An Evening with Regex

Setup

library(tidyverse)

1. College majors

Find all college majors in the data that contain the words “DATA” or “STATISTICS” in the name of the major.

# Read in the data
df = read.csv("majors-list.csv")

# Only need one column
majors <- df |> 
  select(Major)

# Extract all majors that contain DATA or STATISTICS
majors |> 
  filter(str_detect(Major, "\\b(DATA|STATISTICS)\\b"))

                                          Major
1 MANAGEMENT INFORMATION SYSTEMS AND STATISTICS
2      COMPUTER PROGRAMMING AND DATA PROCESSING
3               STATISTICS AND DECISION SCIENCE

2. Proper fruits

Convert a messy string of fruits into a useful character vector!

Note: I interpreted this question to mean “Given this raw string, use regex to convert it into a usable character vector”, so I’m recreating the string as authentically as I can, and then breaking it down into the character vector.

# Create the fruit string!
fruit_string <- "[1] \"bell pepper\"  \"bilberry\"     \"blackberry\"
\"blood orange\"[5] \"blueberry\"    \"cantaloupe\"   \"chili pepper\"
\"cloudberry\"  [9] \"elderberry\"   \"lime\"         \"lychee\"       
\"mulberry\"    [13] \"olive\"        \"salal berry\""

# Reduce the string to each fruit

# Match words with optional spaces between literal \" bookends
pattern <-"(?<=\")(\\b\\w+(\\s\\w+)?\\b)(?=\")"

# Extract the list
fruits <- str_extract_all(fruit_string, pattern)[[1]]

# Turn into a vector
fruits <- unlist(fruits)

print(fruits)

 [1] "bell pepper"  "bilberry"     "blackberry"   "blood orange" "blueberry"   
 [6] "cantaloupe"   "chili pepper" "cloudberry"   "elderberry"   "lime"        
[11] "lychee"       "mulberry"     "olive"        "salal berry" 

3. Expressions

Explain what each regex expression is doing in plain language.

3.1 Expression 1

The full expression matches any character besides a line break followed by a literal \1\1

• (.)\1\1
  • (.)Establish capture group 1 as; Capture any character except a line break
  • \ Match a literal backslash
  • 1 Match a literal 1
  • \ Match a literal backslash
  • 1 Match a literal 1

pattern_1 <- "(.)\1\1"

string_1 <- "a\1\1"

str_view(string_1, pattern_1)

# Suppressed output so file could be rendered

3.2 Expression 2

The full expression matches a double quote, any two characters as capture groups, capture group 2, capture group 1, then a double quote

• "(.)(.)\\2\\1"
  • " Match a double quote
  • (.) Establish capture group 1 as; capture any character except a line break
  • (.) Establish capture group 2 as; capture any character except a line break
  • \\2 Match capture group 2
  • \\1 Match capture group 1
  • " Match a literal double quote

pattern_2 <- '"(.)(.)\\2\\1"'

string_2 <- c('"aaaa"', '"abba"')

str_view(string_2, pattern_2)

[1] │ <"aaaa">
[2] │ <"abba">

3.3 Expression 3

The full expression matches any two characters, a literal backslash, and a 1

• (..)\1
• (..) Match any two characters except line breaks
• \ Match a literal backslash
• 1 Match a one

pattern_3 <- '(..)\1'

string_3 <- c('aa\1', 'X!\1')

str_view(string_3, pattern_3)

# Had to suppress output so file could be rendered

3.4 Expression 4

The full expression matches a double quote, any character as capture group 1, any character, capture group 1, any character, capture group 1, then a double quote

• "(.).\\1.\\1"
• " Match a literal double quote
• (.) Establish capture group 1 as any character except line break
• . Match any character except line break
• \\1 Match capture group 1
• . Match any character
• \\1 Match capture group 1
• " Match a literal double quote

pattern_4 <- '"(.).\\1.\\1"'

string_4 <- c('"aXa!a"', '"2Z2$2"')

str_view(string_4, pattern_4)

[1] │ <"aXa!a">
[2] │ <"2Z2$2">

3.5 Expression 5

This expression matches a literal double quote, any three characters and establishes them as capture groups, any character any amount of times, capture groups 3, 2, and 1 respectively, and a literal double quote.

• "(.)(.)(.).*\\3\\2\\1"

pattern_5 <- '"(.)(.)(.).*\\3\\2\\1"'

string_5 <- c('"ABCXXXXXXXXCBA"', '"AAAXAAA"')

str_view(string_5, pattern_5)

[1] │ <"ABCXXXXXXXXCBA">
[2] │ <"AAAXAAA">

4. More Expressions

Create regular expressions to match words.

# Start and end with the same character

start_end_pattern = "\\b(\\w)\\w*\\1\\b"

start_end_words = c("racecar", "radar", "abba")

str_view(start_end_words, start_end_pattern)

[1] │ <racecar>
[2] │ <radar>
[3] │ <abba>

# Contain repeated pairs of letters

repeat_letters_pattern = "\\b\\w*(.)\\1\\w*\\b" 

repeat_letters_words = c("abba",
                         "apple",
                         "pattern",
                         "banana" # should not be captured
                         )

str_view(repeat_letters_words, repeat_letters_pattern)

[1] │ <abba>
[2] │ <apple>
[3] │ <pattern>