Tolerate blank lines ... at least at the end of input?

[jennybc]

library(readr)
input <- "x,y\n1,2\n\n"
read.csv(text = input)
#>   x y
#> 1 1 2
read_csv(input)
#> Warning: 1 parsing failure.
#> row col  expected    actual
#>   2  -- 2 columns 1 columns
#> Source: local data frame [2 x 2]
#> 
#>       x     y
#>   (int) (int)
#> 1     1     2
#> 2    NA    NA
read_csv(trimws(input))
#> Source: local data frame [1 x 2]
#> 
#>       x     y
#>   (int) (int)
#> 1     1     2

It seems reasonable to tolerate and ignore blank lines ... at least at the end of the input.

[hadley]

Yeah, I think that's totally reasonable (and for empty lines anywhere, not just at the end). It's just a little tricky to implement because you need to think more about what line numbers to report in error messages etc - the original line numbers, or the row numbers in the generated data frame.