package PublicInbox::View;
use strict;
use warnings;
-use CGI qw/escapeHTML escape/;
-use Encode qw/decode encode/;
+use URI::Escape qw/uri_escape/;
+use CGI qw/escapeHTML/;
+use Encode qw/find_encoding/;
use Encode::MIME::Header;
+use Email::MIME::ContentType qw/parse_content_type/;
+use constant MAX_INLINE_QUOTED => 5;
+use constant MAX_TRUNC_LEN => 72;
-# only one public function:
+my $enc_utf8 = find_encoding('utf8');
+my $enc_ascii = find_encoding('us-ascii');
+my $enc_mime = find_encoding('MIME-Header');
+
+# public functions:
sub as_html {
- my ($class, $mime) = @_;
+ my ($class, $mime, $full_pfx) = @_;
headers_to_html_header($mime) .
- multipart_text_as_html($mime) .
- "</pre>\n";
+ multipart_text_as_html($mime, $full_pfx) .
+ '</pre></body></html>';
+}
+
+sub as_feed_entry {
+ my ($class, $mime, $full_pfx) = @_;
+
+ "<pre>" . multipart_text_as_html($mime, $full_pfx) . "</pre>";
}
+
# only private functions below.
+sub enc_for {
+ my ($ct) = @_;
+ defined $ct or return $enc_utf8;
+ my $ct_parsed = parse_content_type($ct);
+ if ($ct_parsed) {
+ if (my $charset = $ct_parsed->{attributes}->{charset}) {
+ my $enc = find_encoding($charset);
+ return $enc if $enc;
+ }
+ }
+ $enc_utf8;
+}
+
sub multipart_text_as_html {
- my ($mime) = @_;
+ my ($mime, $full_pfx) = @_;
my $rv = "";
my $part_nr = 0;
+ my $enc_msg = enc_for($mime->header("Content-Type"));
# scan through all parts, looking for displayable text
$mime->walk_parts(sub {
my ($part) = @_;
return if $part->subparts; # walk_parts already recurses
+ my $enc = enc_for($part->content_type) || $enc_msg || $enc_utf8;
- my $part_type = $part->content_type;
- if ($part_type =~ m!\btext/[a-z0-9\+\._-]+\b!i) {
+ if ($part_nr > 0) {
my $fn = $part->filename;
+ defined($fn) or $fn = "part #" . ($part_nr + 1);
+ $rv .= add_filename_line($enc->decode($fn));
+ }
- if ($part_nr > 0) {
- defined($fn) or $fn = "part #$part_nr";
- $rv .= add_filename_line($fn);
- }
-
- # n.b. $part->body should already be decoded if text
- $rv .= escapeHTML($part->body);
- $rv .= "\n" unless $rv =~ /\n\z/s;
+ if (defined $full_pfx) {
+ $rv .= add_text_body_short($enc, $part, $part_nr,
+ $full_pfx);
} else {
- $rv .= "-- part #$part_nr ";
- $rv .= escapeHTML($part_type);
- $rv .= " skipped\n";
+ $rv .= add_text_body_full($enc, $part, $part_nr);
}
+ $rv .= "\n" unless $rv =~ /\n\z/s;
++$part_nr;
});
$rv;
$len -= length($fn);
$pad x= ($len/2) if ($len > 0);
- "$pad " . escapeHTML($fn) . " $pad\n";
+ "$pad " . ascii_html($fn) . " $pad\n";
+}
+
+sub add_text_body_short {
+ my ($enc, $part, $part_nr, $full_pfx) = @_;
+ my $n = 0;
+ my $s = ascii_html($enc->decode($part->body));
+ $s =~ s!^((?:(?:>[^\n]*)\n)+)!
