R语言：如何获取DataFrame中已剔除常量列的列名及对应值

Got it, let's tackle this problem! You've already removed constant columns from your DataFrame and have their names stored as unmatched_columns—now you want to pair those names with their constant values. Here are a couple of straightforward, practical ways to do this:

方法1：基于已有列名列表提取值 #

If you already have the list of constant column names (unmatched_columns), you can use sapply() to iterate over these columns and pull out their single constant value. This returns a named vector where each name maps to its column's constant:

# 假设你的DataFrame名为df，unmatched_columns是常量列名的向量
constant_values <- sapply(df[unmatched_columns], function(col) {
  # 常量列只有一个唯一值，直接取第一个即可
  unique(col)[1]
})

# 查看结果
constant_values

For example, if unmatched_columns is c("fixed_id", "default_status"), the output might look like:

fixed_id default_status 
         1       "inactive" 

If you prefer a tabular format (easier to read or export), convert it to a data frame:

constant_df <- data.frame(
  column_name = names(constant_values),
  constant_value = as.vector(constant_values),
  stringsAsFactors = FALSE
)

方法2：先识别常量列再获取列名和值 #

If you ever need to re-identify constant columns and grab their values in one go (instead of starting from unmatched_columns), use this approach:

# 先筛选出所有常量列（唯一值数量为1的列）
constant_cols <- sapply(df, function(col) length(unique(col)) == 1)

# 提取这些列的名称和对应值
constant_values <- sapply(df[, constant_cols], function(col) unique(col)[1])

This works seamlessly for all data types—numeric, character, factor, etc.—since unique() handles them consistently.

小技巧 #

• If you use the dplyr package, you can simplify value extraction with first() (since every entry in the column is identical):

  library(dplyr)
  constant_values <- sapply(df[unmatched_columns], first)
  

• For factor columns, unique(col)[1] returns the human-readable factor level, not the underlying integer code—exactly what you'd want for clarity.

内容的提问来源于stack exchange，提问作者Oumab10