Short hand character set

Digit \d [0-9] [[:digit:]]
Word \w [a-zA-Z0-9_]
White space \s [ \t\r\n]
Not digit \D [^0-9]
Not Word \W [^a-zA-Z0-9_]
Not White space \S [^ \t\r\n]
[[:alpha:]] [a-zA-Z]
[[:alnum:]] [a-zA-Z0-9]

Metacharacters

Any character except new line - wildcard | . Line return, new line | \r, \n, \r\n

Repetition mechacharacters

Preceding item, zero or more times *
Preceding item, one or more times +
Preceding item, zero or one time ?
Match lower case words ending in ‘ed’ \s[a-z]+ed\s

Quanlified Repetition

Start quantified repetion of preceding item {
End quantified repetion of preceding item }

Match numbers with 4 to 8 digits | \d{4,8} Match numbers with exact 4 digits | \d{4} Match numbers with 4 or more digits | \d{4,} {0,} same as * {1,} same as + {0,1} same as ?

Greedy Expression

  • Default match strategy is greedy
  • Standard repetition quantifiers are greedy
  • Expression tries to match the longest string

Lazy expression

Mark preceding quantifier lazy | ? Change match strategy is lazy, match as little as possible

*? +? ??

\d+\w+?\d+ > will match \w+ one time only

01_KJ_03_Mass_22_k.txt 01_KJ_07

Grouping

(in)?dependent will match independent, dependent

  • posix sed can’t use \d, gotta use [[:digit:]] echo 780-909-1111 | sed -E ‘s|([[:digit:]]{3})-([[:digit:]]{3}-[[:digit:]]{4})|(\1)-\2|’

Eager: peanut|peanutbutter will match only “peanut” in “peanutbutter” Greedy: peanut(butter)? will match whole word “peanutbutter” eager

Start and End

start of the string ^ or \A (in pcre regex - per compatible regular expression)
end of the string $ or \Z (pcre only)

Line mode & multiline

^[a-z]+$ will match only first line of the text, as carrier return is not matched as alpha characters.

Sed does not care, it’s multiple line mode by default

this: sdf

printf “780-909-1111\n870-243-2312\n” sed -E ‘s ([[:digit:]]{3})-([[:digit:]]{3}-[[:digit:]]{4}) (\1)-\2 ’

Word boundaries

start/end of the word \b

\b is not available in vim, pcre only

vim use <> to define word boundaries

Search: /\v<\w{3}s> will match word end with s and it is 4 characters Substitube: :%s//place/gc will replace any character is highlighted in that’s matched & replace with place global with prompt

Capturing group & backreferences

Backreference can be use right away in matching process

/\v(my) oh \1 will match “my oh my”

/\v<(ok|essm)><\w+><\/\1> will match following records

hello hekd

Following will match below 2 words & surround with * self-reliance reliance Search: ((self-)?reliance) Run: :%s//\1/gc

Supress group capturing

Disable auto capturing for the group ?:
i (?:love like) (.+) will not capture (love like) group as \1 
printf 'i love pizza\ni like pizza\n'  | perl -pe 's/i (love|like) (.+)/i \1/gi'

Look ahead & assertions (pcre)

printf '11-2-2025\n2025\n2024-11-3\n1029-1\n1988-03-09\n' | perl -ne 'print if /(?=\d{4}-\d{2}-\d{2})\d{4}/'

Some text for testing regex in vim

\d.\s[[:alpha:]]{3,4}\n

  1. a
  2. ab
  3. abc
  4. abcd
  5. abcde
  6. abcdef

\d+\w+\d+

01_KJ_03_Mass_22_k.txt 01_KJ_07

”.+”, “.+”

“Princess”, “King”, “Prince”, “Queen”

Page 266

Credit

This note is taken upon doing a online regex course in Linkedin Network from Kevin Skoglund