README.md

In arraytools/rtoy: rtoy

Note: the yaml files may need to be updated once a new version of R comes out. See the latest files on github.com/r-lib/actions. Currently I am using R '3.6'.

A toy (minimal) R package example

The package was created by RStudio IDE (File > New Project > New Directory > R Package).

Test a local package via Docker

You can quickly test the installation of the package using Docker.

$ git clone https://github.com/arraytools/rtoy.git
$ cd rtoys
$ docker run -it --rm -v $(pwd):/rtoy r-base

> system("ls")
bin   dev  home  lib64  mnt  proc  rtoy  sbin  sys  usr
boot  etc  lib   media  opt  root  run   srv   tmp  var

> system("ls rtoy")
DESCRIPTION  man  NAMESPACE  R  rtoy.Rproj
> system("R CMD INSTALL /rtoy")
* installing to library ‘/usr/local/lib/R/site-library’
* installing *source* package ‘rtoy’ ...
** using staged installation
** R
** byte-compile and prepare package for lazy loading
** help
*** installing help indices
** building package indices
** testing if installed package can be loaded from temporary location
** testing if installed package can be loaded from final location
** testing if installed package keeps a record of temporary installation path
* DONE (rtoy)
> library(rtoy)
> hello()
[1] "Hello, world!"
> q()

Add a dependency in our local package

To add a dependency, try to add a new line Imports: Rcpp (needs compilation) or Imports: packrat (no compilation) after Description in DESCRIPTION file. Afterwards, test installing the package by

> install.packages("remotes")
> remotes::install_local("rtoy")

Packrat

Still under a Docker environment.

> install.packages("packrat")
> packrat::init("/home/docker") # root directory won't work?
# For local packages, we can still use remotes::install_local()
# But when we run packrat::snapshot(), it will show an error
# Error: unable to retrieve package records for ...

# packrat issues errors when install local packages that have
# a dependency on other packages on CRAN. See
# https://stackoverflow.com/q/28098785
# Possible solutions are
# 1. manually install dependency packages
# 2. use remotes::install_local() to install dependency
#    package and then use packrat::install_local() again 
#    to install the local package.
> install.packages("remotes")
> remotes::install_local("/rtoy")  # notice the absol directory
> packrat::set_opts(local.repos = "/")
> packrat::install_local("rtoy")
> packrat::snapshot()  # Rcpp, remotes and rtoy are added
> packrat::bundle(include.bundles = FALSE)
The packrat project has been bundled at:
- "/home/docker/packrat/bundles/docker-2019-06-25.tar.gz"

> untar("/home/docker/packrat/bundles/docker-2019-06-25.tar.gz", list = TRUE)
 [1] "docker/.Rprofile"                               
 [2] "docker/packrat/"                                
 [3] "docker/packrat/init.R"                          
 [4] "docker/packrat/packrat.lock"                    
 [5] "docker/packrat/packrat.opts"                    
 [6] "docker/packrat/src/"                            
 [7] "docker/packrat/src/packrat/"                    
 [8] "docker/packrat/src/packrat/packrat_0.5.0.tar.gz"
 [9] "docker/packrat/src/Rcpp/"                       
[10] "docker/packrat/src/Rcpp/Rcpp_1.0.1.tar.gz"      
[11] "docker/packrat/src/remotes/"                    
[12] "docker/packrat/src/remotes/remotes_2.1.0.tar.gz"
[13] "docker/packrat/src/rtoy/"                       
[14] "docker/packrat/src/rtoy/rtoy_0.1.0.tar.gz"
> getwd()
[1] "/home/docker"
> dir()
[1] "packrat"

> args(packrat::bundle)
function (project = NULL, file = NULL, include.src = TRUE, include.lib = FALSE, 
    include.bundles = TRUE, include.vcs.history = FALSE, overwrite = FALSE, 
    omit.cran.src = FALSE, ...) 
NULL

PS.

1. no need to run $ sudo rm .Rprofile; sudo rm -rf packrat unless we mount some local directory to /home/docker.
2. We can use packrat::unbundle() or the tar command to extract the tarball on a new environment and use packrat::restore() to restore all R packages.

3. packrat::bundle() will zip most of things under the current directory.

   3.1 If we just need to reproduce the R enviroment, we use a new directory; such as packrat.init("project"). This will create a new directory "project" and packrat subdirectory will be under it.

   3.2 If we want to zip everything under the current working directory, we can use packrat.init(). The packrat subdirectory will be created under the current directory.

arraytools/rtoy documentation built on April 22, 2020, 5:18 p.m.