From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <e@80x24.org>
X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net
X-Spam-Level: 
X-Spam-ASN: AS62744 199.249.230.0/24
X-Spam-Status: No, score=-3.2 required=3.0 tests=AWL,BAYES_00,RCVD_IN_XBL,
	SPF_FAIL,SPF_HELO_FAIL shortcircuit=no autolearn=no autolearn_force=no
	version=3.4.2
Received: from 80x24.org (tor23.quintex.com [199.249.230.76])
	(using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits))
	(No client certificate requested)
	by dcvr.yhbt.net (Postfix) with ESMTPS id 491DC20305
	for <meta@public-inbox.org>; Thu, 18 Apr 2019 22:47:04 +0000 (UTC)
From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH] linkify: don't get confused by URLs in Perl code, at least
Date: Thu, 18 Apr 2019 22:47:01 +0000
Message-Id: <20190418224701.8135-1-e@80x24.org>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
List-Id: <meta.public-inbox.org>

The URLs at the top of WwwStream.pm weren't getting linkified
correctly.
---
 lib/PublicInbox/Linkify.pm | 15 +++++++++++----
 t/linkify.t                | 19 +++++++++++++++++++
 2 files changed, 30 insertions(+), 4 deletions(-)

diff --git a/lib/PublicInbox/Linkify.pm b/lib/PublicInbox/Linkify.pm
index aa472cd..71193bc 100644
--- a/lib/PublicInbox/Linkify.pm
+++ b/lib/PublicInbox/Linkify.pm
@@ -15,7 +15,7 @@ package PublicInbox::Linkify;
 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 @@ package PublicInbox::Linkify;
 
 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//) {
diff --git a/t/linkify.t b/t/linkify.t
index f0b3a6d..bef4ffd 100644
--- a/t/linkify.t
+++ b/t/linkify.t
@@ -98,4 +98,23 @@ use PublicInbox::Linkify;
 		'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();
-- 
EW