将整数列转换为 R 中的字符时，如何防止前导空格？

Question

我有一个带有字符和整数列的 data.frame。我想将它们全部转换为字符，但我得到了数字列不需要的前导空格：

> example <- data.frame(a=1:10,b=1:10,c=rep("foo",10))
> apply(example,2,format,trim=T)
      a    b    c    
 [1,] " 1" " 1" "foo"
 [2,] " 2" " 2" "foo"
 [3,] " 3" " 3" "foo"
 [4,] " 4" " 4" "foo"
 [5,] " 5" " 5" "foo"
 [6,] " 6" " 6" "foo"
 [7,] " 7" " 7" "foo"
 [8,] " 8" " 8" "foo"
 [9,] " 9" " 9" "foo"
[10,] "10" "10" "foo"

trim=T 参数显然被忽略了。这只发生在字符列'c' 存在的情况下，即如果'c' 不存在（apply(example[,-3],...)），它可以正常工作。

Answer 1

如果我没记错的话，是因为as.matrix，但是你可以使用sapply绕过这个：

> sapply(example, format, trim = TRUE)
      a    b    c    
 [1,] "1"  "1"  "foo"
 [2,] "2"  "2"  "foo"
 [3,] "3"  "3"  "foo"
 [4,] "4"  "4"  "foo"
 [5,] "5"  "5"  "foo"
 [6,] "6"  "6"  "foo"
 [7,] "7"  "7"  "foo"
 [8,] "8"  "8"  "foo"
 [9,] "9"  "9"  "foo"
[10,] "10" "10" "foo"

Answer 2

如果您对字符矩阵作为输出没问题（您似乎基于您对apply 的使用，请尝试）：

do.call(cbind, lapply(example, as.character))

这会产生：

      a    b    c    
 [1,] "1"  "1"  "foo"
 [2,] "2"  "2"  "foo"
 [3,] "3"  "3"  "foo"
 [4,] "4"  "4"  "foo"
 [5,] "5"  "5"  "foo"
 [6,] "6"  "6"  "foo"
 [7,] "7"  "7"  "foo"
 [8,] "8"  "8"  "foo"
 [9,] "9"  "9"  "foo"
[10,] "10" "10" "foo"

Answer 3

正如?apply 中所说，第一个参数被强制转换为矩阵。在这种情况下，由于列c，它将其转换为字符矩阵。对as.matrix 的调用会创建前导空格。随后对format 的调用什么也不做，因为数据已经是字符了。

> as.matrix(example)
      a    b    c    
 [1,] " 1" " 1" "foo"
 [2,] " 2" " 2" "foo"
 [3,] " 3" " 3" "foo"
 [4,] " 4" " 4" "foo"
 [5,] " 5" " 5" "foo"
 [6,] " 6" " 6" "foo"
 [7,] " 7" " 7" "foo"
 [8,] " 8" " 8" "foo"
 [9,] " 9" " 9" "foo"
[10,] "10" "10" "foo"

如果没有c列，则转换为整数矩阵，format将整数转换为字符。

> as.matrix(example[,-3])
       a  b
 [1,]  1  1
 [2,]  2  2
 [3,]  3  3
 [4,]  4  4
 [5,]  5  5
 [6,]  6  6
 [7,]  7  7
 [8,]  8  8
 [9,]  9  9
[10,] 10 10

最好直接使用lapply：

example <- data.frame(a=1:10,b=1:10,c=rep("foo",10))
example[] <- lapply(example, format, trim=TRUE)
# use sapply if you really want a matrix
example <- sapply(example, format, trim=TRUE)