如何在R的lapply中正确提取数据框原名并重命名列？

问题原因 #

你用deparse(substitute(x))无法获取列表中DataFrame的原始名称，是因为lapply的匿名函数里，x只是列表元素的临时引用，不是原始对象名，所以会返回x[[i]]这类内部标识。

可行方案 #

方案1：用purrr::imap（推荐） #

imap会同时传递列表的元素和元素名称，直接就能拿到每个DataFrame的原始名称：

library(purrr)
library(dplyr)
library(tidyr)

df_list <- imap(df_list, function(x, df_name){
  lookup <- c(start = paste(df_name, "_start"),
              end = paste(df_name, "_end"),
              top = paste(df_name, "_top"),
              left = paste(df_name, "_left"),
              height = paste(df_name, "_height"),
              width = paste(df_name, "_width"),
              type = paste(df_name, "_type"),
              value = paste(df_name, "_value"))
  x <- x %>% 
    rename_with(.fn = ~lookup[.x], .cols = intersect(names(.), names(lookup))) %>%
    arrange(creativeId) %>%
    select(-1) %>%
    distinct() %>%
    mutate(counter = with(., ave(creativeId, with(rle(creativeId), rep(seq_along(values), lengths)), FUN = seq_along))) %>%
    relocate(counter) %>%
    pivot_wider(names_from = counter, values_from = -names(.)[1:2])
})

new_df <- Reduce(function(x,y) merge(x,y,all=TRUE), df_list)

方案2：基础R遍历列表名称 #

如果不想依赖purrr包，直接遍历列表的名称，通过名称索引对应DataFrame：

df_names <- names(df_list)

df_list <- lapply(df_names, function(df_name){
  x <- df_list[[df_name]]
  lookup <- c(start = paste(df_name, "_start"),
              end = paste(df_name, "_end"),
              top = paste(df_name, "_top"),
              left = paste(df_name, "_left"),
              height = paste(df_name, "_height"),
              width = paste(df_name, "_width"),
              type = paste(df_name, "_type"),
              value = paste(df_name, "_value"))
  x <- x %>% 
    rename_with(.fn = ~lookup[.x], .cols = intersect(names(.), names(lookup))) %>%
    arrange(creativeId) %>%
    select(-1) %>%
    distinct() %>%
    mutate(counter = with(., ave(creativeId, with(rle(creativeId), rep(seq_along(values), lengths)), FUN = seq_along))) %>%
    relocate(counter) %>%
    pivot_wider(names_from = counter, values_from = -names(.)[1:2])
})

# 可选：恢复列表名称，方便后续查看
names(df_list) <- df_names

new_df <- Reduce(function(x,y) merge(x,y,all=TRUE), df_list)

注意事项 #

• 确保你的df_list是带命名的列表，如果没有命名，先手动添加：names(df_list) <- c("df1", "df2", ...)（替换为你的DataFrame实际名称）。

内容的提问来源于stack exchange，提问作者Jess