Is there a better way to replace multiple patterns inside of a string?

December 12, 2023

Whenever I need to do text replacement, it’s rare that I only have to replace one pattern. I often find myself stringing together multiple str_replace_all() calls, one after another. The example code below feels very repetitive. Is there a way to do this with only one call to str_replace_all()?

library(dplyr)
library(stringr)

text <- r'(mother, father, grandmother, and grandfather)'

# is there a better way to do this?
text %>% 
  str_replace_all(
    pattern = r'(\bmother\b)',
    replacement = 'mom'
  ) %>% 
  str_replace_all(
    pattern = r'(\bfather\b)',
    replacement = 'dad'
  ) %>% 
  str_replace_all(
    pattern = 'grandmother',
    replacement = 'grandma'
  ) %>% 
  str_replace_all(
    pattern = 'grandfather',
    replacement = 'grandpa'
  ) %>% 
  str_replace_all(
    pattern = r'(\band\b)',
    replacement = '&'
  )
#> [1] "mom, dad, grandma, & grandpa"

^{Created on 2023-12-12 with reprex v2.0.2}

I’m imagining something with a lookup table.

df_lookup <-
  tribble(
    ~pattern,           ~replacement,
    r'(\bmother\b)',    'mom',
    r'(\bfather\b)',    'dad',
    'grandmother',      'grandma',
    'grandfather',      'grandpa',
    r'(\band\b)',       '&'
  )

>Solution :

You can perform multiple str_replace as follows:

str_replace_all(text, 
    c(
        r'(\bmother\b)' = "mom", 
         r'(\bfather\b)' = "dad", 
        'grandmother' = "grandma", 
        'grandfather' = "grandpa",
        r'(\band\b)' = "&")
    )