R String Functions — Advanced Text Manipulation

Learning Objectives

By the end of this tutorial, you will be able to:

• Use base R string functions for manipulation
• Apply stringr package for consistent string operations
• Write regular expressions for pattern matching
• Perform advanced text processing tasks

Base R String Functions

Character Vector Operations

# Length
nchar("hello")         # [1] 5
nchar(c("a", "bb", "ccc"))  # [1] 1 2 3

# Case conversion
toupper("hello")       # [1] "HELLO"
tolower("HELLO")       # [1] "hello"

# Substring
substr("hello world", 1, 5)  # [1] "hello"
substring("hello world", 7)  # [1] "world"

# Split
strsplit("a,b,c", ",")  # [[1]] "a" "b" "c"

# Concatenate
paste("hello", "world")       # [1] "hello world"
paste0("hello", "world")      # [1] "helloworld"
paste(c("a", "b", "c"), collapse = ", ")  # [1] "a, b, c"

Pattern Matching

# grep — find matches
grep("e", c("hello", "world", "test"))  # [1] 1 3

# grepl — logical
grepl("e", c("hello", "world", "test"))  # [1] TRUE FALSE TRUE

# sub — replace first
sub("o", "0", "hello")  # [1] "hell0"

# gsub — replace all
gsub("o", "0", "hello")  # [1] "hell0"

# regexpr — find position
regexpr("o", "hello")  # [1] 5

stringr Package

library(stringr)

Core Functions

# String length
str_length("hello")  # [1] 5

# Concatenation
str_c("hello", "world", sep = " ")  # [1] "hello world"

# Substring
str_sub("hello world", 1, 5)  # [1] "hello"
str_sub("hello world", -6, -1)  # [1] "world"

# Split
str_split("a,b,c", ",")
# [[1]]
# [1] "a" "b" "c"

# Trim
str_trim("  hello  ")  # [1] "hello"
str_pad("hi", 10, pad = " ")  # [1] "        hi"

# Duplicate
str_dup("ab", 3)  # [1] "ababab"

# Truncate
str_trunc("long string here", 10)  # [1] "long st..."

Pattern Matching

# Detect
str_detect(c("apple", "banana", "cherry"), "an")
# [1] FALSE TRUE FALSE

# Count
str_count("mississippi", "s")  # [1] 4

# Locate
str_locate("hello world", "world")
#      start end
# [1,]     7  11

# Extract
str_extract("Order #12345", "\\d+")  # [1] "12345"

# Extract all
str_extract_all("Order #12345 shipped #67890", "\\d+")
# [[1]]
# [1] "12345" "67890"

# Replace
str_replace("hello world", "world", "R")  # [1] "hello R"
str_replace_all("hello", "l", "L")  # [1] "heLLo"

# Match
str_match("2024-01-15", "(\\d{4})-(\\d{2})-(\\d{2})")
#      [,1]         [,2]  [,3]  [,4]
# [1,] "2024-01-15" "2024" "01"  "15"

# Start/End with
str_starts_with(c("hello", "world"), "h")  # [1] TRUE FALSE
str_ends_with(c("hello", "world"), "o")  # [1] TRUE TRUE

Regular Expressions

Basic Patterns

# Any character (except newline)
grepl("h.llo", "hello")  # [1] TRUE

# Start and end
grepl("^hello", "hello world")  # [1] TRUE
grepl("world$", "hello world")  # [1] TRUE

# Character classes
grepl("[aeiou]", "hello")  # [1] TRUE (has vowel)
grepl("[^aeiou]", "hello")  # [1] TRUE (has non-vowel)

# Quantifiers
grepl("lo+", "hello")  # [1] TRUE (one or more o)
grepl("lo*", "hell")  # [1] TRUE (zero or more o)
grepl("lo?", "hell")  # [1] TRUE (zero or one o)

