CODEWITHSUNDEEP
regexperl
Kamal Nayan
Kamal Nayan

Reputation: 1940

Need to fix the perl regex to handle multiple cases

I'm trying to handle some cases of strings with the regex: (.*note(?:'|")?\s*=>\s*)("|')?(.*?)\2(.*)

Strings:

There can be multiple spaces in start/end/in between. I need to capture " (or '), $note, ,or whatever is left after note_section. There can be # in beginning if this line is a comment, so, I've included .* in beginning.
Given regex is failing in case 3 as there is \2 as null.

Edit: Requirement is that I'm reading a file, and replacing the value of note with some tag say NOTETAG, and all other things around remain same, including inverted commas and spaces. For that,

e.g. note => "kamal" , will become note => "NOTETAG" , (notice we didnt ate , from last)

Upvotes: 0

Views: 52

Answers (2)

ikegami
ikegami

Reputation: 385764

s{
   \b
   note
   \s*
   =>
   \s*
   \K
   (?: (.*)
   |   '[^']*'
   |   "[^"]*"
   )
}{
   defined($1)
   ?  $1 =~ s{\$note\b}{"NOTETAG"}gr
   : '"NOTETAG"'
}exg;

Upvotes: 3

Michał Turczyn
Michał Turczyn

Reputation: 37367

Yuo could try (note\s*=>\s*(?:"|')?)[^'",]+

Explanation:

(...) - capturing group

note - match note literally

\s* - match zero or more of whitespaces

=> - match => literally

(?:..) - non-capturing group

"|' - alternation: match either ' or "

? - match preceding pattern zero or one time

[^'",]+ - negated character class - match one or more chraacters (due to + operator) other than ', ", ,

Demo

As a replacement use \1NOTETAG, where \1 means first capturing group

Upvotes: 1

Related Questions