R语言data.frame指定列拆分求助：separate_wider_delim执行报错

我有一个名为table_8[1:2,]的小型data.frame，结构如下：

structure(list(X1 = c("", ""), X2 = c("Wealth", "Ratio (75/25)"), X3 = c("", "Gini"), X4 = c("Non-Land Wealth", "Ratio (75/25) Gini"), X5 = c("Income", "Ratio (75/25)"), X6 = c("", "Gini"), X7 = c("Consumption", "Ratio (75/25) Gini")), row.names = 1:2, class = "data.frame")

需要将X4和X7列的第二行内容各自拆分为两列，使其格式与X2、X3列一致。尝试了以下代码：

table_8_n <- table_8[1:2,] |>
mutate(across(X4:X7, ~str_replace_all(.x, ") G", ")_G")))|>
separate_wider_delim(cols = c(X4,X7), names=LETTERS[1:2], names_sep = "", delim="_", too_few = "align_start")

但出现报错：

Error in `mutate()`:
ℹ In argument: `across(X4:X7, ~str_replace_all(.x, ") G", ")_G"))`.
Caused by error in `across()`:
! Can't compute column `X4`.
Caused by error in `stri_replace_all_regex()`:
! Incorrectly nested parentheses in regex pattern. (U_REGEX_MISMATCHED_PAREN, context=`) G`)

不确定该替换across函数还是用其他方式解决。

错误根源 #

报错是因为str_replace_all的第一个参数是正则表达式，而)属于正则特殊字符，未转义会被当作分组结束符，导致括号匹配错误。另外你不需要用across处理X4到X7的所有列，只需要处理X4和X7即可。

方法一：转义正则特殊字符 #

把)转义为\\)，让正则引擎识别为普通字符，同时缩小across的处理范围：

library(tidyverse)

table_8 <- structure(list(X1 = c("", ""), X2 = c("Wealth", "Ratio (75/25)"), X3 = c("", "Gini"), X4 = c("Non-Land Wealth", "Ratio (75/25) Gini"), X5 = c("Income", "Ratio (75/25)"), X6 = c("", "Gini"), X7 = c("Consumption", "Ratio (75/25) Gini")), row.names = 1:2, class = "data.frame")

table_8_n <- table_8 |>
  mutate(across(c(X4, X7), ~str_replace_all(.x, "\\) G", "\\)_G"))) |>
  separate_wider_delim(
    cols = c(X4, X7), 
    names = c(str_remove(cur_column(), "X"), str_remove(cur_column(), "X")), 
    delim = "_", 
    too_few = "align_start"
  )

这里用cur_column()自动生成对应列名，避免用LETTERS导致命名混乱，同时只处理需要修改的X4和X7列。

方法二：直接处理目标行，规避正则 #

更直观的方式是定位X4和X7的第二行，直接拆分后插入新列，不需要处理正则转义：

library(tidyverse)

table_8 <- structure(list(X1 = c("", ""), X2 = c("Wealth", "Ratio (75/25)"), X3 = c("", "Gini"), X4 = c("Non-Land Wealth", "Ratio (75/25) Gini"), X5 = c("Income", "Ratio (75/25)"), X6 = c("", "Gini"), X7 = c("Consumption", "Ratio (75/25) Gini")), row.names = 1:2, class = "data.frame")

# 拆分X4列并生成新列
x4_split <- str_split(table_8$X4[2], " ", n=2)[[1]]
table_8 <- table_8 |>
  mutate(X4a = if_else(row_number() == 1, "", x4_split[1]),
         X4b = if_else(row_number() == 1, "Gini", x4_split[2])) |>
  select(-X4)

# 拆分X7列并生成新列
x7_split <- str_split(table_8$X7[2], " ", n=2)[[1]]
table_8 <- table_8 |>
  mutate(X7a = if_else(row_number() == 1, "", x7_split[1]),
         X7b = if_else(row_number() == 1, "Gini", x7_split[2])) |>
  select(-X7)

# 调整列顺序，对齐原表格格式
table_8 <- table_8 |>
  select(X1, X2, X3, X4a, X4b, X5, X6, X7a, X7b)

这种方法直接针对需要修改的内容操作，逻辑更清晰，也不会遇到正则特殊字符的问题。

内容的提问来源于stack exchange，提问作者hks