[PATCH] linkify: implement Markdown link compatibility

[PATCH] linkify: implement Markdown link compatibility

[Eric Wong]

Although unescaped parentheses in URLs are technically allowed,
they are uncommon.  However, Markdown-like syntaxes are
unfortunately common for URLs, so we might as well support them.
---
 lib/PublicInbox/Linkify.pm | 15 +++++++++++----
 t/linkify.t                |  9 +++++++++
 2 files changed, 20 insertions(+), 4 deletions(-)

diff --git a/lib/PublicInbox/Linkify.pm b/lib/PublicInbox/Linkify.pm
index ea7fd71..cc0f7e3 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\-\._~!\$\&\';\(\)\*\+,;=:@/%]+)?
@@ -27,15 +27,22 @@ sub new { bless {}, shift }
 sub linkify_1 {
 	my ($self, $s) = @_;
 	$s =~ s!$LINK_RE!
-		my $url = $1;
+		my $beg = $1 || '';
+		my $url = $2;
 		my $end = '';
 
+		# Markdown compatibility:
+		if ($beg eq '(') {
+			$url =~ s/\)\z//;
+			$end = ')';
+		}
+
 		# it's fairly common to end URLs in messages with
 		# '.', ',' or ';' to denote the end of a statement;
 		# assume the intent was to end the statement/sentence
 		# in English
 		if ($url =~ s/([\.,;])\z//) {
-			$end = $1;
+			$end = $1 . $end;
 		}
 
 		# salt this, as this could be exploited to show
@@ -45,7 +52,7 @@ sub linkify_1 {
 		# only escape ampersands, others do not match LINK_RE
 		$url =~ s/&/&/g;
 		$self->{$key} = $url;
-		'PI-LINK-'. $key . $end;
+		$beg . 'PI-LINK-'. $key . $end;
 	!ge;
 	$s;
 }
diff --git a/t/linkify.t b/t/linkify.t
index 49cbbd6..a794c78 100644
--- a/t/linkify.t
+++ b/t/linkify.t
@@ -57,4 +57,13 @@ use PublicInbox::Linkify;
 	is($s, qq(hello <a\nhref="$u">$u</a> world), "root + fragment");
 }
 
+{
+	my $l = PublicInbox::Linkify->new;
+	my $u = 'http://example.com/';
+	my $s = "[markdown]($u)";
+	$s = $l->linkify_1($s);
+	$s = $l->linkify_2($s);
+	is($s, qq![markdown](<a\nhref="$u">$u</a>)!, 'markdown compatible');
+}
+
 done_testing();
-- 
EW

Re: [PATCH] linkify: implement Markdown link compatibility

[Eric Wong]

Oops, reverted :x  v2 coming.

[PATCH] linkify: implement Markdown link compatibility (again)

[Eric Wong]

Although unescaped parentheses in URLs are technically allowed,
they are uncommon.  However, Markdown-like syntaxes are
unfortunately common for URLs, so we might as well support them.

This fixes parentheses detection at sentence endings, as seen
in practice on emails.
---
 lib/PublicInbox/Linkify.pm | 14 ++++++++++----
 t/linkify.t                | 22 ++++++++++++++++++++++
 2 files changed, 32 insertions(+), 4 deletions(-)

diff --git a/lib/PublicInbox/Linkify.pm b/lib/PublicInbox/Linkify.pm
index ea7fd71..acd2a47 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\-\._~!\$\&\';\(\)\*\+,;=:@/%]+)?
@@ -27,14 +27,20 @@ sub new { bless {}, shift }
 sub linkify_1 {
 	my ($self, $s) = @_;
 	$s =~ s!$LINK_RE!
-		my $url = $1;
+		my $beg = $1 || '';
+		my $url = $2;
 		my $end = '';
 
 		# it's fairly common to end URLs in messages with
 		# '.', ',' or ';' to denote the end of a statement;
 		# assume the intent was to end the statement/sentence
 		# in English
-		if ($url =~ s/([\.,;])\z//) {
+		# Markdown compatibility:
+		if ($beg eq '(') {
+			if ($url =~ s/(\)[\.,;]?)\z//) {
+				$end = $1;
+			}
+		} elsif ($url =~ s/([\.,;])\z//) {
 			$end = $1;
 		}
 
@@ -45,7 +51,7 @@ sub linkify_1 {
 		# only escape ampersands, others do not match LINK_RE
 		$url =~ s/&/&/g;
 		$self->{$key} = $url;
-		'PI-LINK-'. $key . $end;
+		$beg . 'PI-LINK-'. $key . $end;
 	!ge;
 	$s;
 }
diff --git a/t/linkify.t b/t/linkify.t
index 49cbbd6..99acf17 100644
--- a/t/linkify.t
+++ b/t/linkify.t
@@ -57,4 +57,26 @@ use PublicInbox::Linkify;
 	is($s, qq(hello <a\nhref="$u">$u</a> world), "root + fragment");
 }
 
+# Markdown compatibility
+{
+	my $l = PublicInbox::Linkify->new;
+	my $u = 'http://example.com/';
+	my $s = "[markdown]($u)";
+	$s = $l->linkify_1($s);
+	$s = $l->linkify_2($s);
+	is($s, qq![markdown](<a\nhref="$u">$u</a>)!, 'Markdown-compatible');
+
+	$s = qq![markdown]($u "title")!;
+	$s = $l->linkify_1($s);
+	$s = $l->linkify_2($s);
+	is($s, qq![markdown](<a\nhref="$u">$u</a> "title")!,
+		'Markdown title compatible');
+
+	$s = qq![markdown]($u).!;
+	$s = $l->linkify_1($s);
+	$s = $l->linkify_2($s);
+	is($s, qq![markdown](<a\nhref="$u">$u</a>).!,
+		'Markdown-compatible end of sentence');
+}
+
 done_testing();
-- 
EW