatibilityAPI_1.1.0 foreach_1.4.4          grid_3.4.1             R6_2.2.0              
  #>[6] jsonlite_1.4           curl_2.6               iterators_1.0.8        tools_3.4.1            yaml_2.1.14           
 #>[11] compiler_3.4.1         mrupdate_1.0.1

packageVersion("dbplyr")
 #>‘1.1.0’

RStudio IDE has as mirror Global (CDN) - RStudio but I guess this is overruled by MRAN?

enter image description here

Solution

  • So you are running Microsoft R Server, and that has an R version of 3.4.1. I believe what happens is that when you try to upgrade using MRAN, the upgrade takes into consideration which version of R you're on - and (I believe again) the latest version of dbplyr for R 3.4.1 was 1.1.0. You can read more about snapshots and checkpoints here.

    What you can do if you want is to install the package straight from the CRAN repo, using code something like so:

    install.packages("dbplyr", repos = "https://cloud.r-project.org/", dependencies=TRUE, lib = pkg.dir)

    Hope this helps!

    Niels