How can I use the str_replace() function to replace matched patterns in a string with a new value?

R str_replace() and str_replace_all() are used to replace values of a string column based on matched patterns ( pattern matching with regex – regular expression), also used to replace with a specific string or character value. This function can be used on both DataFrame columns and a vector.

In order to use this str_replace() method, first, you need to load its library using library("stringr"). In case you don’t have this package, install it using install.packages("stringr"). The stringr package provides a set of functions to work with strings as easily as possible. 

1. str_replace() Syntax

Following is a syntax of str_replace() method from stringr package.

# Syntax of str_replace()
str_replace(string, pattern, replacement)
str_replace_all(string, pattern, replacement)

Let’s create an R dataframe and explore some examples using str_replace() method.

# Create DataFrame
df <- data.frame(id=c(1,2,3,4),
    address=c('Orange St','Anton Blvd','Jefferson Pkwy','Main St'))
print(df)

#Output
#  id        address
#1  1      Orange St
#2  2     Anton Blvd
#3  3 Jefferson Pkwy
#4  4        Main St

2. Use str_replace() to Replace Part of String with Another String

str_replace() method from stringr package is used to replace a part of a string on column with another string or replace column with pattern matching. The following example replaces string St with Street on column address. Here, df$address is a vector. Note that every column in a DataFrame is a vector.

If you wanted to replace NA with Empty String in R dataframe use methods from dplyr package.

# Replace String with anotehr String
library(stringr)
df$address <- str_replace(df$address, "St", "Street")
print(df)

Yields below output.

# Output
  id        address
1  1  Orange Street
2  2     Anton Blvd
3  3 Jefferson Pkwy
4  4    Main Street

3. str_replace_all() to Match on Multiple Strings

Use str_replace_all() method of stringr package to replace multiple string values at a time on a single column. The following example takes vector c() with mapping of values to be replaced on address column.

# Replace multiple strings at a time
rep_str = c('St'='Street','Blvd'='Boulevard','Pkwy'='Parkway')
df$address <- str_replace_all(df$address, rep_str)
print(df)

Yields below output

# Output
  id           address
1  1     Orange Street
2  2   Anton Boulevard
3  3 Jefferson Parkway
4  4       Main Street

4. Replace With Pattern Matching (regex)

Since these methods are used on vector, let’s create a R vector and replace values in it with pattern matching. Since every column in a DataFrame is a vector, you can also use pattern matching on DataFrame columns.

# Create Vector
numbers <- c("one", "two", "three")
num <- str_replace(numbers, "[oe]", "-")
print(num)

# Output
[1] "-ne"   "tw-"   "thr-e"

Now let’s use str_replace_all() on the same example.

# Create Vector
numbers <- c("one", "two", "three")
num2 <- str_replace_all(numbers, "[oe]", "-")
print(num2)

# Output
[1] "-n-"   "tw-"   "thr--"

5. Conclusion

In this article, you have learned str_replace() and str_replace_all() methods from sringr package are used to replace part of a string from a column to another string. Also, these methods are used to replace the string with another string based on pattern matching.

Related Articles

References