在R中按优先级从多列提取列表指定字符串的实现需求

Got it, let's work through this problem together. You need to pull strings from multiple columns in R based on a target list, following a strict priority order (check Column A first, then B, and so on), then store the first match in a new column. Here's how to implement this cleanly:

Step 1: Define Sample Data & Target List #

First, let's set up a sample dataset and target list to test with (matches your 2-column example):

# Sample data frame with 2 columns (A = higher priority)
df <- data.frame(
  A = c("apple", NA, "banana", "orange"),
  B = c("grape", "banana", NA, "apple"),
  stringsAsFactors = FALSE
)

# List of strings we want to match
target_list <- c("apple", "banana", "cherry")

Step 2: Tidyverse Approach (Recommended for Readability) #

Using dplyr and purrr makes this logic straightforward and easy to scale to more columns:

library(dplyr)
library(purrr)

# Define your column priority order (adjust this for more columns!)
priority_cols <- c("A", "B")

# Create the new matched column
df <- df %>%
  rowwise() %>%
  mutate(matched = {
    # Grab values from priority columns in order for the current row
    row_values <- c_across(all_of(priority_cols))
    # Find the first value that exists in our target list
    detect(row_values, ~ .x %in% target_list)
  }) %>%
  ungroup()

# Check the result
print(df)

How This Works: #

• priority_cols: This vector sets the order we check columns (A first, then B—add more columns here if needed)
• rowwise(): Ensures we process each row individually
• c_across(all_of(priority_cols)): Pulls the values from our priority columns for the current row
• detect(): Scans the row values in order and returns the first match from target_list (returns NA if no matches are found)
• ungroup(): Resets the data frame to its regular structure after row-wise processing

Step 3: Base R Alternative #

If you prefer not to use tidyverse packages, here's a base R solution:

# Define priority columns and target list (same as above)
priority_cols <- c("A", "B")
target_list <- c("apple", "banana", "cherry")

# Apply logic row-by-row
df$matched <- apply(df[, priority_cols], 1, function(row) {
  # Find the index of the first matching value in the row
  first_match_idx <- which(row %in% target_list)[1]
  # Return the match if it exists, else NA
  if (!is.na(first_match_idx)) row[first_match_idx] else NA
})

# View the result
print(df)

Expected Output #

Running either solution will give you this result, where matched takes the first valid match from the priority columns:

A B matched
1 apple grape apple
2 banana banana
3 banana banana
4 orange apple apple

内容的提问来源于stack exchange，提问作者Molia