我们需要na.rm = TRUE

anymatch <- function(vars){
  rowSums(select(cur_data(), all_of(vars))=="Yes", na.rm = TRUE)
}

现在它给出了正确的计数

>  mydata %>%
   mutate(afin = anymatch(avars)) %>% 
   select(all_of(avars), afin) 
# A tibble: 100 × 5
   a1    a2    a3    a4     afin
   <chr> <chr> <chr> <chr> <dbl>
 1 <NA>  <NA>  <NA>  <NA>      0
 2 <NA>  Yes   <NA>  Yes       2
 3 Yes   <NA>  <NA>  <NA>      1
 4 <NA>  Yes   Yes   <NA>      2
 5 Yes   Yes   <NA>  <NA>      2
 6 Yes   Yes   Yes   Yes       4
 7 <NA>  Yes   <NA>  <NA>      1
 8 <NA>  <NA>  <NA>  <NA>      0
 9 Yes   Yes   <NA>  Yes       3
10 <NA>  Yes   <NA>  <NA>      1
# … with 90 more rows
# ℹ Use `print(n = ...)` to see more rows

在将来的版本中，我们可能会使用pick而不是cur_data()

anymatch <- function(vars){
   rowSums(pick(all_of(vars))=="Yes", na.rm = TRUE)
 }
mydata %>%
   mutate(afin = anymatch(avars)) %>% 
    select(all_of(avars), afin) 
# A tibble: 100 × 5
   a1    a2    a3    a4     afin
   <chr> <chr> <chr> <chr> <dbl>
 1 <NA>  <NA>  <NA>  <NA>      0
 2 <NA>  Yes   <NA>  Yes       2
 3 Yes   <NA>  <NA>  <NA>      1
 4 <NA>  Yes   Yes   <NA>      2
 5 Yes   Yes   <NA>  <NA>      2
 6 Yes   Yes   Yes   Yes       4
 7 <NA>  Yes   <NA>  <NA>      1
 8 <NA>  <NA>  <NA>  <NA>      0
 9 Yes   Yes   <NA>  Yes       3
10 <NA>  Yes   <NA>  <NA>      1
# … with 90 more rows