# Escape special characters
grepl("\\.", "hello.world")  # [1] TRUE (literal dot)

Common Regex Patterns

Practical Examples

# Email validation
email <- "user@example.com"
grepl("^[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}$", email)
# [1] TRUE

# Phone number
phone <- "(555) 123-4567"
grepl("^\\(\\d{3}\\) \\d{3}-\\d{4}$", phone)
# [1] TRUE

# URL
url <- "https://www.example.com/path"
grepl("^https?://[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}", url)
# [1] TRUE

# Extract numbers
text <- "I have 3 cats and 5 dogs"
as.integer(regmatches(text, gregexpr("\\d+", text))[[1]])
# [1] 3 5

# Extract words
words <- "hello world foo bar"
str_extract_all(words, "\\b\\w+\\b")
# [[1]]
# [1] "hello" "world" "foo"   "bar"

Practical Examples

Example 1: Clean Data

library(stringr)

# Messy data
raw <- c("  Alice Smith  ", "BOB JONES", "charlie brown")

# Clean up
clean <- raw |>
  str_trim() |>
  str_to_title()

clean
# [1] "Alice Smith"    "Bob Jones"      "Charlie Brown"

Example 2: Parse CSV Line

library(stringr)

line <- "John,Doe,30,New York"

fields <- str_split(line, ",", simplify = TRUE)
data.frame(
  first = fields[1],
  last = fields[2],
  age = as.integer(fields[3]),
  city = fields[4]
)

Example 3: Extract Domain

library(stringr)

emails <- c("alice@gmail.com", "bob@yahoo.com", "charlie@company.org")

domains <- str_extract(emails, "(?<=@)[^.]+")
domains
# [1] "gmail"    "yahoo"    "company"

Example 4: Word Count

library(stringr)

count_words <- function(text) {
  str_count(text, "\\S+")
}

count_words("hello world")           # [1] 2
count_words("  one   two   three  ") # [1] 3

Common Mistakes

1. Forgetting escape characters

# Wrong
grepl(".", "hello")  # [1] TRUE (matches anything)

# Right
grepl("\\.", "hello.world")  # [1] TRUE (literal dot)

2. Case sensitivity

# Wrong
grepl("hello", "Hello World")  # [1] FALSE

# Right
grepl("hello", "Hello World", ignore.case = TRUE)  # [1] TRUE

3. Not handling NA

x <- c("hello", NA, "world")
str_to_upper(x)  # [1] "HELLO" NA      "WORLD"

# Better
x[!is.na(x)] <- str_to_upper(x[!is.na(x)])

Practice Exercises

Exercise 1: Email Extractor

Write a function that extracts all email addresses from a text.

Solution

library(stringr)

extract_emails <- function(text) {
  str_extract_all(text, "[a-zA-Z0-9._%+-]+@[a-zA-Z0-9.-]+\\.[a-zA-Z]{2,}")[[1]]
}

text <- "Contact alice@example.com or bob@yahoo.com for info"
extract_emails(text)
# [1] "alice@example.com" "bob@yahoo.com"

Exercise 2: Camel Case Converter

Write a function that converts snake_case to camelCase.

Solution

library(stringr)

to_camel <- function(snake) {
  parts <- str_split(snake, "_")[[1]]
  parts <- str_to_title(parts)
  paste0(parts[1], paste(parts[-1], collapse = ""))
}

to_camel("hello_world")       # [1] "helloWorld"
to_camel("my_variable_name")  # [1] "myVariableName"

Key Takeaways

• Base R has basic string functions — nchar(), substr(), paste(), strsplit()
• stringr provides consistency — all functions start with str_
• Regular expressions are powerful — learn \\d, \\w, \\s, [], ()
• str_extract() and str_match() pull out matches
• str_replace() and str_replace_all() substitute patterns
• Always handle NA in string operations
• Use ignore.case = TRUE for case-insensitive matching
• Practice regex with online tools like regex101.com

Next: Learn about R Date and Time — temporal data manipulation.