+ my $cur = $1;
+ my @lines = split(/\n/, $cur);
+ if (@lines > MAX_INLINE_QUOTED) {
+ # show a short snippet of quoted text
+ $cur = join(' ', @lines);
+ $cur =~ s/^>\s*//;
+
+ my @sum = split(/\s+/, $cur);
+ $cur = '';
+ do {
+ my $tmp = shift(@sum);
+ my $len = length($tmp) + length($cur);
+ if ($len > MAX_TRUNC_LEN) {
+ @sum = ();
+ } else {
+ $cur .= $tmp . ' ';
+ }
+ } while (@sum && length($cur) < MAX_TRUNC_LEN);
+ $cur =~ s/ \z/ .../;
+ "> <<a href=\"${full_pfx}#q${part_nr}_" . $n++ .
+ "\">$cur<\/a>>\n";
+ } else {
+ $cur;
+ }
+ !emg;
+ $s;
+}
+
+sub add_text_body_full {
+ my ($enc, $part, $part_nr) = @_;
+ my $n = 0;
+ my $s = ascii_html($enc->decode($part->body));
+ $s =~ s!^((?:(?:>[^\n]*)\n)+)!
+ my $cur = $1;
+ my @lines = split(/\n/, $cur);
+ if (@lines > MAX_INLINE_QUOTED) {
+ "<a name=q${part_nr}_" . $n++ . ">$cur</a>";
+ } else {
+ $cur;
+ }
+ !emg;
+ $s;
+}
+
+sub trim_message_id {
+ my ($mid) = @_;
+ $mid =~ s/\A\s*<//;
+ $mid =~ s/>\s*\z//;
+ my $html = ascii_html($mid);
+ my $href = ascii_html(uri_escape($mid));
+
+ ($html, $href);
+}
+
+sub ascii_html {
+ $enc_ascii->encode(escapeHTML($_[0]), Encode::HTMLCREF);
}
sub headers_to_html_header {
foreach my $h (qw(From To Cc Subject Date)) {
my $v = $simple->header($h);
defined $v or next;
- $v = decode("MIME-Header", $v);
- $v = encode("utf8", $v);
- $v = escapeHTML($v);
- $v =~ tr/\n/ /;
+ $v =~ tr/\n/ /s;
+ $v =~ tr/\r//d;
+ my $raw = $enc_mime->decode($v);
+ $v = ascii_html($raw);
$rv .= "$h: $v\n";
- if ($h eq "From" || $h eq "Subject") {
- push @title, $v;
+ if ($h eq 'From') {
+ my @from = Email::Address->parse($raw);
+ $raw = $from[0]->name;
+ unless (defined($raw) && length($raw)) {
+ $raw = '<' . $from[0]->address . '>';
+ }
+ $title[1] = ascii_html($raw);
+
+ } elsif ($h eq 'Subject') {
+ $title[0] = $v;
}
}
- foreach my $h (qw(Message-ID In-Reply-To)) {
- my $v = $simple->header($h);
- defined $v or next;
- $v =~ tr/<>//d;
- my $html = escapeHTML($v);
- my $href = escapeHTML(escape($v));
- $rv .= "$h: <a href=\"$href\">$html</a>\n";
+ my $mid = $simple->header('Message-ID');
+ if (defined $mid) {
+ my ($html, $href) = trim_message_id($mid);
+ $rv .= "Message-ID: <$html> ";
+ $rv .= "(<a href=\"$href.txt\">original</a>)\n";
}
+ my $irp = $simple->header('In-Reply-To');
+ if (defined $irp) {
+ my ($html, $href) = trim_message_id($irp);
+ $rv .= "In-Reply-To: <";
+ $rv .= "<a href=\"$href.html\">$html</a>>\n";
+ }
$rv .= "\n";
- return ("<html><head><title>".
- join(' - ', @title) .
- '</title></head><body><pre style="white-space:pre-wrap">' .
- $rv);
+ ("<html><head><title>". join(' - ', @title) .
+ '</title></head><body><pre style="white-space:pre-wrap">' . $rv);
}
1;