hval: routines for attribute escaping

We'll use HTML attributes + anchor links to link to filenames
in coming commits.

diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm
index 53810b338a0cdc801a558b936e3c7c6f79dd72d2..95a0f7095b59f2a6683b3921f06cfb07b13939f0 100644 (file)
--- a/lib/PublicInbox/Hval.pm
+++ b/lib/PublicInbox/Hval.pm
@@ -9,8 +9,8 @@ use warnings;
 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/;
-
+our @EXPORT_OK = qw/ascii_html obfuscate_addrs to_filename src_escape
+               to_attr from_attr/;
 my $enc_ascii = find_encoding('us-ascii');
 
 sub new {
@@ -122,4 +122,35 @@ sub to_filename ($) {
        $s
 }
 
+# convert a filename (or any string) to HTML attribute
+
+my %ESCAPES = map { chr($_) => sprintf('::%02x', $_) } (0..255);
+$ESCAPES{'/'} = ':'; # common
+
+sub to_attr ($) {
+       my ($str) = @_;
+
+       # git would never do this to us:
+       return if index($str, '//') >= 0;
+
+       my $first = '';
+       if ($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;
+       $first . $str;
+}
+
+# reverse the result of to_attr
+sub from_attr ($) {
+       my ($str) = @_;
+       my $first = '';
+       if ($str =~ s/\AZ([a-f0-9]{2})//ms) {
+               $first = chr(hex($1));
+       }
+       $str =~ s!::([a-f0-9]{2})!chr(hex($1))!egms;
+       $str =~ tr!:!/!;
+       $first . $str;
+}
+
 1;

diff --git a/t/hval.t b/t/hval.t
index bfc9a856f50ba83886388f34e57ad1f3fa91ddbe..f93d6be6f0bd03df0fc04911348241b4a7fdc358 100644 (file)
--- a/t/hval.t
+++ b/t/hval.t
@@ -1,9 +1,9 @@
-# Copyright (C) 2017-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2017-2019 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 use strict;
 use warnings;
 use Test::More;
-use_ok 'PublicInbox::Hval';
+use_ok 'PublicInbox::Hval', qw(from_attr to_attr);
 
 my $ibx = {
        -no_obfuscate_re => qr/(?:example\.com)\z/i,
@@ -47,4 +47,14 @@ my $s = "\0\x07\n";
 PublicInbox::Hval::src_escape($s);
 is($s, "\\0\\a\n", 'src_escape works as intended');
 
+foreach my $s ('Hello/World.pm', 'Zcat', 'hello world.c', 'Eléanor', '$at') {
+       my $attr = to_attr($s);
+       is(from_attr($attr), $s, "$s => $attr => $s round trips");
+}
+
+{
+       my $bad = to_attr('foo//bar');
+       ok(!$bad, 'double-slash rejected');
+}
+
 done_testing();