CODEWITHSUNDEEP
rdata.tabledplyr
pachadotdev
pachadotdev

Reputation: 3775

Write data.table syntax as its dplyr equivalent to fix 9218868437227407266's integer64

I have the integer64 issue that instead of NA it shows 9218868437227407266 as it's described here: fread() fails with missing values in integer64 columns

I have loaded bit64 package that with data.table 1.9.5 or greater supposedly displays NAs instead of 9218868437227407266s but it doesn't work. The issue is described here and is a bit64 issue: https://github.com/Rdatatable/data.table/issues/488

I wrote this as a "solution":

dt[as.character(my_col) == "9218868437227407266", my_col := as.integer64(NA)]

It works but I wonder how I can write that in dplyr syntax to use the %>% without unexpected crashes.

Many thanks in advance !

Upvotes: 0

Views: 568

Answers (1)

akrun
akrun

Reputation: 887501

We can use replace

library(dplyr)
dt %>%
   mutate(my_col = replace(my_col, as.character(my_col) == "9218868437227407266",
                         as.integer64(NA))

Upvotes: 1

Related Questions