R语言：如何筛选任意列含指定值的行？支持数值与字符串

Hey there! Let's work through this problem together. First, let's recreate the dataset using your provided code to make sure we're on the same page:

set.seed(1234) 
m3 <- matrix(12:1,nrow=6,ncol=4) 
m4 <- as.data.frame(m3) 
m5 <- m4[sample(nrow(m4)),]

Solution 1: Base R (works for both numeric and string columns) #

We can use rowSums() combined with sapply() to check if any element in a row matches your target values (12, 9, 7). This approach works seamlessly for string data too—just adjust your target values to strings when needed.

# Define your target values (use strings like c("12", "9", "7") for string columns)
target_values <- c(12, 9, 7)

# Filter rows where ANY column contains a target value
filtered_m5 <- m5[rowSums(sapply(m5, function(col) col %in% target_values)) > 0, ]

Here's what's happening:

• sapply(m5, function(col) col %in% target_values) creates a logical matrix where each cell is TRUE if the element matches a target value, otherwise FALSE.
• rowSums() converts those logical values to 1s and 0s, then sums them per row. A sum greater than 0 means at least one element in the row is a match.

Solution 2: Alternative Base R with apply() #

If you prefer a more explicit row-wise check, use apply() with any():

filtered_m5 <- m5[apply(m5, 1, function(row) any(row %in% target_values)), ]

This iterates over each row and checks if any element in the row is in your target list—again, works for both numeric and string data.

Solution 3: Tidyverse (dplyr) for cleaner syntax #

If you're using the tidyverse, if_any() makes this super intuitive and readable:

library(dplyr)

filtered_m5 <- m5 %>%
  filter(if_any(everything(), ~ .x %in% target_values))

if_any(everything(), ...) tells dplyr to check every column, and keep rows where at least one column meets the condition (~ .x %in% target_values). This method also supports both numeric and string columns without any extra adjustments.

Testing with string columns #

To confirm this works for strings, just convert your dataset to string type and update the target values:

# Convert m5 to string columns
m5_str <- m5 %>% mutate(across(everything(), as.character))

# Use string targets
target_strings <- c("12", "9", "7")

# Same filter code works!
filtered_m5_str <- m5_str %>%
  filter(if_any(everything(), ~ .x %in% target_strings))

All these methods will give you the rows where any column contains your target values, and they're fully adaptable to string data as requested.

内容的提问来源于stack exchange，提问作者Mark