R : using filter in a for loop and change the data value-CodePudding

I have a list called 'col_missing' like below.

col_missing <- list(
  RAgeCat = c(-1, 8)
  , Married = c(-1, 9)
  , skipmeal = c(-1, 8, 9)
)

and I want to filter the dataframe if the column has above values, and replace them as NA. Below code is working but if it is in a for loop, it is not working.

Working

temp <- df %>%  
  filter(skipmeal %in% col_missing$skipmeal)
temp$skipmeal <- NA

Not working

for (col in names(col_missing)) {
  temp <- df %>%
    filter(col %in% col_missing$col)
  temp$col <- NA
}

Could you tell me how to do this in a for loop?

CodePudding user response：

We may do this using across and modify the values without any filtering

library(dplyr)
df <- df %>%
    mutate(across(all_of(names(col_missing)),
     ~ replace(.x, .x %in% col_missing[[cur_column()]], NA)))

In the for loop, temp is getting updated in each iteration. Perhaps it should be created outside Also, use [[ instead of $ i.e. col_missing[[col]] instead of col_missing$col

If we need to filter and create multiple datasets in a list

library(purrr)
imap(col_missing, ~ df %>%
         filter(.data[[.y]] %in% .x) %>%
         mutate(!! .y := NA)
)