Argument is of length 0, no NA's - r

Question

I have a dataframe that looks like this:
logentrytime ord_lat_dt0 ord_lat_dt1 ord_lat_dt2 ord_lat_dt3 ord_lat_dt4 ord_lat_dt5 ord_lat_dt6 ord_lat_dt7 ord_lat_dt8 ord_lat_dt9 ord_num0 ord_num1 ord_num2
1 2016-11-10 14:23:36 0 0 0 0 0 0 2016-02-12 0 0 0 0 0 0
2 2016-11-10 14:22:22 0 0 0 0 0 0 2016-02-12 0 0 0 0 0 0
3 2016-11-07 16:02:45 0 0 0 0 0 0 2016-02-12 0 0 0 0 0 0
4 2016-11-07 21:10:00 0 0 0 0 0 0 2016-02-12 0 0 0 0 0 0
5 2016-11-07 16:03:29 0 0 0 0 0 0 2016-02-12 0 0 0 0 0 0
6 2016-11-10 14:23:05 0 0 0 0 0 0 2016-02-12 0 0 0 0 0 0
Where ord_lat_dt columns are last purchase date of a customer. ord_lat_dt[0-9] were pulled from different database tables. Thus each row represents one customer, and their last order date will be indicated in one of the 9 columns.
I would like to merge these, but before I do, want to calculate "months_since_last_purchase" based on the date in each column.
Thus, I have converted the date columns into character strings, and am looping through using these functions:
elapsed_time <- function(end_date, start_date) {
ed <- as.POSIXlt(end_date)
sd <- as.POSIXlt(start_date)
12 * (ed$year - sd$year) + (ed$mon - sd$mon)
}
convert_time <- function(data, column){
for(i in seq(1,length(data$column))){
if((data$column[i]!= "0") ==TRUE){
data$column[i] <- elapsed_months(Sys.time(), as.Date(data$column[i], format="%Y/%m/%d"))
}
}
return(data)
}
test1 <- convert_time(test2, ord_lat_dt0)
But I obtain the error
Error in if ((data$column[i] != "0") == TRUE) { :
argument is of length zero
I've also tried changing the if statement to check :
grepl("[-]", data$column[i])==FALSE)
But I obtain the same error.
Any ideas?
If you decide to down vote, please explain to me what is wrong with my question. I am trying to learn and would like to make sure I am asking correctly.
NOTE: I am having a different issue and have completely changed the question. Thus some of the comments below do not apply. Because of the down votes, I could not open a new question.

Answer 1

The problem here is that when you do data_theme[is.na(data_theme)] <- 0, NA in the date columns will be replaced. But date columns are in a POSIXct format, and if you try as.POSIXct(0), it will through an error.
One solution could be to do it in two step. First replace NA from numeric column first, then do whatever you want with POSIXct values :
library(dplyr)
df %>%
mutate_if("is.numeric", funs(if_else(is.na(.), 0, .))

Answer 2

You can only replace all NAs by the value 0 if all columns are numeric first. This could for instance be achieved by writing a little function to first convert a column to numeric if needed, and then replace the NAs. Using lapply you can loop over the columns, and make the resulting list of columns a data frame again afterwards.
f <- function(x) {
x <- as.numeric(x)
x[is.na(x)] <- 0
x
}
data_theme <- as.data.frame(lapply(data_theme, f))
Of course, this will also convert any meaningful datetimes to numbers.