Regular Expressions

Flags can be combined

s/cat/Dog/gi replace every occurrence of cat (ignoring case, so it matches Cat, cAt, etc) with Dog (note that i doesn't affect the replacement string Dog)

For more info, :h :s_flags

^ start matching from beginning of a line
- /^This match This only at beginning of line
$ match pattern should terminate at end of a line
- /)$ match ) only at end of line
- /^$ match empty line
. match any single character, excluding new line
- /c.t match 'cat' or 'cot' or 'c2t' or 'c^t' but not 'cant'

For more info, :h pattern-atoms

*greedy match preceding character 0 or more times
- /abc* match 'ab' or 'abc' or 'abccc' or 'abcccccc' etc
\+ greedy match preceding character 1 or more times
- /abc\+ match 'abc' or 'abccc' but not 'ab'
\? match preceding character 0 or 1 times (\= can also be used)
- /abc\? match 'ab' or 'abc' but not 'abcc'
\{-} non-greedy match preceding character 0 or more times
- Consider this line of text 'This is a sample text'
- /h.\{-}s will match: 'his'
- /h.*s will match: 'his is a s'
- Read more on non-greedy matching
\{min,max} greedy match preceding character min to max times (including min and max)
- min or max can be left unspecified as they default to 0 and infinity respectively
- greedy match, tries to match as much as possible
\{-min,max} non-greedy match, tries to match as less as possible
\{number} match exactly with specified number
- /c\{5} match exactly 'ccccc'

For more info, :h pattern-overview

[abcde] match any of 'a' or 'b' or 'c' or 'd' or 'e' ONE time
- use [a-e] as shortform
[^abcde] match any character other than 'a' or 'b' or 'c' or 'd' or 'e'
- use [^a-e] as shortform
[aeiou] match vowel character
[^aeiou] match consonant character
\a matches alphabet character, short-cut for [a-zA-Z]
\A matches other than alphabet [^a-zA-Z]
\l matches lowercase alphabets [a-z]
\L matches other than lowercase alphabets [^a-z]
\u matches uppercase alphabets [A-Z]
\U matches other than uppercase alphabets [^A-Z]
\d matches digit character [0-9]
\D matches other than digit [^0-9]
\x matches hexademical character [0-9a-fA-F]
\X matches other than hexademical [^0-9a-fA-F]
\w matches any alphanumeric character or underscore [a-zA-Z0-9_]
\W match other than alphanumeric character or underscore [^a-zA-Z0-9_]
\s matches white-space characters space and tab
\S matches other than white-space characters
\t used in replacestring to insert a Tab character
\r used in replacestring to insert a newline character

For more info, :h /character-classes

\| allows to specify two or more patterns to be matched
- /min\|max match 'min' or 'max'
$pattern$ allows to group matched patterns and use special variables \1, \2, etc to represent them in same searchpattern and/or replacestring when using substitute command
- /hand$y\|ful$ match 'handy' or 'handful'
- /$\a$\1 match repeated alphabets

\<pattern Bind the searchpattern to necessarily be starting characters of a word
- /\<his matches 'his' and 'history' but not 'this'
pattern\> Bind the searchpattern to necessarily be ending characters of a word
- /his\> matches 'his' and 'this' but not 'history'
\<pattern\> Bind the searchpattern to exactly match whole word
- /\<his\> matches 'his' and not 'this' or 'history'

\v helps to avoid \ for pattern qualifiers, grouping pattern, etc
- /\vc{5} match exactly 'ccccc'
- /\vabc+ match 'abc' or 'abccc' but not 'ab'
- /\vabc? match 'ab' or 'abc' but not 'abcc'
- /\v<his> match whole word 'his', not 'this' or 'history'
- /\vmin|max match 'min' or 'max'
- /\vhand(y|ful) match 'handy' or 'handful'
- /\v(\a)\1 match repeated alphabets
- s/\v(\d+) (\d+)/\2 \1/ swap two numbers separated by space
\V no need to use \ when trying to match special characters
- /V^.*$ match the exact sequence ^.*$
\c case insensitive search
- /\cthis matches 'this', 'This', 'thiS', etc
\C case sensitive search
- /\Cthis match exactly 'this', not 'This', 'thiS', etc
\%V only inside visually selected area
- s/\%Vcat/dog/g replace 'cat' with 'dog' only in visually selected region

For more info

These are used in the replacestring section

Example:

:% s/\v(\a+)/\u\1/g will Capitalize all the words in current file (i.e only first character of word is capitalized)
:% s/\v(\a+)/\U\1/g will change to all letters to UPPERCASE in current file
:% s/\v(\w)_(\a+)/\1\u\2/g change variable_name to camelCase
- for ex: 'min_max' will change to 'minMax', 'array_sum' will change to 'arraySum' and so on
Changing case with regular expressions

One can also use other characters like #^$ instead of /

:% s#/project/adder/#/verilog/project/high_speed_adder/#g this avoids mess of having to use \/ for every / character

results matching ""