blib/lib/Text/FindLinks.pm

filter: Statementstmt 100% 35 / 35 Branchbran 83.3% 5 / 6 Conditioncond 100% 2 / 2 MC/DCmcdc 100% 1 / 1 Subroutinesub 100% 8 / 8 Podpod 100% 3 / 3 totaltotal 98.1% 54 / 55 CC 5 SCAR 7.7
CC
5
Cov
98%
decorate_link
4.0
markup_links
2.0
BEGIN
1.0
CRAP 2.2
↑ Summary
1 package Text::FindLinks;
2  
3 =encoding utf8
4  
5 =head1 NAME
6  
7 Text::FindLinks - Find and markup URLs in plain text
8  
9 =cut
10  
11 5 ▶use warnings;
Subroutine BEGIN: called CC=1 SCAR=0.0
12 5 ▶use strict;
Subroutine BEGIN: called CC=1 SCAR=0.0
13 5 ▶use Exporter;
Subroutine BEGIN: called CC=1 SCAR=0.0
14 5 ▶use Params::Validate qw/validate CODEREF/;
Subroutine BEGIN: called CC=1 SCAR=0.0
15  
16 5 our @ISA = 'Exporter';
17 5 our @EXPORT_OK = qw/find_links markup_links/;
18 5 our $VERSION = '0.04';
19  
20 =head1 SYNOPSIS
21  
22 use Text::FindLinks 'markup_links';
23 my $text = "Have you seen www.foo.com yet?";
24 # Have you seen <a href="http://www.foo.com">http://www.foo.com</a> yet?
25 print markup_links(text => $text);
26  
27 =head1 FUNCTIONS
28  
29 =head2 markup_links(text => ..., [handler => sub { ... }])
30  
31 Finds all URLs in the given text and replaces them using
32 the given handler. The handler gets passed three arguments:
33 the URL itself, all the text to the left from it and all the
34 text to the right. (The context is passed in case you would
35 like to keep some URLs untouched.) If no handler is given,
36 the default handler will be used that simply creates a link
37 to the URL and skips URLs already turned into links.
38  
39 =cut
40  
41 sub markup_links
42 12 ▶{
Subroutine markup_links: called CC=2 SCAR=6.9
Pod documented
43 12 validate(@_,
44 {
45 text => 1,
46 handler =>
47 {
48 type => CODEREF,
49 optional => 1,
50 },
51 });
52  
53 12 my %args = @_;
54 12 my $text = $args{'text'};
55 12 ▶ my $decorator = $args{'handler'} || \&decorate_link;
Condition100%2/2
$args{'handler'} || \&Text::FindLinks::decorate_link
l!l
93
MC/DC100%1/1
$args{'handler'} || \&Text::FindLinks::decorate_link
$args{'handler'}
Truth table100%2/2
A
A= $args{'handler'}
Aresult
11
00
56  
57 12 $text =~ s{(
58 (
59 (((https?)|(ftp))://) # either a schema...
60 | (www\.) # or the ‘www’ token
61 )
62 [^\s<]+ # anything except whitespace and ‘<’
63 (?<![,.;?!]) # must not end with given characters
64 )}
65 13 {&$decorator($1, $`, $')}gex;
66  
67 12 return $text;
68 }
69  
70 =head2 find_links(text => ...)
71  
72 Returns an array with all the URLs found in given text.
73 Just a simple wrapper around C<markup_links>, see the
74 sources.
75  
76 =cut
77  
78 sub find_links
79 1 ▶{
Subroutine find_links: called CC=1 SCAR=0.0
Pod documented
80 1 validate(@_, { text => 1 });
81 1 my %args = @_;
82 1 my @links;
83 markup_links(
84 text => $args{'text'},
85 1 ▶ handler => sub { push @links, shift });
Subroutine __ANON__: called CC=1 SCAR=0.0
86 1 return @links;
87 }
88  
89 =head2 decorate_link($url, $left, $right)
90  
91 Default URL handler that will be used if you don’t pass your
92 own to the C<markup_links> sub using the C<handler> attribute.
93 It turns an URL into a HTML link and skips URLs that are
94 apparently already links. Not exported.
95  
96 =cut
97  
98 sub decorate_link
99 3 ▶{
Subroutine decorate_link: called CC=4 SCAR=13.9
Pod documented
100 3 my ($url, $left, $right) = @_;
101  
102 # Skip already marked links.
103 3 ▶ return $url if ($left =~ /href=["']$/);
Branch: 1 branch
BranchTrueFalseTotal
if $left =~ /href=["']$/ 1 2 3
104 2 ▶ return $url if ($right =~ qr|^</a>|);
Branch: 1 branch
BranchTrueFalseTotal
if $right =~ qr"^</a>" 1 1 2
105  
106 1 my $label = $url;
107 1 ▶ $url = "http://$url" if ($url =~ /^www/i);
Branch: 1 branch
BranchTrueFalseTotal
if $url =~ /^www/i 1 0 1
108 1 return qq|<a href="$url">$label</a>|;
109 }
110  
111 =head1 BUGS
112  
113 The algorithm is extremely naive, a simple regex. It is almost
114 certain that some URLs will not be recognized and some things
115 that are not URLs will (to keep the balance). I’d be glad to
116 hear if there is some URL that misbehaves.
117  
118 L<http://rt.cpan.org/NoAuth/Bugs.html?Dist=Text-FindLinks>
119  
120 =head1 AUTHOR
121  
122 Tomáš Znamenáček, zoul@fleuron.cz
123  
124 L<http://github.com/zoul/Text-FindLinks>
125  
126 =head1 COPYRIGHT & LICENSE
127  
128 Copyright 2009 Tomáš Znamenáček
129  
130 This program is free software; you can redistribute it and/or modify it
131 under the terms of either: the GNU General Public License as published
132 by the Free Software Foundation; or the Artistic License.
133  
134 See L<http://dev.perl.org/licenses/> for more information.
135  
136 =cut
137  
138 5 1;
↑ Summary