From c242a28351d3a9a39224d35031d71afc02993c7a Mon Sep 17 00:00:00 2001 From: Eric Wong Date: Sun, 19 Jan 2020 09:40:51 +0000 Subject: hval: to_attr: support wide characters We need to escape wide characters when making attribute names from filename-looking things in diffstats. --- lib/PublicInbox/Hval.pm | 3 +++ 1 file changed, 3 insertions(+) (limited to 'lib/PublicInbox/Hval.pm') diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm index 7e007027..39256ee0 100644 --- a/lib/PublicInbox/Hval.pm +++ b/lib/PublicInbox/Hval.pm @@ -139,10 +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 . $str; } @@ -155,6 +157,7 @@ sub from_attr ($) { } $str =~ s!::([a-f0-9]{2})!chr(hex($1))!egms; $str =~ tr!:!/!; + utf8::decode($str); $first . $str; } -- cgit v1.2.3-24-ge0c7