regex-replace

Substitute for replace expressions.

Alt-*

This command functions like query-replace, but starts in regular expression mode.

pat1|pat2: matches either pat1 or pat2.
pat*: matches zero or more matches of pat.
pat+: matches one or more matches of pat.
pat?: matches zero or one matches of pat.
[abx]: matches any of the characters a, b, or x.
[^abx]: matches any but a, b, or x.
[a-z3]: matches a, b, c, ... z, or 3.
.: matches any character except newline.
( ): group patterns for +, *, ?, and |.
^: only matches at the beginning of a line.
$: only matches at the end of a line.
<#50>: means the character with ASCII code 50.
%: removes the special meaning from the following character, so that %$ matches only $.
!: marks the end of the match. The command does not change any characters that match the pattern after the exclamation point.

In the replacement text, #1 means substitute the part of the text that matched the first parenthesized pattern piece. For example, using regex-replace to replace "([A-Z][a-z]+)([.!?])" with "#2 ends #1" changes the text "Howard!" to "! ends Howard". #0 means to substitute the whole match. #U forces any following replacement text to uppercase, #L and #C to lowercase or capitalized. #E ends such case modifications; the remaining replacement text will be substituted as-is. #S substitutes the next alternative, when the search pattern consists of simple alternative bits of fixed text separated by |'s. Characters may be included by name in replacement text using the syntax #<Newline>, which substitutes a <Newline> character.

More info:

Replacing
Regular Expressions
Regular Expression Commands

Previous Up Next
refresh-files Command Reference regex-search