R-根据ID填充空白值

Question

小问题：我想根据id的分组填充id2的值。

由此

> head(dta)
    id     id2
1 B10388W4       0
2 B10388W4 B10388W
3 B10388W4 B10388W

简单地说

    id     id2
1 B10388W4 B10388W
2 B10388W4 B10388W
3 B10388W4 B10388W

根据组id填充值的巧妙方法是什么？

我想到了类似的东西

dta %>% 
  group_by(id) %>% 
  mutate( id3 = ifelse(id2 == 0, lead(id2), id2) )

但这不是很聪明，因为在其他地方可以找到相同id 。

任何想法欢迎。

数据

dta = structure(list(id = c("B10388W4", "B10388W4", "B10388W4"), 
id2 = c("0", "B10388W", "B10388W")), row.names = c(NA, -3L), 
class = "data.frame", .Names = c("id", "id2"))

Answer 1

对于每个id （组），您可以使用唯一的id2值（不是0 ，并更新id2列。 假设对于每个唯一的id值，只有一个唯一的id2值，也可能是0值。

# my example dataset
dt = data.frame(id = c("B10388W4","B10388W4","B10388W4","A10388W4","A10388W4","A10388W4"),
                id2 = c(0,"B10388W","B10388W","A10388W",0,0),
                stringsAsFactors = F)

dt

#         id     id2
# 1 B10388W4       0
# 2 B10388W4 B10388W
# 3 B10388W4 B10388W
# 4 A10388W4 A10388W
# 5 A10388W4       0
# 6 A10388W4       0


library(dplyr)

dt %>% 
group_by(id) %>%
mutate(id2_new = unique(id2[id2 != 0])) %>%
select(-id2) %>%
ungroup

#          id id2_new
#       (chr)   (chr)
# 1 A10388W4 A10388W
# 2 A10388W4 A10388W
# 3 A10388W4 A10388W
# 4 B10388W4 B10388W
# 5 B10388W4 B10388W
# 6 B10388W4 B10388W

Answer 2

dta %>%
  filter(id2 != 0) %>%
  distinct %>%
  full_join(dta %>% select(id1) )