R String Functions

R provides a rich set of built-in string functions for searching, replacing, splitting, and formatting text. These functions process entire character vectors at once, making them ideal for cleaning and transforming text columns in a dataset.

Searching Within Strings

# grepl() — returns TRUE/FALSE for each element
emails <- c("alice@gmail.com","bob@yahoo.com","cena@gmail.com","dev@hotmail.com")

grepl("@gmail", emails)
# TRUE FALSE TRUE FALSE

# Filter: get only Gmail addresses
gmail <- emails[grepl("@gmail", emails)]
# "alice@gmail.com" "cena@gmail.com"

# grep() — returns the positions (indices) of matches
grep("@gmail", emails)   # 1 3  (positions 1 and 3)

# grep() with value=TRUE returns the actual strings
grep("@gmail", emails, value=TRUE)
# "alice@gmail.com" "cena@gmail.com"

Finding Position of a Pattern

text <- "Learning R is fun and rewarding"

regexpr("R", text)         # 10 (first occurrence)
gregexpr("r", text, ignore.case=TRUE)  # all occurrences

Replacing Text

# sub() — replace FIRST match only
sub("o", "0", "too cool for school")
# "t0o cool for school"

# gsub() — replace ALL matches
gsub("o", "0", "too cool for school")
# "t00 c00l f0r sch00l"

# Practical: clean phone numbers
phones <- c("(91) 98765-43210", "(91) 87654-32109")
clean_phones <- gsub("[^0-9]", "", phones)  # keep digits only
print(clean_phones)
# "919876543210" "918765432109"

Splitting Strings

csv_row <- "Priya,28,Mumbai,85000"
strsplit(csv_row, ",")
# [["Priya", "28", "Mumbai", "85000"]]

# Split a sentence into words
sentence <- "R is great for data analysis"
strsplit(sentence, " ")[[1]]
# "R" "is" "great" "for" "data" "analysis"

# Split multiple strings at once
data <- c("Jan-2024", "Feb-2024", "Mar-2024")
strsplit(data, "-")
# [["Jan","2024"], ["Feb","2024"], ["Mar","2024"]]

Formatting Numbers as Strings

formatC(12345.678, format="f", digits=2)  # "12345.68"
formatC(12345,     format="d", big.mark=",")  # "12,345"
format(0.00123, scientific=TRUE)          # "1.23e-03"
format(12345678, big.mark=",", scientific=FALSE)  # "12,345,678"

Padding Strings

formatC("R", width=10)          # "         R"  (right-aligned)
formatC("R", width=10, flag="-")# "R         "  (left-aligned)
formatC(42, width=6, flag="0")  # "000042"       (zero-padded)

# Practical: create padded IDs
ids <- 1:5
formatC(ids, width=4, flag="0")
# "0001" "0002" "0003" "0004" "0005"

String Manipulation Cheat Sheet

Function         Purpose                        Example
─────────────────────────────────────────────────────────────────
nchar(x)         Count characters               nchar("hello") = 5
toupper/lower    Case conversion                 toupper("hi") = "HI"
trimws(x)        Remove whitespace              trimws("  hi  ") = "hi"
substr(x,s,e)    Extract substring              substr("hello",1,3) = "hel"
paste(...)       Combine strings                paste("a","b") = "a b"
paste0(...)      Combine without sep            paste0("a","b") = "ab"
gsub(p,r,x)      Replace all matches            gsub("a","@","banana")
sub(p,r,x)       Replace first match            sub("a","@","banana")
grepl(p,x)       TRUE if pattern found          grepl("@","email@x")
grep(p,x)        Indices of matches             grep("@", emails)
strsplit(x,s)    Split by separator             strsplit("a,b",",")
sprintf(fmt,...) Format string                  sprintf("%.2f", 3.14)

Practical: Clean a Messy Name Column

raw_names <- c("  ALICE sharma ", "BOB   Kumar", "cena PATEL  ")

clean_names <- raw_names |>
  trimws() |>                                          # remove whitespace
  tolower() |>                                         # all lowercase
  gsub("\\s+", " ", x = _) |>                        # collapse spaces
  gsub("(\\w)(\\w*)", "\\U\\1\\L\\2", x=_, perl=TRUE) # title case

print(clean_names)
# "Alice Sharma" "Bob Kumar" "Cena Patel"

String functions handle the messiest real-world data cleaning tasks. Survey responses, imported CSVs, scraped web data, and database fields all require string cleaning before analysis. These functions give you the tools to handle most situations without writing complex custom code.