Is it possible to check an error of a function with an if()?

Question

I’m using the package gtrends to pull some data from google..

This package allows you to download 5 Keywords at a time. That’s why I’m using a for()

However, some Keywords I need to query are not in the period I seek.

When the function gtrends() cannot find the word, it cannot find the error. Examples of Keywords with error: DDD, MMM.

So I’d like to know: is there a way to create a if(), in case of error in the function, it allocate the keyword in a list and, if not, do the rest of the procedure?

library(readr)
library(gtrendsR)


kwlist = readLines("keywords.csv")


resultslist <- list() 

for (keywords in kwlist){
  
  country=c('IT')
  time=("2018-08-01 2018-08-27")
  channel='web'
  trends = gtrends(keywords, gprop =channel,geo=country, time = time )
  resultslist[[keywords]] <- trends$interest_over_time
  
}

The mistake is this:

Error in [<-.data.frame(*tmp*, , timevar, value = "Subject") means: Placement has 1 Row, data has 0

I was told that in older versions of the package, the error does not appear. But I was curious to know if() can be created for an error.

Answer 1

The following code reads the keyword array and creates two lists, a list of results and a list of errors.

library(readr)
library(gtrendsR)

keywords <- c("FLWS","TXG","RETC","YSGG","EFSH","ONEM","ISOL",
              "FNRC","TWGL","TWOU","TDCH","MRJT","DPSM","THDS",
              "DDDX","MMM","FTPM","CATV","FFNT","SSOF","AOS")

resultslist <- vector("list", length = length(keywords))
errorslist <- vector("list", length = length(keywords))
names(resultslist) <- keywords
names(errorslist) <- keywords

country <- 'IT'
time <- "2018-08-01 2018-08-27"
channel <- 'web'

for(keyword in keywords){
  cat("Processando:", keyword, "\n")
  flush.console()
  trends <- tryCatch(gtrends(keyword, gprop = channel, geo = country, time = time),
                     warning = function(w) warning(w),
                     error = function(e) e
                    )
  if(inherits(trends, "error")){
    errorslist[[keyword]] <- trends
  } else {
    resultslist[[keyword]] <- trends$interest_over_time
  }
}

resultslist <- resultslist[!sapply(resultslist, is.null)]
errorslist <- errorslist[!sapply(errorslist, is.null)]

errorslist
#$MMM
#<simpleError in `[<-.data.frame`(`*tmp*`, , timevar, value = "subject"): replacement has 1 row, data has 0>
#
#$AOS
#<simpleError in `[<-.data.frame`(`*tmp*`, , timevar, value = "subject"): replacement has 1 row, data has 0>