r - 使用 dplyr 修改列中的一行

Question

说我有这两个tibbles

library(dplyr)
library(lubridate)

my_date <- as_date(c("2021-10-30","2021-10-29","2021-10-28"))
my_t1 <- tibble(date = my_date) %>% 
  mutate("col1" = NA, 
         "col2" = NA, 
         "col3" = NA, 
         "col4" = NA)
my_t2 <- tibble("column" =  c("col1","col2","col3","col4"), 
                "value" = c(10,20,30,40))

所以

> my_t1
# A tibble: 3 × 5
  date       col1  col2  col3  col4 
  <date>     <lgl> <lgl> <lgl> <lgl>
1 2021-10-30 NA    NA    NA    NA   
2 2021-10-29 NA    NA    NA    NA   
3 2021-10-28 NA    NA    NA    NA  

> my_t2
# A tibble: 4 × 2
  column value
  <chr>  <dbl>
1 col1      10
2 col2      20
3 col3      30
4 col4      40

我想复制的值

my_t2 %>% select(value)

进入 my_t1 的特定行，比如中间一行(date == "2021-10-29")。我想避免对每个元素一个一个地做(即基数 R)，这样 my_t1 就变成了

# A tibble: 3 × 5
  date       col1  col2  col3  col4 
  <date>     <lgl> <lgl> <lgl> <lgl>
1 2021-10-30 NA    NA    NA    NA   
2 2021-10-29 10    20    30    40   
3 2021-10-28 NA    NA    NA    NA   

Answer 1

我们可以使用 match 和 cur_column 将第二行替换为“my_t2”中相应的“column”值

library(dplyr)
my_t1_new <- my_t1 %>% 
    mutate(across(-date, ~ replace(., date == "2021-10-29", 
         my_t2$value[match(cur_column(), my_t2$column)])))

-输出

my_t1_new
# A tibble: 3 × 5
  date        col1  col2  col3  col4
  <date>     <dbl> <dbl> <dbl> <dbl>
1 2021-10-30    NA    NA    NA    NA
2 2021-10-29    10    20    30    40
3 2021-10-28    NA    NA    NA    NA

---

如果列是有序的，base R会更容易

my_t1_new <- as.data.frame(my_t1)
my_t1_new[2, -1] <- my_t2$value