在R中如何根据另一列文本存在情况填充数据框指定列？

需求说明 #

当first_column存在文本时，用second_column的非空值填充对应行；first_column为空的行，保持second_column原有的空值。

问题分析 #

直接用fill()函数（无论向上还是向下）无法满足需求，因为fill()会批量填充所有空值（或NA），不会区分first_column是否有文本。需要先传播second_column的非空值，再根据first_column的内容过滤填充范围。

解决方案1：使用dplyr + tidyr（推荐） #

通过临时列向上传播非空值，再仅对first_column非空的行应用填充值：

library(dplyr)
library(tidyr)

# 构造原始数据框
current_dataframe <- data.frame(
  first_column = c("eeer","","","sdfdfdsf", "sdfsdfdsf","faffa", "we", "", "", "", "", "", "eeee", "", "wqwq", "", "ttetxg", "", "sf", "sf","f", "dfsdf", "sqdweg", "sfsfdsdvghhh", ""),
  second_column = c("","", "","", "", "", "", "renato paulo", "", "", "", "", "", "carlos gomes", "", "", "", "vitor", "","","","","","", "Cassia rabelo"),
  stringsAsFactors = FALSE
)

# 执行填充逻辑
result <- current_dataframe %>%
  # 将空字符串转为NA，方便填充操作
  mutate(temp_col = ifelse(second_column == "", NA, second_column)) %>%
  # 向上传播非空值，覆盖上方的NA
  fill(temp_col, .direction = "up") %>%
  # 仅在first_column非空的行，用传播后的值替换second_column
  mutate(second_column = ifelse(first_column != "", temp_col, second_column)) %>%
  # 清理临时列，将NA转回空字符串
  select(-temp_col) %>%
  mutate(second_column = ifelse(is.na(second_column), "", second_column))

# 验证结果是否符合预期
desired_output <- data.frame(
  first_column = c("eeer","","","sdfdfdsf", "sdfsdfdsf","faffa", "we", "", "", "", "", "", "eeee", "", "wqwq", "", "ttetxg", "", "sf", "sf","f", "dfsdf", "sqdweg", "sfsfdsdvghhh", ""),
  second_column = c("","", "","renato paulo", "renato paulo", "renato paulo", "renato paulo", "", "", "", "", "", "carlos gomes", "", "", "", "vitor", "", "Cassia rabelo","Cassia rabelo","Cassia rabelo","Cassia rabelo","Cassia rabelo","Cassia rabelo", ""),
  stringsAsFactors = FALSE
)

all.equal(result, desired_output) # 返回TRUE表示结果正确

解决方案2：Base R实现（无需额外包） #

通过循环定位second_column的非空值，再精准填充对应first_column非空的行：

# 构造原始数据框
first_column <- c("eeer","","","sdfdfdsf", "sdfsdfdsf","faffa", "we", "", "", "", "", "", "eeee", "", "wqwq", "", "ttetxg", "", "sf", "sf","f", "dfsdf", "sqdweg", "sfsfdsdvghhh", "")
second_column <- c("","", "","", "", "", "", "renato paulo", "", "", "", "", "", "carlos gomes", "", "", "", "vitor", "","","","","","", "Cassia rabelo")
current_dataframe <- data.frame(first_column, second_column, stringsAsFactors = FALSE)

# 初始化结果列
result_col <- second_column
# 找到second_column非空值的位置
non_empty_pos <- which(second_column != "")
# 起始填充位置
start_pos <- 1

# 循环处理每个非空值
for (pos in non_empty_pos) {
  current_val <- second_column[pos]
  # 确定当前非空值上方的填充范围
  fill_range <- start_pos:(pos - 1)
  # 筛选范围内first_column非空的行
  target_rows <- fill_range[first_column[fill_range] != ""]
  # 填充对应行
  result_col[target_rows] <- current_val
  # 更新下一次填充的起始位置
  start_pos <- pos + 1
}

# 生成结果数据框
result <- current_dataframe
result$second_column <- result_col

# 验证结果
desired_output <- data.frame(
  first_column = c("eeer","","","sdfdfdsf", "sdfsdfdsf","faffa", "we", "", "", "", "", "", "eeee", "", "wqwq", "", "ttetxg", "", "sf", "sf","f", "dfsdf", "sqdweg", "sfsfdsdvghhh", ""),
  second_column = c("","", "","renato paulo", "renato paulo", "renato paulo", "renato paulo", "", "", "", "", "", "carlos gomes", "", "", "", "vitor", "", "Cassia rabelo","Cassia rabelo","Cassia rabelo","Cassia rabelo","Cassia rabelo","Cassia rabelo", ""),
  stringsAsFactors = FALSE
)

all.equal(result, desired_output) # 返回TRUE表示结果正确

内容的提问来源于stack exchange，提问作者Francisco1988