line |
stmt |
bran |
cond |
sub |
pod |
time |
code |
1
|
|
|
|
|
|
|
package Perl::Critic::Policy::RegularExpressions::ProhibitEscapedMetacharacters; |
2
|
|
|
|
|
|
|
|
3
|
40
|
|
|
40
|
|
27832
|
use 5.010001; |
|
40
|
|
|
|
|
192
|
|
4
|
40
|
|
|
40
|
|
280
|
use strict; |
|
40
|
|
|
|
|
129
|
|
|
40
|
|
|
|
|
862
|
|
5
|
40
|
|
|
40
|
|
246
|
use warnings; |
|
40
|
|
|
|
|
119
|
|
|
40
|
|
|
|
|
1100
|
|
6
|
|
|
|
|
|
|
|
7
|
40
|
|
|
40
|
|
269
|
use Readonly; |
|
40
|
|
|
|
|
143
|
|
|
40
|
|
|
|
|
2202
|
|
8
|
|
|
|
|
|
|
|
9
|
40
|
|
|
40
|
|
346
|
use Perl::Critic::Utils qw( :severities hashify ); |
|
40
|
|
|
|
|
141
|
|
|
40
|
|
|
|
|
2272
|
|
10
|
40
|
|
|
40
|
|
5825
|
use parent 'Perl::Critic::Policy'; |
|
40
|
|
|
|
|
142
|
|
|
40
|
|
|
|
|
265
|
|
11
|
|
|
|
|
|
|
|
12
|
|
|
|
|
|
|
our $VERSION = '1.150'; |
13
|
|
|
|
|
|
|
|
14
|
|
|
|
|
|
|
#----------------------------------------------------------------------------- |
15
|
|
|
|
|
|
|
|
16
|
|
|
|
|
|
|
Readonly::Scalar my $DESC => q{Use character classes for literal metachars instead of escapes}; |
17
|
|
|
|
|
|
|
Readonly::Scalar my $EXPL => [247]; |
18
|
|
|
|
|
|
|
|
19
|
|
|
|
|
|
|
Readonly::Hash my %REGEXP_METACHARS => hashify(split / /xms, '{ } ( ) . * + ? |'); |
20
|
|
|
|
|
|
|
|
21
|
|
|
|
|
|
|
#----------------------------------------------------------------------------- |
22
|
|
|
|
|
|
|
|
23
|
89
|
|
|
89
|
0
|
1674
|
sub supported_parameters { return qw() } |
24
|
74
|
|
|
74
|
1
|
352
|
sub default_severity { return $SEVERITY_LOWEST } |
25
|
84
|
|
|
84
|
1
|
419
|
sub default_themes { return qw( core pbp cosmetic ) } |
26
|
30
|
|
|
30
|
1
|
127
|
sub applies_to { return qw(PPI::Token::Regexp::Match |
27
|
|
|
|
|
|
|
PPI::Token::Regexp::Substitute |
28
|
|
|
|
|
|
|
PPI::Token::QuoteLike::Regexp) } |
29
|
|
|
|
|
|
|
|
30
|
|
|
|
|
|
|
#----------------------------------------------------------------------------- |
31
|
|
|
|
|
|
|
|
32
|
|
|
|
|
|
|
sub violates { |
33
|
0
|
|
|
0
|
1
|
|
my ( $self, $elem, $document ) = @_; |
34
|
|
|
|
|
|
|
|
35
|
|
|
|
|
|
|
# optimization: don't bother parsing the regexp if there are no escapes |
36
|
0
|
0
|
|
|
|
|
return if $elem !~ m/\\/xms; |
37
|
|
|
|
|
|
|
|
38
|
0
|
0
|
|
|
|
|
my $re = $document->ppix_regexp_from_element( $elem ) or return; |
39
|
0
|
0
|
|
|
|
|
$re->failures() and return; |
40
|
0
|
0
|
|
|
|
|
my $qr = $re->regular_expression() or return; |
41
|
|
|
|
|
|
|
|
42
|
0
|
0
|
|
|
|
|
my $exacts = $qr->find( 'PPIx::Regexp::Token::Literal' ) or return; |
43
|
0
|
|
|
|
|
|
foreach my $exact( @{ $exacts } ) { |
|
0
|
|
|
|
|
|
|
44
|
0
|
0
|
|
|
|
|
$exact->content() =~ m/ \\ ( . ) /xms or next; |
45
|
