-# Copyright (C) 2015, all contributors <meta@public-inbox.org>
+# Copyright (C) 2015 all contributors <meta@public-inbox.org>
# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
-# Streaming interface for formatting messages as an mbox
+
+# Streaming interface for formatting messages as an mboxrd.
+# Used by the web interface
package PublicInbox::Mbox;
use strict;
use warnings;
-use PublicInbox::MID qw/mid_clean mid_compressed mid2path/;
-use Fcntl qw(SEEK_SET);
+use PublicInbox::MID qw/mid_clean mid_escape/;
+use PublicInbox::Hval qw/to_filename/;
+use Email::Simple;
+use Email::MIME::Encode;
-sub thread_mbox {
- my ($ctx, $srch) = @_;
- sub {
- my ($response) = @_; # Plack callback
- my $w = $response->([200, ['Content-Type' => 'text/plain']]);
- emit_mbox($w, $ctx, $srch);
+sub subject_fn ($) {
+ my ($simple) = @_;
+ my $fn = $simple->header('Subject');
+ return 'no-subject' unless defined($fn);
+
+ # no need for full Email::MIME, here
+ if ($fn =~ /=\?/) {
+ eval { $fn = Encode::decode('MIME-Header', $fn) };
+ $fn = 'no-subject' if $@;
}
+ $fn =~ s/^re:\s+//i;
+ $fn = to_filename($fn);
+ $fn eq '' ? 'no-subject' : $fn;
+}
+
+sub emit1 {
+ my ($ctx, $msg) = @_;
+ $msg = Email::Simple->new($msg);
+ my $fn = subject_fn($msg);
+ my @hdr = ('Content-Type');
+ if ($ctx->{-inbox}->{obfuscate}) {
+ # obfuscation is stupid, but maybe scrapers are, too...
+ push @hdr, 'application/mbox';
+ $fn .= '.mbox';
+ } else {
+ push @hdr, 'text/plain';
+ $fn .= '.txt';
+ }
+ push @hdr, 'Content-Disposition', "inline; filename=$fn";
+
+ # single message should be easily renderable in browsers,
+ # unless obfuscation is enabled :<
+ [ 200, \@hdr, [ msg_str($ctx, $msg) ] ]
}
-sub emit_msg {
- my ($fh, $simple) = @_; # Email::Simple object
+sub msg_str {
+ my ($ctx, $simple) = @_; # Email::Simple object
+ my $header_obj = $simple->header_obj;
# drop potentially confusing headers, ssoma already should've dropped
# Lines and Content-Length
- foreach my $d (qw(Lines Content-Length Status)) {
- $simple->header_set($d);
+ foreach my $d (qw(Lines Bytes Content-Length Status)) {
+ $header_obj->header_set($d);
}
+ my $ibx = $ctx->{-inbox};
+ my $base = $ibx->base_url($ctx->{env});
+ my $mid = mid_clean($header_obj->header('Message-ID'));
+ $mid = mid_escape($mid);
+ my @append = (
+ 'Archived-At', "<$base$mid/>",
+ 'List-Archive', "<$base>",
+ 'List-Post', "<mailto:$ibx->{-primary_address}>",
+ );
+ my $crlf = $simple->crlf;
+ my $buf = "From mboxrd\@z Thu Jan 1 00:00:00 1970\n" .
