r - 合并数据框中的行而不分组

Question

tibble::tibble(
  col1 = c("A","","C","",""),
  col2 = c("string1 part 1","string1 part 2",
           "string2 part 1", "string2 part 2",
           "string3"),
  col3 = c(1, "", 2, "", 3)
)

我想合并 col2 中的行, 像这样获得一个没有空字符串的 tibble

tibble::tibble(
  col1 = c("A","C",NA),
  col2 = c("string1 part 1 string1 part 2", "string2 part 1 string2 part 2", "string3"),
  col3 = c(1,2,3)
)

# A tibble: 3 x 3
  col1  col2                           col3
  <chr> <chr>                         <dbl>
1 A     string1 part 1 string1 part 2     1
2 C     string2 part 1 string2 part 2     2
3 NA    string3                           3

正如我在其他答案中发现的那样，分组似乎不是一个选项，而我没有任何引用列

Answer 1

这适用于 dplyr

tibble::tibble(
  col1 = c("A","","C","",""),
  col2 = c("string1 part 1","string1 part 2",
           "string2 part 1", "string2 part 2",
           "string3"),
  col3 = c(1, "", 2, "", 3)
) %>% 
# fill empty values on col3
mutate(col3 = case_when(
    col3!="" ~ 1,
    T ~ 0
)) %>%
mutate(col3 = cumsum(col3)) %>%
# fill empty values on col1
group_by(col3) %>%
mutate(col1 = first(col1)) %>%
# group & summarise
group_by(col1, col3) %>%
summarise(col2 = paste(col2, collapse=' ')) %>%
# replace empty string by NA & arrange by col3
ungroup() %>%
mutate(col1 = case_when(
    col1=="" ~ as.character(NA),
    T ~ col1
)) %>%
arrange(col3) %>%
select(col1, col2, col3)

输出 :

# A tibble: 3 x 3
#  col1  col2                           col3
#  <chr> <chr>                         <dbl>
#1 A     string1 part 1 string1 part 2     1
#2 C     string2 part 1 string2 part 2     2
#3 <NA>  string3                           3