3 Packages

A package is an archive of files that conforms to a certain format and structure and that provides extra functionality, usually extending R in a particular direction. The R community has produced many high-quality R packages for performing specific tasks,

Any package is in one of three states:

Installed and loaded. A package that is loaded is directly available to your R session. Find out which packages are loaded using sessionInfo().
Installed but not loaded. A package that is installed is available for loading but its contents are not available until it is loaded. The function help.start() gives details of the packages that are installed on your computer.
Not installed. These packages cannot be loaded. If a package is not installed then the library() function produces an error. If the install status is uncertain at the time of calling library (for example if you are writing a function that requires the package), then use the require() function, which returns FALSE if the package is not installed, rather than an error.

Installing all available packages would be a waste of space and time, as you would never use most of them. Similarly, loading all installed packages every time you start R would take some time, so by default R only loads the base packages when it starts and requires the user to load any others as and when they are needed.

Packages are divided into three groups:

Base. Base packages are installed along with R, and their objects are always available.
Recommended. Recommended packages are installed along with R but must be loaded before they can be used.
Other. Other packages are not installed by default, and must be installed separately.

The command to find out what packages are available for loading is installed.packages. The output of the function is quite verbose, but we only need the first column:

Example:

installed.packages()[1:5, 1] # Returns only the first five packages

    anytime     askpass   backports   base64enc          BH 
  "anytime"   "askpass" "backports" "base64enc"        "BH"

All the packages that are available at a repository, and whose requirements are matched by the currently running version of R, can be listed using the command available.packages(). A package that is available in the repository but has not yet been installed may be installed using the install.packages() function. If we include the argument dependencies = TRUE, then the function will also install packages that are necessary to run the package or packages of interest; such packages are called dependencies.

Most of the packages frequently get updates. This includes new features, improvements, and error fixing. The packageVersion function returns the version details of the input package:

Example:

packageVersion("tidyverse")

[1] '2.0.0'

The status of the packages that are installed can be compared with the repository using the old.packages() function, and easily updated using the update.packages() function.

R can be easily updated with installr package. To update R on MacOS, you need to use updateR” package instead.

Example:

install.packages("installr")
library(installr)
updateR()