代码之家 › 专栏 › 技术社区 › jay.sf

如何用reduce合并多个data.frames并获得有序输出?

reduce merge list r

1

jay.sf · 技术社区 · 7 年前

有一个经典的方法 how to simultaneously merge multiple data.frames in a list .

然而,产量有点混乱。

例子

> L
[[1]]
  a b c  d e
1 5 2 4 10 1

[[2]]
  a b c d e
1 6 7 4 6 1

[[3]]
  a b c d
1 7 3 5 5

[[4]]
  a b c d
1 5 2 6 5

[[5]]
  a b c d
1 4 4 2 8

输出的行 Reduce(.) 由5,1,4,2,3排序,这可能意味着从外部到内部的某种缩减工作。

> Reduce(function(...) merge(..., all=TRUE), L)
> Reduce(function(x, y) merge(x, y, all=TRUE, by=intersect(names(x), names(y))), L)  # same
  a b c  d  e
1 4 4 2  8 NA
2 5 2 4 10  1
3 5 2 6  5 NA
4 6 7 4  6  1
5 7 3 5  5 NA

不管怎样,有没有一种方法可以稍微更改代码以获得如下所示的有序输出?

#   a b c  d  e
# 1 5 2 4 10  1
# 2 6 7 4  6  1
# 3 7 3 5  5 NA
# 4 5 2 6  5 NA
# 5 4 4 2  8 NA

数据

L <- list(structure(list(a = 5L, b = 2L, c = 4L, d = 10L, e = 1L), class = "data.frame", row.names = c(NA, 
-1L)), structure(list(a = 6L, b = 7L, c = 4L, d = 6L, e = 1L), class = "data.frame", row.names = c(NA, 
-1L)), structure(list(a = 7L, b = 3L, c = 5L, d = 5L), class = "data.frame", row.names = c(NA, 
-1L)), structure(list(a = 5L, b = 2L, c = 6L, d = 5L), class = "data.frame", row.names = c(NA, 
-1L)), structure(list(a = 4L, b = 4L, c = 2L, d = 8L), class = "data.frame", row.names = c(NA, 
-1L)))

2 回复 | 直到 7 年前

1

Julius Vainora 7 年前

这是因为 sort 属于 merge :

排序-逻辑。结果是否应按列排序?

所以,你可以用

Reduce(function(...) merge(..., all = TRUE, sort = FALSE), L)
#   a b c  d  e
# 1 5 2 4 10  1
# 2 6 7 4  6  1
# 3 7 3 5  5 NA
# 4 5 2 6  5 NA
# 5 4 4 2  8 NA

2

1

Dan 7 年前

在这里,我用 bind_rows 从 dplyr 打包而不是 merge .

L <- list(structure(list(a = 5L, b = 2L, c = 4L, d = 10L, e = 1L), class = "data.frame", row.names = c(NA, 
          -1L)), structure(list(a = 6L, b = 7L, c = 4L, d = 6L, e = 1L), class = "data.frame", row.names = c(NA, 
          -1L)), structure(list(a = 7L, b = 3L, c = 5L, d = 5L), class = "data.frame", row.names = c(NA, 
          -1L)), structure(list(a = 5L, b = 2L, c = 6L, d = 5L), class = "data.frame", row.names = c(NA, 
          -1L)), structure(list(a = 4L, b = 4L, c = 2L, d = 8L), class = "data.frame", row.names = c(NA, 
          -1L)))

library(dplyr)

Reduce(bind_rows, L) 
#>   a b c  d  e
#> 1 5 2 4 10  1
#> 2 6 7 4  6  1
#> 3 7 3 5  5 NA
#> 4 5 2 6  5 NA
#> 5 4 4 2  8 NA

^{创建日期:2019-02-09

reprex package

(V0.2.1.9000)}