Splitting Strings: A Beginner’s Guide to strsplit() in R

The strsplit() function in R splits elements of a character vector into a list of substrings based on a specified delimiter or regular expression pattern.

In the above figure, we are splitting a character vector “string” into three substrings based on the space in between them.

string <- ("Hello dystopian world")

strsplit(string, split = " ")

# Output:

# [[1]] 

# [1] "Hello" "dystopian" "world"

Syntax

strsplit(x, split, fixed = FALSE, perl = FALSE, useBytes = FALSE)

Parameters

Splitting into individual character vectors

You can split a string into individual character vectors by passing an empty string as a delimiter, which will separate each character, and then use the unlist() function to convert the list into a vector.

In the above figure, we split a single string into individual characters separated by an empty string.

string <- "hello"

list <- strsplit(string, split = "")

char_vector <- unlist(list)

print(char_vector)

# Output

# [1] "h" "e" "l" "l" "o"

Multiple strings

This function is not limited to a single string; you can use it to split the multiple strings.

# Multiple strings
strings <- c("one,two,three", "four,five,six")

# Split each string by ','
splitted <- strsplit(strings, split = ",")

# Print the result
print(splitted)

# Output:

# [[1]] 
# [1] "one" "two" "three" 

# [[2]] 
# [1] "four" "five" "six"

Passing delimiter

You can pass a delimiter (a symbol or special character) that separates the words or text in the data.

The above figure illustrates how we split a string based on the “&” delimiter.

string <- "Hello&dystopian&world"

strsplit(string, split = "&")

# Output:

# [[1]]
# [1] "Hello" "dystopian" "world"

Regular expression

Regular expressions are a compact and flexible way of describing patterns in strings. You can provide the specific regular expression you need to split the string.

In the above figure, we split a string based on regular expression as a separator. Basically, we split a string where R found numerical values.

string <- ("Hello19dystpoian21world")

strsplit(string, split = "[0-9]+")

# Output:
# [[1]] 
# [1] "Hello" "dystpoian" "world"

Passing fixed = TRUE

If you set fixed = TRUE, it tells R to treat the split pattern as a fixed string, not as a regular expression.

fruits <- c("apple, banana, cherry", "orange, peach, grape")

split_fruits <- strsplit(fruits, ", ", fixed = TRUE)

print(split_fruits)

# Output
# [[1]] 
# [1] "apple" "banana" "cherry" 

# [[2]] 
# [1] "orange" "peach" "grape"

Handling NA

When it comes to NA values, the strsplit() function handles NA values gracefully.

When an input vector contains NA, this function includes NA in the output list at the corresponding position without causing errors or altering the structure of the result.

input_with_na <- c("data, info", NA, "red,blue")

output_including_na <- strsplit(input_with_na, ",")

print(output_including_na)

# Output:
# [[1]]
# [1] "data" " info"

# [[2]]
# [1] NA

# [[3]]
# [1] "red" "blue"

Perl-Compatible Regular Expressions

If you want to use a Perl-compatible regular expression, you can do it by passing perl=TRUE in the argument.

complex_string <- "audi123bmw456bugatti"

perl_pattern_output <- strsplit(complex_string, "\\d+", perl = TRUE)

print(perl_pattern_output)

# Output
# [[1]]
# [1] "audi" "bmw" "bugatti"

In this code, the Perl-compatible regex \\d+ matches one or more digits, splitting the string at numeric sequences.

Byte-Level splitting

For non-standard encodings, you might have to split the string at the byte level.

non_standard_string <- "café"

# Splitting the string byte-by-byte
byte_split <- strsplit(non_standard_string, "", useBytes = TRUE)

print(byte_split)

# Output
# [[1]]
# [1] "c" "a" "f" "\xc3" "\xa9"

On most UTF-8 systems, the output looks like above program. It helps manage non-UTF-8 encodings.

That’s all!