Common errors

An error prints Error in <call> : <message> and stops the script (Execution halted). A warning prints Warning message: and the script goes on — but it usually means a result is not what you think.

• object 'x' not found — a typo, a variable that was never assigned, or a column name used outside its data frame (age instead of people$age).
• could not find function "foo" — a typo, a file not loaded with source(), or a function from a package that is not available.
• non-numeric argument to binary operator — text used in arithmetic; convert with as.numeric().
• subscript out of bounds — [[i]] or a matrix index past the end.
• missing value where TRUE/FALSE needed — the condition of if is NA; check with is.na() first.
• the condition has length > 1 — if got a whole vector; use any(), all() or ifelse().
• there is no package called 'dplyr' — only base R and the recommended packages are installed.
• Warning NAs introduced by coercion — as.numeric() met text that is not a number.

values <- c("10", "x", "7")
total <- 0
for (v in values) {
  n <- suppressWarnings(as.numeric(v))
  if (is.na(n)) {
    cat(sprintf("skipped '%s'\n", v))
    next
  }
  total <- total + n
}
cat(sprintf("total: %g\n", total))

result <- tryCatch(stop("boom"), error = function(e) conditionMessage(e))
print(result)

skipped 'x'
total: 17
[1] "boom"