View on
Jeffrey Ryan Thalhammer > Perl-Critic-1.122 > Perl::Critic::Policy::RegularExpressions::ProhibitEscapedMetacharacters



Annotate this POD


View/Report Bugs
Module Version: 1.122   Source   Latest Release: Perl-Critic-1.127_01


Perl::Critic::Policy::RegularExpressions::ProhibitEscapedMetacharacters - Use character classes for literal meta-characters instead of escapes.


This Policy is part of the core Perl::Critic distribution.


Ever heard of leaning toothpick syndrome? That comes from writing regular expressions that match on characters that are significant in regular expressions. For example, the expression to match four forward slashes looks like:


Well, this policy doesn't solve that problem (write it as m{////} instead!) but solves a related one. As seen above, the escapes make the expression hard to parse visually. One solution is to use character classes. You see, inside of character classes, the only characters that are special are \, ], ^ and -, so you don't need to escape the others. So instead of the following loose IPv4 address matcher:

    m/ \d+ \. \d+ \. \d+ \. \d+ /x;

You could write:

    m/ \d+ [.] \d+ [.] \d+ [.] \d+ /x;

which is certainly more readable, if less recognizable prior the publication of Perl Best Practices. (Of course, you should really use Regexp::Common::net to match IPv4 addresses!)

Specifically, this policy forbids backslashes immediately prior to the following characters:

    { } ( ) . * + ? | #

We make special exception for $ because /[$]/ turns into /[5.008006/ for Perl 5.8.6. We also make an exception for ^ because it has special meaning (negation) in a character class. Finally, [ and ] are exempt, of course, because they are awkward to represent in character classes.

Note that this policy does not forbid unnecessary escaping. So go ahead and (pointlessly) escape ! characters.


This Policy is not configurable except for the standard options.


Perl treats m/[#]/x in unexpected ways. I think it's a bug in Perl itself, but am not 100% sure that I have not simply misunderstood...

This part makes sense:

    "#f" =~ m/[#]f/x;     # match
    "#f" =~ m/[#]a/x;     # no match

This doesn't:

    $qr  = qr/f/;
    "#f" =~ m/[#]$qr/x; # no match

Neither does this:

    print qr/[#]$qr/x;  # yields '(?x-ism:[#]$qr


Initial development of this policy was supported by a grant from the Perl Foundation.


Chris Dolan <>


Copyright (c) 2007-2011 Chris Dolan. Many rights reserved.

This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself. The full text of this license can be found in the LICENSE file included with this module

syntax highlighting: