view: do not redundantly obfuscate addresses

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

diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm
index 53810b338a0cdc801a558b936e3c7c6f79dd72d2..23233f0e7e979c8ff524e1864a46489a747512c9 100644 (file)
--- a/lib/PublicInbox/Hval.pm
+++ b/lib/PublicInbox/Hval.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2014-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2014-2020 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 #
 # represents a header value in various forms.  Used for HTML generation
@@ -9,33 +9,14 @@ 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 prurl mid_href/;
 my $enc_ascii = find_encoding('us-ascii');
 
-sub new {
-       my ($class, $raw, $href) = @_;
+# safe-ish acceptable filename pattern for portability
+our $FN = '[a-zA-Z0-9][a-zA-Z0-9_\-\.]+[a-zA-Z0-9]'; # needs \z anchor
 
-       # we never care about trailing whitespace
-       $raw =~ s/\s*\z//;
-       bless {
-               raw => $raw,
-               href => defined $href ? $href : $raw,
-       }, $class;
-}
-
-sub new_msgid {
-       my ($class, $msgid) = @_;
-       $class->new($msgid, mid_escape($msgid));
-}
-
-sub new_oneline {
-       my ($class, $raw) = @_;
-       $raw = '' unless defined $raw;
-       $raw =~ tr/\t\n / /s; # squeeze spaces
-       $raw =~ tr/\r//d; # kill CR
-       $class->new($raw);
-}
+sub mid_href { ascii_html(mid_escape($_[0])) }
 
 # some of these overrides are standard C escapes so they're
 # easy-to-understand when rendered.
@@ -63,31 +44,30 @@ 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/&apos;/&#39;/sg; # workaround https://bugs.debian.org/927409
        $_[0] =~ s/([\x7f\x00-\x1f])/$xhtml_map{$1}/sge;
        $_[0] = $enc_ascii->encode($_[0], Encode::HTMLCREF);
 }
 
 sub ascii_html {
        my ($s) = @_;
-       $s =~ s/\r\n/\n/sg; # fixup bad line endings
        $s =~ s/([<>&'"\x7f\x00-\x1f])/$xhtml_map{$1}/sge;
        $enc_ascii->encode($s, Encode::HTMLCREF);
 }
 
-sub as_html { ascii_html($_[0]->{raw}) }
-
-sub raw {
-       if (defined $_[1]) {
-               $_[0]->{raw} = $_[1];
-       } else {
-               $_[0]->{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;
 }
 
@@ -98,7 +78,7 @@ sub prurl {
 # However, &#8226; was chosen to make copy+paste errors more obvious
 sub obfuscate_addrs ($$;$) {
        my $ibx = $_[0];
-       my $repl = $_[2] || '&#8226;';
+       my $repl = $_[2] // '&#8226;';
        my $re = $ibx->{-no_obfuscate_re}; # regex of domains
        my $addrs = $ibx->{-no_obfuscate}; # { adddress => 1 }
        $_[1] =~ s/(([\w\.\+=\-]+)\@([\w\-]+\.[\w\.\-]+))/
@@ -122,4 +102,25 @@ sub to_filename ($) {
        $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 $first = '';
+       utf8::encode($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;
+}
+
 1;