如何处理多DataFrame：合并指定列、删除原列及转换POSIXct格式？

1. Binding Two Columns from Multiple DataFrames in a List #

Suppose you have a list of DataFrames (let’s call it df_list) and want to pull two specific columns (say date_col and time_col) from each, then combine them into a single DataFrame. Here are two practical approaches using base R and the tidyverse:

Base R Method #

# Extract the target columns from each DataFrame in the list
extracted_cols <- lapply(df_list, function(df) df[, c("date_col", "time_col")])
# Bind all extracted columns into one DataFrame
combined_df <- do.call(rbind, extracted_cols)

Tidyverse (purrr) Method #

If you prefer a more concise syntax, map_dfr handles the extraction and binding in one line:

library(purrr)
combined_df <- map_dfr(df_list, ~ .x[, c("date_col", "time_col")])

This automatically stacks the results row-wise into a single DataFrame.

2. Merging Columns, Dropping Originals, and Converting to POSIXct #

Let’s say you have multiple DataFrames loaded in your global environment (e.g., df1, df2, df3). For each, you want to merge date_col and time_col into a datetime column, remove the original two columns, and convert the new column to POSIXct type.

First, get the names of your target DataFrames. If they follow a naming pattern (like starting with "df"), use:

df_names <- ls(pattern = "^df") # Adjust the pattern to match your DataFrame names

Then, loop through each DataFrame to make the changes:

# Base R approach
for (name in df_names) {
  df <- get(name)
  # Combine date and time (adjust the format string to match your data)
  df$datetime <- as.POSIXct(paste(df$date_col, df$time_col), format = "%Y-%m-%d %H:%M:%S")
  # Drop original columns
  df <- df[, !names(df) %in% c("date_col", "time_col")]
  # Update the DataFrame in your environment
  assign(name, df)
}

# Tidyverse (dplyr) approach for cleaner code
library(dplyr)
for (name in df_names) {
  df <- get(name) %>%
    mutate(datetime = as.POSIXct(paste(date_col, time_col), format = "%Y-%m-%d %H:%M:%S")) %>%
    select(-date_col, -time_col)
  assign(name, df)
}

Important: Double-check that the format argument matches the actual structure of your date/time strings (e.g., "%m/%d/%Y" for month/day/year formats).

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