r - 使用 `tidyr::unnest_longer()` 取消嵌套/矩形化/展平嵌套列表

Question

我一直在尝试了解 tidyr 和 tibblify 中的取消嵌套函数。我相信您应该能够使用 unnest_longer() 来复制下面将这种嵌套列表转换为 tibble 的更多手动方法，但我一直在处理这些文档。如何执行此操作的正确示例将对我有很大帮助:

# Example nested list
nl <- list(time = list("2023-02-06", "2023-02-07", "2023-02-08",
                       "2023-02-09", "2023-02-10", "2023-02-11",
                       "2023-02-12"), 
           precipitation_sum = list(0.9, 0, 0, 0.3, 0, 0, 0))

# one way to do it (extract colnames and construct)
tibble(!!! setNames(map(nl, unlist),names(nl)))

# another way (collect & reduce each sublist)
as_tibble(lapply(nl, function(x) Reduce(c, x)))

# how to use tidyr and unnest_longer? (below is incorrect)
unnest_longer(tibble(nl), col = everything())

Answer 1

我们可以使用

library(tibble)
library(tidyr)
as_tibble(nl) %>% 
    unnest(cols = where(is.list))

-输出

# A tibble: 7 × 2
  time       precipitation_sum
  <chr>                  <dbl>
1 2023-02-06               0.9
2 2023-02-07               0  
3 2023-02-08               0  
4 2023-02-09               0.3
5 2023-02-10               0  
6 2023-02-11               0  
7 2023-02-12               0  

---

或更紧凑

library(purrr)
map_dfc(nl, unlist)
# A tibble: 7 × 2
  time       precipitation_sum
  <chr>                  <dbl>
1 2023-02-06               0.9
2 2023-02-07               0  
3 2023-02-08               0  
4 2023-02-09               0.3
5 2023-02-10               0  
6 2023-02-11               0  
7 2023-02-12               0