Summing data frames with non-numeric values-CodePudding

I would like to sum dataframes that contain non-numeric arguments. When I do the simple add function I receive the following error: "Error in FUN(left, right) : non-numeric argument to binary operator".

How can I solve this?

ex1:

`04:00` `04:10` `04:20`
  <chr>   <chr>   <chr>  
1 a       0       a      
2 0       a       a      
3 0       0       a

ex2:

`04:00` `04:10` `04:20`
  <chr>   <chr>     <dbl>
1 0       b             0
2 b       0             0
3 b       b             0

Desired outcome:

`04:00` `04:10` `04:20`
      <chr>   <chr>     <dbl>
    1 a       b             a
    2 b       a             a
    3 b       b             a

Sample code:

sum = ex1   ex2

Sample data:

ex1<-structure(list(`04:00` = c("a", "0", "0"), `04:10` = c("0", "a", 
"0"), `04:20` = c("a", "a", "a")), spec = structure(list(cols = list(
    `04:00` = structure(list(), class = c("collector_character", 
    "collector")), `04:10` = structure(list(), class = c("collector_character", 
    "collector")), `04:20` = structure(list(), class = c("collector_character", 
    "collector"))), default = structure(list(), class = c("collector_guess", 
"collector")), delim = ","), class = "col_spec"),  row.names = c(NA, 
-3L), class = c("spec_tbl_df", "tbl_df", "tbl", "data.frame"))
    

ex2<-structure(list(`04:00` = c("0", "b", "b"), `04:10` = c("b", "0", 
"b"), `04:20` = c(0, 0, 0)), spec = structure(list(cols = list(
    `04:00` = structure(list(), class = c("collector_character", 
    "collector")), `04:10` = structure(list(), class = c("collector_character", 
    "collector")), `04:20` = structure(list(), class = c("collector_double", 
    "collector"))), default = structure(list(), class = c("collector_guess", 
"collector")), delim = ","), class = "col_spec"),  row.names = c(NA, 
-3L), class = c("spec_tbl_df", "tbl_df", "tbl", "data.frame"))

CodePudding user response：

For the test case this works:

data.frame(
  Map(\(x, y) ifelse(x == "0", y, x), 
      ex1, ex2), 
  check.names = FALSE)
#  04:00 04:10 04:20
#1     a     b     a
#2     b     a     a
#3     b     b     a

This doesn't check if column names match. It simply iterates over the columns of both data.frames simultaneously.

CodePudding user response：

You could first set your zeros to NAs, and then use dplyr::coalesce :

l <- map(list(ex1,ex2), ~ mutate(.x, across(.fns = ~ replace(., 0, NA_character_))))
do.call(coalesce, l)

# A tibble: 3 x 3
  `04:00` `04:10` `04:20`
1 a       b       a      
2 b       a       a      
3 b       b       a