Go to next iteration of lapply() if error in r

Question

I'm using RSelenium and lapply() to scrape a fairly complex set of pages. I occasionally run into problems with the page not loading as expected and thus the code failing.

It happens so rarely that rather than try and deal with every possible error, I would just like to skip the current iteration and go on to the next. It looks like tryCatch() is what I'm looking for, but I'm not sure where to put it in the code.

I know this is not complete but I hope this is enough to go on as it looks like tryCatch() is wrapped around lapply() or with the return() statement. Thanks in advance.

team_id <- c(1:10)
df_list <- lapply (1:length(team_id), function(x) {
        # complex navigation and scraping of multiple sub tables
        # to create a final teamtable
        <code>
        return(teamtable)
})
df <- data.table::rbindlist(df_list)

Answer 1

Let's take an example to take square root of a number.

x <- list(1, 3, 4, 'a', 5)

do.call(rbind, lapply(x, function(p) {
       sqrt(p)
}))

Error in sqrt(p) : non-numeric argument to mathematical function

To avoid the error you can use tryCatch in the following way.

do.call(rbind, lapply(x, function(p) {
    tryCatch(sqrt(p), error = function(e) return(NULL))
}))

#         [,1]
#[1,] 1.000000
#[2,] 1.732051
#[3,] 2.000000
#[4,] 2.236068

Depending on how you want your final output you can decide whether you want to send NULL in final output or NA. When you rbind NULL values they are ignored whereas NA's would still remain in the data indicating that input is not what you expected and there was some error.

---

For your case you can do :

df_list <- do.call(rbind, lapply(x, function(p) {
               tryCatch({
                <code>
                 return(teamtable)
            }, error = function(e) return(NULL))
           }))