Regular expressions in R in R Programming

Introduction

Regular expressions help you find patterns in text. They make searching and changing text easier and faster.

You want to check if a word or phrase appears in a sentence.

You need to extract phone numbers or emails from a list of text.

You want to replace all spaces with dashes in a string.

You want to split a sentence into words based on punctuation.

You want to validate if a text matches a specific format, like a date.

Syntax

R Programming

grepl(pattern, text)
sub(pattern, replacement, text)
gsub(pattern, replacement, text)
regexpr(pattern, text)
regmatches(text, regexpr(pattern, text))

pattern is the regular expression you want to use.

text is the string or vector of strings to search.

Examples

Checks if "cat" is in the text. Returns TRUE or FALSE.

R Programming

grepl("cat", "The cat is here")

Replaces the first "cat" with "dog".

R Programming

sub("cat", "dog", "The cat is here")

Replaces all spaces with dashes.

R Programming

gsub(" ", "-", "Hello world")

Extracts the first number sequence from the text.

R Programming

regmatches("abc123", regexpr("[0-9]+", "abc123"))

Sample Program

This program checks which words contain numbers, replaces numbers with '#', and extracts the first number found in each word.

R Programming

text <- c("apple", "banana", "cherry123", "date")
pattern <- "[0-9]+"

# Find which words have numbers
has_numbers <- grepl(pattern, text)
print(has_numbers)

# Replace numbers with '#'
replaced <- gsub(pattern, "#", text)
print(replaced)

# Extract numbers from words
numbers <- regmatches(text, regexpr(pattern, text))
print(numbers)

Important Notes

Regular expressions use special symbols like ., *, +, and [] to match patterns.

Use double backslashes \\ in R to escape special characters in patterns.

grepl() returns TRUE or FALSE, while regexpr() gives position info.

Summary

Regular expressions help find and change text patterns easily.

Use grepl() to check if a pattern exists.

Use sub() and gsub() to replace text based on patterns.

Practice

1. What does the grepl() function do in R when used with regular expressions?

easy

A. It checks if a pattern exists in a string and returns TRUE or FALSE.

B. It replaces all occurrences of a pattern in a string.

C. It extracts all matches of a pattern from a string.

D. It splits a string into parts based on a pattern.

Solution

1. Step 1: Understand the purpose of grepl()

   grepl() returns a logical value indicating if the pattern is found in the string.

2. Step 2: Compare with other functions

   Functions like gsub() replace text, and strsplit() splits strings, so they do not match grepl()'s behavior.

3. Final Answer:

   It checks if a pattern exists in a string and returns TRUE or FALSE. -> Option A

4. Quick Check:

   grepl() = logical check [OK]

Hint: Remember: grepl() returns TRUE/FALSE if pattern found [OK]

Common Mistakes:

• Confusing grepl() with gsub() which replaces text
• Thinking grepl() extracts matched text
• Assuming grepl() splits strings

2. Which of the following is the correct syntax to replace all occurrences of "cat" with "dog" in the string "cat and cat" using regular expressions in R?

easy

A. sub("cat", "dog", "cat and cat")

B. str_replace_all("cat and cat", "cat", "dog")

C. gsub("cat", "dog", "cat and cat")

D. replace("cat", "dog", "cat and cat")

Solution

1. Step 1: Identify function for replacing all occurrences

   gsub() replaces all matches, while sub() replaces only the first.

2. Step 2: Check syntax correctness

   gsub(pattern, replacement, string) is the correct order and usage in R.

3. Final Answer:

   gsub("cat", "dog", "cat and cat") -> Option C

4. Quick Check:

   Replace all = gsub() [OK]

Hint: Use gsub() to replace all matches, sub() only first [OK]

Common Mistakes:

• Using sub() when all occurrences need replacement
• Using non-existent replace() function in base R
• Confusing argument order in functions

3. What is the output of the following R code?

text <- c("apple", "banana", "apricot", "berry")
matches <- grepl("^a", text)
matches

medium

A. [FALSE, TRUE, FALSE, TRUE]

B. [FALSE, FALSE, FALSE, FALSE]

C. [TRUE, TRUE, TRUE, TRUE]

D. [TRUE, FALSE, TRUE, FALSE]

Solution

1. Step 1: Understand the pattern "^a"

   The caret (^) means the string must start with 'a'. So strings starting with 'a' match.

2. Step 2: Check each string in text

   "apple" and "apricot" start with 'a' (TRUE), "banana" and "berry" do not (FALSE).

3. Final Answer:

   [TRUE, FALSE, TRUE, FALSE] -> Option D

4. Quick Check:

   ^a matches start with 'a' = [TRUE, FALSE, TRUE, FALSE] [OK]

Hint: ^ means start of string; check which words start with 'a' [OK]

Common Mistakes:

• Ignoring ^ and matching 'a' anywhere
• Confusing TRUE/FALSE order in output
• Assuming grepl returns matched strings instead of logical

4. The following R code is intended to replace the first digit in a string with "#" but does not work as expected:

text <- "abc123"
result <- sub("[0-9]", "#", text)
print(result)

What is the problem and how to fix it?

medium

A. Pattern should be "\\d" instead of "[0-9]" for digits in R regex.

B. No problem; code works correctly and replaces first digit.

C. Use gsub() instead of sub() to replace all digits, not just first.

D. The replacement string "#" must be escaped as "\\#".

Solution

1. Step 1: Check the pattern and function

   [0-9] matches any digit; sub() replaces the first match only.

2. Step 2: Verify the output

   Replacing first digit '1' with '#' in "abc123" results in "abc#23", which is expected.

3. Final Answer:

   No problem; code works correctly and replaces first digit. -> Option B

4. Quick Check:

   sub() replaces first digit correctly [OK]

Hint: sub() replaces first match; [0-9] matches digits [OK]

Common Mistakes:

• Thinking [0-9] is invalid in R regex
• Confusing sub() with gsub() for all replacements
• Believing replacement '#' needs escaping

5. You have a vector of email addresses:

emails <- c("[e-mail]", "admin@sitecom", "[e-mail]")

Which regular expression pattern would correctly identify emails ending with ".com" using grepl()?

hard

A. "\\.com$"

B. ".com"

C. "com$"

D. "\\.com"

Solution

1. Step 1: Understand the pattern requirements

   We want to match emails ending exactly with ".com". The dot must be escaped as "\\." to match a literal dot, and "$" means end of string.

2. Step 2: Analyze each option

   "\\.com$" uses "\\.com$" which matches '.com' at the end. ".com" matches any character before 'com', not necessarily end. "com$" misses the dot. "\\.com" matches '.com' anywhere, not necessarily at end.

3. Final Answer:

   "\\.com$" -> Option A

4. Quick Check:

   Escape dot and use $ for end match [OK]

Hint: Escape dot with \\. and use $ to match string end [OK]

Common Mistakes:

• Not escaping dot, so matches any character
• Missing $ causing partial matches
• Confusing string end anchor with start anchor