连接每组一列的值[重复]

Question

这个问题在这里已有答案：

• Collapse / concatenate / aggregate a column to a single comma separated string within each group 3回答

我想连接现在在一列中的每个组的值。以下是我想要争论的数据框的简短版本。

library(tidyverse)

df <- tibble::tribble(
  ~county,  ~party,
      "A",   "VVD",
      "A",    "GL",
      "A", "Local",
      "B",   "D66",
      "B", "Local"
  )

现在我想为每个县创建一行，其中所有各方都在他们自己的列中：

df2 <- tibble::tribble(
  ~county, ~party1, ~party2, ~party3,
      "A",   "VVD",    "GL", "Local",
      "B",   "D66", "Local",      NA
  )

稍后与unite()连接并替换逗号的下划线并删除NA。

df2 %>%
  unite(party, c("party1", "party2", "party3")) %>%
  mutate(party = gsub("_NA", "", party),
         party = gsub("_", ", ", party))

我想要的df输出：

  county party         
  <chr>  <chr>         
1 A      VVD, GL, Local
2 B      D66, Local

Answer 1

我们可以通过创建序列列和spread来实现

library(tidyverse)
df %>%
   group_by(county) %>% 
   mutate(v1 = paste0('party', row_number())) %>% 
   spread(v1, party)
# A tibble: 2 x 4
# Groups:   county [2]
#  county party1 party2 party3
#  <chr>  <chr>  <chr>  <chr> 
#1 A      VVD    GL     Local 
#2 B      D66    Local  <NA>  

---

对于第二个输出，我们将'county'和paste分组为'party'的元素

df %>%
  group_by(county) %>%
  summarise(party = toString(party))
# A tibble: 2 x 2
#  county party         
#  <chr>  <chr>         
#1 A      VVD, GL, Local
#2 B      D66, Local   

Answer 2

df %>%
    group_by(county) %>%
    dplyr::summarise( paste0(party, collapse = ", "))

如果有任何不清楚的地方，你应该将?group_by（?paste0等）提示给R控制台。