www: use WwwStream for dumping thread and search views

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

diff --git a/lib/PublicInbox/Hval.pm b/lib/PublicInbox/Hval.pm
index 26a2d0bb018f631dbd117af8da36bbe9d86ac7e8..9ba780975cf2d5b1a5ca5188affbe3efa891acf2 100644 (file)
--- a/lib/PublicInbox/Hval.pm
+++ b/lib/PublicInbox/Hval.pm
@@ -1,35 +1,39 @@
-# Copyright (C) 2014, Eric Wong <normalperson@yhbt.net> and all contributors
+# Copyright (C) 2014-2015 all contributors <meta@public-inbox.org>
 # License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
 #
-# represents a header value in various forms
+# represents a header value in various forms.  Used for HTML generation
+# in our web interface(s)
 package PublicInbox::Hval;
 use strict;
 use warnings;
-use fields qw(raw -as_utf8);
 use Encode qw(find_encoding);
-use CGI qw(escapeHTML);
-use URI::Escape qw(uri_escape);
+use URI::Escape qw(uri_escape_utf8);
+use PublicInbox::MID qw/mid_clean/;
+use base qw/Exporter/;
+our @EXPORT_OK = qw/ascii_html/;
+
+# for user-generated content (UGC) which may have excessively long lines
+# and screw up rendering on some browsers.  This is the only CSS style
+# feature we use.
+use constant STYLE => '<style>pre{white-space:pre-wrap}</style>';
 
-my $enc_utf8 = find_encoding('utf8');
 my $enc_ascii = find_encoding('us-ascii');
-my $enc_mime = find_encoding('MIME-Header');
 
 sub new {
-       my ($class, $raw) = @_;
-       my $self = fields::new($class);
+       my ($class, $raw, $href) = @_;
 
-       # we never care about leading/trailing whitespace
-       $raw =~ s/\A\s*//;
+       # we never care about trailing whitespace
        $raw =~ s/\s*\z//;
-       $self->{raw} = $raw;
-       $self;
+       bless {
+               raw => $raw,
+               href => defined $href ? $href : $raw,
+       }, $class;
 }
 
 sub new_msgid {
-       my ($class, $raw) = @_;
-       $raw =~ s/\A<//;
-       $raw =~ s/>\z//;
-       $class->new($raw);
+       my ($class, $msgid, $no_compress) = @_;
+       $msgid = mid_clean($msgid);
+       $class->new($msgid, $msgid);
 }
 
 sub new_oneline {
@@ -40,14 +44,35 @@ sub new_oneline {
        $class->new($raw);
 }
 
-sub as_utf8 {
-       my ($self) = @_;
-       $self->{-as_utf8} ||= $enc_utf8->encode($self->{raw});
+my %xhtml_map = (
+       '"' => '&#34;',
+       '&' => '&#38;',
+       "'" => '&#39;',
+       '<' => '&lt;',
+       '>' => '&gt;',
+);
+
+sub ascii_html {
+       my ($s) = @_;
+       $s =~ s/\r\n/\n/sg; # fixup bad line endings
+       $s =~ s/([<>&'"])/$xhtml_map{$1}/ge;
+       $enc_ascii->encode($s, Encode::HTMLCREF);
 }
 
-sub ascii_html { $enc_ascii->encode(escapeHTML($_[0]), Encode::HTMLCREF) }
+sub as_html { ascii_html($_[0]->{raw}) }
+sub as_href { ascii_html(uri_escape_utf8($_[0]->{href})) }
+
+sub raw {
+       if (defined $_[1]) {
+               $_[0]->{raw} = $_[1];
+       } else {
+               $_[0]->{raw};
+       }
+}
 
-sub as_html { ascii_html($_[0]->as_utf8) }
-sub as_href { ascii_html(uri_escape($_[0]->as_utf8)) }
+sub prurl {
+       my ($env, $u) = @_;
+       index($u, '//') == 0 ? "$env->{'psgi.url_scheme'}:$u" : $u;
+}
 
 1;