use Digest::SHA qw/sha1_hex/;
my $SALT = rand;
-my $LINK_RE = qr{(\()?\b((?:ftps?|https?|nntps?|gopher)://
+my $LINK_RE = qr{([\('!])?\b((?:ftps?|https?|nntps?|gopher)://
[\@:\w\.-]+(?:/
(?:[a-z0-9\-\._~!\$\&\';\(\)\*\+,;=:@/%]*)
(?:\?[a-z0-9\-\._~!\$\&\';\(\)\*\+,;=:@/%]+)?
sub new { bless {}, $_[0] }
+# try to distinguish paired punctuation chars from the URL itself
+# Maybe other languages/formats can be supported here, too...
+my %pairs = (
+ "(" => qr/(\)[\.,;\+]?)\z/, # Markdown (,), Ruby (+) (, for arrays)
+ "'" => qr/('[\.,;\+]?)\z/, # Perl / Ruby
+ "!" => qr/(![\.,;\+]?)\z/, # Perl / Ruby
+);
+
sub linkify_1 {
$_[1] =~ s^$LINK_RE^
my $beg = $1 || '';
# '.', ',' or ';' to denote the end of a statement;
# assume the intent was to end the statement/sentence
# in English
- # Markdown compatibility:
- if ($beg eq '(') {
- if ($url =~ s/(\)[\.,;]?)\z//) {
+ if (defined(my $re = $pairs{$beg})) {
+ if ($url =~ s/$re//) {
$end = $1;
}
} elsif ($url =~ s/([\.,;])\z//) {
'Markdown-compatible end of sentence');
}
+# Perl and Ruby code compatibility
+{
+ my $l = PublicInbox::Linkify->new;
+ my $u = 'http://example.com/';
+ foreach my $q ("'%s'", '"%s"', 'q!%s!', 'q(%s)') {
+ # Perl
+ my $s = sprintf("my \$var = $q;", $u);
+ $s = $l->linkify_1($s);
+ $s = $l->linkify_2($s);
+ like($s, qr/>\Q$u\E</, "no quote($q) in URL");
+
+ # applies to Ruby, too
+ $s = sprintf("$q,", $u);
+ $s = $l->linkify_1($s);
+ $s = $l->linkify_2($s);
+ like($s, qr/>\Q$u\E</, "no quote($q) in URL array");
+ }
+}
+
done_testing();