From 348f6b2b271d3562509420070bd94d6ac00619dc Mon Sep 17 00:00:00 2001 From: Eric Wong Date: Thu, 18 Apr 2019 22:47:01 +0000 Subject: linkify: don't get confused by URLs in Perl code, at least The URLs at the top of WwwStream.pm weren't getting linkified correctly. --- lib/PublicInbox/Linkify.pm | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'lib') diff --git a/lib/PublicInbox/Linkify.pm b/lib/PublicInbox/Linkify.pm index aa472cdb..71193bc2 100644 --- a/lib/PublicInbox/Linkify.pm +++ b/lib/PublicInbox/Linkify.pm @@ -15,7 +15,7 @@ use warnings; 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\-\._~!\$\&\';\(\)\*\+,;=:@/%]+)? @@ -25,6 +25,14 @@ my $LINK_RE = qr{(\()?\b((?:ftps?|https?|nntps?|gopher):// sub new { bless {}, $_[0] } +# try to distinguish paired punctuation chars from the URL itself +# Maybe other languages/formats can be supported here, too... +my %pairs = ( + "(" => qr/(\)[\.,;\+]?)\z/, # Markdown (,), Ruby (+) (, for arrays) + "'" => qr/('[\.,;\+]?)\z/, # Perl / Ruby + "!" => qr/(![\.,;\+]?)\z/, # Perl / Ruby +); + sub linkify_1 { $_[1] =~ s^$LINK_RE^ my $beg = $1 || ''; @@ -35,9 +43,8 @@ sub linkify_1 { # '.', ',' or ';' to denote the end of a statement; # assume the intent was to end the statement/sentence # in English - # Markdown compatibility: - if ($beg eq '(') { - if ($url =~ s/(\)[\.,;]?)\z//) { + if (defined(my $re = $pairs{$beg})) { + if ($url =~ s/$re//) { $end = $1; } } elsif ($url =~ s/([\.,;])\z//) { -- cgit v1.2.3-24-ge0c7