如何从自定义存储库强制安装 R 包？

Question

There's a dependency of dependency which fails to compile when pulled from CRAN at the moment, so the maintainer provided an alternative Cannot compile RcppArmadillo .

I've tried to install RcppArmadillo from that path like so:

if(!require("RcppArmadillo")) install.packages("RcppArmadillo", repos="https://rcppcore.github.io/drat", lib = LIB_PATH)
library(RcppArmadillo)

if(!require("sqjin/scAI")) devtools::install_github("sqjin/scAI", lib = LIB_PATH)
library(scAI, lib.loc = LIB_PATH)

However, I can see in the logs that RcppArmadillo is still being pulled from CRAN .

How do I force the installation from custom repo?

Answer 1

By setting your options("repos") (named vector) argument in one of your (user or system) startup files you make the alternate location known to R. After that R will pick the highest (== "newest") version (number).

Here is a block from help(Startup) showing one way to do it:

 ## Example of Rprofile.site
 local({
   # add MASS to the default packages, set a CRAN mirror
   old <- getOption("defaultPackages"); r <- getOption("repos")
   r["CRAN"] <- "http://my.local.cran"
   options(defaultPackages = c(old, "MASS"), repos = r)
   ## (for Unix terminal users) set the width from COLUMNS if set
   cols <- Sys.getenv("COLUMNS")
   if(nzchar(cols)) options(width = as.integer(cols))
   # interactive sessions get a fortune cookie (needs fortunes package)
   if (interactive())
     fortunes::fortune()
 })

where you could just use

   r["CRAN"] <- "https://cloud.r-project.org"
   r["Rcpp"] <- "https://RcppCore.github.io/drat"

As the RcppCore repo is a drat repo you could also consult the drat documentation which has other examples and helpers.