如何批量删除多CSV文件指定行？现有R代码失效求助

测试数据（用于复现场景） #

data <- data.frame(
  "stim" = c("face", "object", " ", "pareidolia"),
  "RT" = c(23, 24, 22, 25),
  "Opac" = c(70, 60, 80, 65)
)
write.csv(data, "data1.csv", row.names = FALSE)

data <- data.frame(
  "stim" = c("face", "pareidolia", " ", "pareidolia"),
  "RT" = c(83, 24,52, 85),
  "Opac" = c(70, 87, 8, 6)
)
write.csv(data, "data2.csv", row.names = FALSE)

原代码核心问题 #

• lapply(files, read.csv)生成的列表无命名，导致names(datalist)为空，最终生成无效的.csv文件名
• 清理空白行的逻辑错误：with(datalist, subset(datalist, !("stim" == "")))无法遍历列表内的每个数据框，无法实现批量清理
• 未关联原文件名与输出文件，导致路径指向混乱

修复后的可行代码 #

library(tidyverse)

# 替换为你的实际工作路径
setwd("your_filepath")

# 获取所有csv文件，保留文件名
files <- list.files(path = ".", recursive = TRUE, pattern = "\\.csv$", full.names = FALSE)

# 读取文件并为列表元素命名（用原文件名作为标识）
datalist <- lapply(files, read.csv, stringsAsFactors = FALSE)
names(datalist) <- tools::file_path_sans_ext(files)

# 批量清理：删除stim列含空白（含纯空格）的行
datalist_cleaned <- lapply(datalist, function(df) {
  df %>%
    filter(trimws(stim) != "")
})

# 生成带标识的输出文件名
file_out <- paste0("cleaned_", names(datalist_cleaned), ".csv")

# 批量写入处理后的文件
mapply(function(x, y) write_csv(x, y, na = ""), datalist_cleaned, file_out)

更高效的tidyverse风格方案 #

用purrr函数链简化流程，代码更简洁可读：

library(tidyverse)

setwd("your_filepath")

# 一站式完成：读取→清理→拆分→写入
list.files(path = ".", recursive = TRUE, pattern = "\\.csv$", full.names = TRUE) %>%
  set_names(nm = str_remove(., "\\.csv$") %>% str_c("_cleaned")) %>%
  map_dfr(read_csv, .id = "source_file") %>%
  filter(trimws(stim) != "") %>%
  group_split(source_file) %>%
  walk2(., str_c(names(.), ".csv"), ~write_csv(.x, .y))

方案优势 #

• 函数链结构避免嵌套代码，逻辑更清晰
• 合并后统一清理再拆分，减少重复遍历开销
• 自动关联原文件名与输出文件，无需手动命名列表

内容的提问来源于stack exchange，提问作者thefriendly_plague.doctor