0
|
0
|
|
|
|
|
return $self->violation( $DESC, $EXPL, $elem ) if $REGEXP_METACHARS{$1}; |
46
|
|
|
|
|
|
|
} |
47
|
|
|
|
|
|
|
|
48
|
0
|
|
|
|
|
|
return; # OK |
49
|
|
|
|
|
|
|
} |
50
|
|
|
|
|
|
|
|
51
|
|
|
|
|
|
|
1; |
52
|
|
|
|
|
|
|
|
53
|
|
|
|
|
|
|
__END__ |
54
|
|
|
|
|
|
|
|
55
|
|
|
|
|
|
|
#----------------------------------------------------------------------------- |
56
|
|
|
|
|
|
|
|
57
|
|
|
|
|
|
|
=pod |
58
|
|
|
|
|
|
|
|
59
|
|
|
|
|
|
|
=for stopwords IPv4 |
60
|
|
|
|
|
|
|
|
61
|
|
|
|
|
|
|
=head1 NAME |
62
|
|
|
|
|
|
|
|
63
|
|
|
|
|
|
|
Perl::Critic::Policy::RegularExpressions::ProhibitEscapedMetacharacters - Use character classes for literal meta-characters instead of escapes. |
64
|
|
|
|
|
|
|
|
65
|
|
|
|
|
|
|
|
66
|
|
|
|
|
|
|
=head1 AFFILIATION |
67
|
|
|
|
|
|
|
|
68
|
|
|
|
|
|
|
This Policy is part of the core L<Perl::Critic|Perl::Critic> |
69
|
|
|
|
|
|
|
distribution. |
70
|
|
|
|
|
|
|
|
71
|
|
|
|
|
|
|
|
72
|
|
|
|
|
|
|
=head1 DESCRIPTION |
73
|
|
|
|
|
|
|
|
74
|
|
|
|
|
|
|
Ever heard of leaning toothpick syndrome? That comes from writing |
75
|
|
|
|
|
|
|
regular expressions that match on characters that are significant in |
76
|
|
|
|
|
|
|
regular expressions. For example, the expression to match four |
77
|
|
|
|
|
|
|
forward slashes looks like: |
78
|
|
|
|
|
|
|
|
79
|
|
|
|
|
|
|
m/\/\/\/\//; |
80
|
|
|
|
|
|
|
|
81
|
|
|
|
|
|
|
Well, this policy doesn't solve that problem (write it as C<m{////}> |
82
|
|
|
|
|
|
|
instead!) but solves a related one. As seen above, the escapes make |
83
|
|
|
|
|
|
|
the expression hard to parse visually. One solution is to use |
84
|
|
|
|
|
|
|
character classes. You see, inside of character classes, the only |
85
|
|
|
|
|
|
|
characters that are special are C<\>, C<]>, C<^> and C<->, so you |
86
|
|
|
|
|
|
|
don't need to escape the others. So instead of the following loose |
87
|
|
|
|
|
|
|
IPv4 address matcher: |
88
|
|
|
|
|
|
|
|
89
|
|
|
|
|
|
|
m/ \d+ \. \d+ \. \d+ \. \d+ /x; |
90
|
|
|
|
|
|
|
|
91
|
|
|
|
|
|
|
You could write: |
92
|
|
|
|
|
|
|
|
93
|
|
|
|
|
|
|
m/ \d+ [.] \d+ [.] \d+ [.] \d+ /x; |
94
|
|
|
|
|
|
|
|
95
|
|
|
|
|
|
|
which is certainly more readable, if less recognizable prior the |
96
|
|
|
|
|
|
|
publication of Perl Best Practices. (Of course, you should really use |
97
|
|
|
|
|
|
|
L<Regexp::Common::net|Regexp::Common::net> to match IPv4 addresses!) |
98
|
|
|
|
|
|
|
|
99
|
|
|
|
|
|
|
Specifically, this policy forbids backslashes immediately prior to the |
100
|
|
|
|
|
|
|
following characters: |
101
|
|
|
|
|
|
|
|
102
|
|
|
|
|
|
|
{ } ( ) . * + ? | # |
103
|
|
|
|
|
|
|
|
104
|
|
|
|
|
|
|
