tryCatch() Vs if else

Question

I have recently started exploring R programming and have a very trivial doubt. I want to write a function which would try to load the required packages and install the same in case its not installed already.

I have written it using if else logic. My question is will it be more efficient if I use tryCatch()? Is there any other advantage than the ability to handle errors as it is not important to handle errors in this context.

Below is the code I am currently using:

Inst.Pkgs=function(){

  if (!require(tm)) 
    {install.packages("tm")
    require(tm)}

   if (!require(SnowballC)) 
    {install.packages("SnowballC")
     require(SnowballC)}

  if (!require(wordcloud)) 
    {install.packages("wordcloud")
    require(wordcloud)}

  if (!require(RWeka)) 
    {install.packages("RWeka")
    require(RWeka)}

  if (!require(qdap)) 
    {install.packages("qdap") 
    require(qdap)}

  if (!require(timeDate)) 
    {install.packages("timeDate") 
    require(timeDate)}

  if (!require(googleVis)) 
  {install.packages("googleVis") 
    require(googleVis)}

  if (!require(rCharts)) 
  { 
    if (!require(downloader)) 
    {install.packages("downloader") 
      require(downloader)}

    download("https://github.com/ramnathv/rCharts/archive/master.tar.gz", "rCharts.tar.gz")
    install.packages("rCharts.tar.gz", repos = NULL, type = "source")
    require(rCharts)
  }

}

Answer 1

You can check at once and install missing packages.

# Definition of function out, the opposite of in
"%out%" <- function(x, table) match(x, table, nomatch = 0) == 0

# Storing target packages
pkgs <- c("tm", "SnowballC", "wordcloud", "RWeka", "qdap",
          "timeDate", "googleVis")

# Finding which target packages are already installed using
# function installed.packages(). Names of packages are stored
# in the first column
idx <- pkgs %out% as.vector(installed.packages()[,1])

# Installing missing target packages
install.packages(pkgs[idx])

Answer 2

We can in fact use tryCatch for this. If the program tries to load a library that is not installed, it will throw an error - an error that can be caught and resolved with a function called by tryCatch() .

Here's how I would do it:

needed_libs <- c("tm", "SnowballC", "wordcloud", "RWeka", "qdap", "timeDate", "googleVis")
install_missing <- function(lib){install.packages(lib,repos="https://cran.r-project.org/", dependencies = TRUE); library(lib, character.only = TRUE)}
for (lib in needed_libs) tryCatch(library(lib, character.only=TRUE), error = function(e) install_missing(lib))

This will install the missing packages and load the required libraries, as requested in the OP.