hval: to_attr: support wide characters

[public-inbox.git] / lib / PublicInbox / Hval.pm

diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm
index c134e297280eeff734f0a2170eeac48386c395ec..39256ee01e0380bdfafd8c655265a82ceb796100 100644 (file)
--- a/lib/PublicInbox/Hval.pm
+++ b/lib/PublicInbox/Hval.pm
@@ -10,7 +10,7 @@ use Encode qw(find_encoding);
 use PublicInbox::MID qw/mid_clean mid_escape/;
 use base qw/Exporter/;
 our @EXPORT_OK = qw/ascii_html obfuscate_addrs to_filename src_escape
-               to_attr from_attr/;
+               to_attr from_attr prurl/;
 my $enc_ascii = find_encoding('us-ascii');
 
 # safe-ish acceptable filename pattern for portability
@@ -58,8 +58,11 @@ my %xhtml_map = (
 $xhtml_map{chr($_)} = sprintf('\\x%02x', $_) for (0..31);
 %xhtml_map = (%xhtml_map, %escape_sequence);
 
+# for post-processing the output of highlight.pm and perhaps other
+# highlighers in the future
 sub src_escape ($) {
        $_[0] =~ s/\r\n/\n/sg;
+       $_[0] =~ s/'/'/sg; # workaround https://bugs.debian.org/927409
        $_[0] =~ s/([\x7f\x00-\x1f])/$xhtml_map{$1}/sge;
        $_[0] = $enc_ascii->encode($_[0], Encode::HTMLCREF);
 }
@@ -81,8 +84,15 @@ sub raw {
        }
 }
 
-sub prurl {
+# returns a protocol-relative URL string
+sub prurl ($$) {
        my ($env, $u) = @_;
+       if (ref($u) eq 'ARRAY') {
+               my $h = $env->{HTTP_HOST} // $env->{SERVER_NAME};
+               my @host_match = grep(/\b\Q$h\E\b/, @$u);
+               $u = $host_match[0] // $u->[0];
+               # fall through to below:
+       }
        index($u, '//') == 0 ? "$env->{'psgi.url_scheme'}:$u" : $u;
 }
 
@@ -129,10 +139,12 @@ sub to_attr ($) {
        return if index($str, '//') >= 0;
 
        my $first = '';
+       utf8::encode($str); # to octets
        if ($str =~ s/\A([^A-Ya-z])//ms) { # start with a letter
                  $first = sprintf('Z%02x', ord($1));
        }
        $str =~ s/([^A-Za-z0-9_\.\-])/$ESCAPES{$1}/egms;
+       utf8::decode($str); # allow wide chars
        $first . $str;
 }
 
@@ -145,6 +157,7 @@ sub from_attr ($) {
        }
        $str =~ s!::([a-f0-9]{2})!chr(hex($1))!egms;
        $str =~ tr!:!/!;
+       utf8::decode($str);
        $first . $str;
 }