We make special exception for C<$> because C</[$]/> turns into |
105
|
|
|
|
|
|
|
C</[5.008006/> for Perl 5.8.6. We also make an exception for C<^> |
106
|
|
|
|
|
|
|
because it has special meaning (negation) in a character class. |
107
|
|
|
|
|
|
|
Finally, C<[> and C<]> are exempt, of course, because they are awkward |
108
|
|
|
|
|
|
|
to represent in character classes. |
109
|
|
|
|
|
|
|
|
110
|
|
|
|
|
|
|
Note that this policy does not forbid unnecessary escaping. So go |
111
|
|
|
|
|
|
|
ahead and (pointlessly) escape C<!> characters. |
112
|
|
|
|
|
|
|
|
113
|
|
|
|
|
|
|
|
114
|
|
|
|
|
|
|
=head1 CONFIGURATION |
115
|
|
|
|
|
|
|
|
116
|
|
|
|
|
|
|
This Policy is not configurable except for the standard options. |
117
|
|
|
|
|
|
|
|
118
|
|
|
|
|
|
|
|
119
|
|
|
|
|
|
|
=head1 BUGS |
120
|
|
|
|
|
|
|
|
121
|
|
|
|
|
|
|
Perl treats C<m/[#]/x> in unexpected ways. |
122
|
|
|
|
|
|
|
I think it's a bug in Perl itself, but am not 100% sure that I have |
123
|
|
|
|
|
|
|
not simply misunderstood... |
124
|
|
|
|
|
|
|
|
125
|
|
|
|
|
|
|
This part makes sense: |
126
|
|
|
|
|
|
|
|
127
|
|
|
|
|
|
|
"#f" =~ m/[#]f/x; # match |
128
|
|
|
|
|
|
|
"#f" =~ m/[#]a/x; # no match |
129
|
|
|
|
|
|
|
|
130
|
|
|
|
|
|
|
This doesn't: |
131
|
|
|
|
|
|
|
|
132
|
|
|
|
|
|
|
$qr = qr/f/; |
133
|
|
|
|
|
|
|
"#f" =~ m/[#]$qr/x; # no match |
134
|
|
|
|
|
|
|
|
135
|
|
|
|
|
|
|
Neither does this: |
136
|
|
|
|
|
|
|
|
137
|
|
|
|
|
|
|
print qr/[#]$qr/x; # yields '(?x-ism:[#]$qr |
138
|
|
|
|
|
|
|
)' |
139
|
|
|
|
|
|
|
|
140
|
|
|
|
|
|
|
=head1 CREDITS |
141
|
|
|
|
|
|
|
|
142
|
|
|
|
|
|
|
Initial development of this policy was supported by a grant from the |
143
|
|
|
|
|
|
|
Perl Foundation. |
144
|
|
|
|
|
|
|
|
145
|
|
|
|
|
|
|
|
146
|
|
|
|
|
|
|
=head1 AUTHOR |
147
|
|
|
|
|
|
|
|
148
|
|
|
|
|
|
|
Chris Dolan <cdolan@cpan.org> |
149
|
|
|
|
|
|
|
|
150
|
|
|
|
|
|
|
|
151
|
|
|
|
|
|
|
=head1 COPYRIGHT |
152
|
|
|
|
|
|
|
|
153
|
|
|
|
|
|
|
Copyright (c) 2007-2023 Chris Dolan |
154
|
|
|
|
|
|
|
|
155
|
|
|
|
|
|
|
This program is free software; you can redistribute it and/or modify |
156
|
|
|
|
|
|
|
it under the same terms as Perl itself. The full text of this license |
157
|
|
|
|
|
|
|
can be found in the LICENSE file included with this module |
158
|
|
|
|
|
|
|
|
159
|
|
|
|
|
|
|
=cut |
160
|
|
|
|
|
|
|
|
161
|
|
|
|
|
|
|
# Local Variables: |
162
|
|
|
|
|
|
|
# mode: cperl |
163
|
|
|
|
|
|
|
# cperl-indent-level: 4 |
164
|
|
|
|
|
|
|
# fill-column: 78 |
165
|
|
|
|
|
|
|
# indent-tabs-mode: nil |
166
|
|
|
|
|
|
|
# c-indentation-style: bsd |
167
|
|
|
|
|
|
|
# End: |
168
|
|
|
|
|
|
|
# ex: set ts=8 sts=4 sw=4 tw=78 ft=perl expandtab shiftround : |