Best way to deal with warnings

[ldecicco-USGS]

I have many data sets that appear to have integers for > 1000 rows, but then it turns out they are numeric. I'm trying to come up with a nice way to handle that, but I'm stuck trying to figure out how to dynamically use the cols_only feature (and maybe that's just not the right way to go anyway):

readr.data <- suppressWarnings(read_delim(doc, skip = 2, delim="\t", col_names = FALSE))
badCols <- problems(readr.data)$col
badCols
[1] "X4" "X4"
unique.bad.cols <- unique(badCols)

readr.cols <- read_delim(doc, skip =2, delim="\t",col_names = FALSE, 
                               col_types = cols_only(unique.bad.cols = "n"))
Warning message:
The following named parsers don't match the column names: unique.bad.cols 

#I've also tried:
readr.cols <- read_delim(doc, skip = 2, delim="\t",col_names = FALSE, 
                               col_types = cols_only(setNames("n",unique.bad.cols)))
Error: not compatible with STRSXP
In addition: Warning message:
Unnamed `col_types` should have the same length as `col_names`. Using smaller of the two.

#Other attempts, re-grabbing and parsing all the data instead of just the bad columns:
#a: would like to use the variable unique.bad.cols in col_double
readr.data <- read_delim(doc, skip = 2, delim="\t",col_names = FALSE, 
                                col_types = col_double(unique.bad.cols))
Error in col_double(unique.bad.cols) : unused argument (unique.bad.cols)

#b: use setNames
readr.data <- read_delim(doc, skip = 2, delim="\t",col_names = FALSE, 
+                          col_types = cols(setNames("n",unique.bad.cols)))
Error: not compatible with STRSXP
In addition: Warning message:
Unnamed `col_types` should have the same length as `col_names`. Using smaller of the two. 

#This works just fine, but it won't always be "X4":
readr.cols <- read_delim(doc, skip = 2,delim="\t",col_names = FALSE, 
                          col_types = cols_only("X4"="n"))

After more digging, I realize I had assumed:
cols_only("X4"="n")
was the same as
cols_only(c("X4"="n"))
which it is not (hence the setName function doesn't help).

Any advice would be greatly appreciated. Thanks for the fantastic package!

[ldecicco-USGS]

So here's what I ended up doing, it seems to work, I just don't want to miss something obvious from readr that would work:

    readr.data <- suppressWarnings(read_delim(doc, skip = 2, delim="\t", col_names = FALSE))
    badCols <- problems(readr.data)$col
    if(length(badCols) > 0){
      unique.bad.cols <- unique(badCols)
      readr.data.char <- read_delim(doc, skip = 2, delim="\t", col_names = FALSE, 
                                    col_types = cols(.default = "c"))
      readr.data[,unique.bad.cols] <- lapply(readr.data.char[,unique.bad.cols], parse_number)
    }

[hadley]

I think #401 will solve this problem