+ $header_obj->as_string;
+ for (my $i = 0; $i < @append; $i += 2) {
+ my $k = $append[$i];
+ my $v = $append[$i + 1];
+ my @v = $header_obj->header($k);
+ foreach (@v) {
+ if ($v eq $_) {
+ $v = undef;
+ last;
+ }
+ }
+ $buf .= "$k: $v$crlf" if defined $v;
+ }
+ $buf .= $crlf;
+
+ # mboxrd quoting style
+ # ref: http://www.qmail.org/man/man5/mbox.html
+ my $body = $simple->body;
+ $body =~ s/^(>*From )/>$1/gm;
+ $buf .= $body;
+ $buf .= "\n";
+}
+
+sub thread_mbox {
+ my ($ctx, $srch, $sfx) = @_;
+ eval { require IO::Compress::Gzip };
+ return sub { need_gzip(@_) } if $@;
+
+ my $cb = sub { $srch->get_thread($ctx->{mid}, @_) };
+ PublicInbox::MboxGz->response($ctx, $cb);
+}
+
+sub emit_range {
+ my ($ctx, $range) = @_;
- my $buf = $simple->header_obj->as_string;
- unless ($buf =~ /\AFrom /) {
- $fh->write("From a\@a Thu Jan 1 00:00:00 1970\n");
+ eval { require IO::Compress::Gzip };
+ return sub { need_gzip(@_) } if $@;
+ my $query;
+ if ($range eq 'all') { # TODO: YYYY[-MM]
+ $query = '';
+ } else {
+ return [404, [qw(Content-Type text/plain)], []];
}
- $fh->write($buf .= $simple->crlf);
- $buf = $simple->body;
- $simple->body_set('');
- $buf =~ s/^(From )/>$1/gm;
- $buf .= "\n" unless $buf =~ /\n\z/s;
+ my $cb = sub { $ctx->{srch}->query($query, @_) };
+ PublicInbox::MboxGz->response($ctx, $cb);
+}
+
+sub need_gzip {
+ my $fh = $_[0]->([501, ['Content-Type' => 'text/html']]);
+ my $title = 'gzipped mbox not available';
+ $fh->write(<<EOF);
+<html><head><title>$title</title><body><pre>$title
+The administrator needs to install the IO::Compress::Gzip Perl module
+to support gzipped mboxes.
+<a href="../">Return to index</a></pre></body></html>
+EOF
+ $fh->close;
+}
+
+1;
+
+package PublicInbox::MboxGz;
+use strict;
+use warnings;
- $fh->write($buf);
+sub new {
+ my ($class, $ctx, $cb) = @_;
+ my $buf = '';
+ bless {
+ buf => \$buf,
+ gz => IO::Compress::Gzip->new(\$buf, Time => 0),
+ cb => $cb,
+ ctx => $ctx,
+ msgs => [],
+ opts => { offset => 0 },
+ }, $class;
}
-sub emit_mbox {
- my ($fh, $ctx, $srch) = @_;
+sub response {
+ my ($class, $ctx, $cb) = @_;
+ my $body = $class->new($ctx, $cb);
+ # http://www.iana.org/assignments/media-types/application/gzip
+ $body->{hdr} = [ 'Content-Type', 'application/gzip' ];
+ my $hdr = $body->getline; # fill in Content-Disposition filename
+ [ 200, $hdr, $body ];
+}
- require PublicInbox::GitCatFile;
- require Email::Simple;
- my $mid = mid_compressed($ctx->{mid});
- my $git = PublicInbox::GitCatFile->new($ctx->{git_dir});
- my %opts = (offset => 0);
- my $nr;
+# called by Plack::Util::foreach or similar
+sub getline {
+ my ($self) = @_;
+ my $ctx = $self->{ctx} or return;
+ my $res;
+ my $ibx = $ctx->{-inbox};
+ my $gz = $self->{gz};
do {
- my $res = $srch->get_thread($mid, \%opts);
- my $msgs = $res->{msgs};
- $nr = scalar @$msgs;
- while (defined(my $smsg = shift @$msgs)) {
- my $msg = eval {
- my $p = 'HEAD:'.mid2path($smsg->mid);
- Email::Simple->new($git->cat_file($p));
- };
- emit_msg($fh, $msg) if $msg;
+ # work on existing result set
+ while (defined(my $smsg = shift @{$self->{msgs}})) {
+ my $msg = eval { $ibx->msg_by_smsg($smsg) } or next;
+ $msg = Email::Simple->new($msg);
+ $gz->write(PublicInbox::Mbox::msg_str($ctx, $msg));
+
+ # use subject of first message as subject
+ if (my $hdr = delete $self->{hdr}) {
+ my $fn = PublicInbox::Mbox::subject_fn($msg);
+ push @$hdr, 'Content-Disposition',
+ "inline; filename=$fn.mbox.gz";
+ return $hdr;
+ }
+ my $bref = $self->{buf};
+ if (length($$bref) >= 8192) {
+ my $ret = $$bref; # copy :<
+ ${$self->{buf}} = '';
+ return $ret;
+ }
+
+ # be fair to other clients on public-inbox-httpd:
+ return '';
}
- $opts{offset} += $nr;
- } while ($nr > 0);
+ # refill result set
+ $res = $self->{cb}->($self->{opts});
+ $self->{msgs} = $res->{msgs};
+ $res = scalar @{$self->{msgs}};
+ $self->{opts}->{offset} += $res;
+ } while ($res);
+ $gz->close;
+ delete $self->{ctx};
+ ${delete $self->{buf}};
}
+sub close {} # noop
+
1;