X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FMbox.pm;h=1c97f959c84502f6529310af2b5f6704f4858c30;hb=5422a844b7384c32b3532d128e15e0b50d24435b;hp=2ec50656e36e7f02cd093db88168270a65d4bdff;hpb=f71e9e9b67a6ff23642ccd119390bd6b3cb0d91e;p=public-inbox.git

diff --git a/lib/PublicInbox/Mbox.pm b/lib/PublicInbox/Mbox.pm
index 2ec50656..1c97f959 100644
--- a/lib/PublicInbox/Mbox.pm
+++ b/lib/PublicInbox/Mbox.pm
@@ -1,55 +1,155 @@
-# 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/;
+use URI::Escape qw/uri_escape_utf8/;
+use Plack::Util;
+require Email::Simple;
 
-sub thread_mbox {
-	my ($ctx, $srch) = @_;
-	my $mid = mid_compressed($ctx->{mid});
-	my $res = $srch->get_thread($mid);
-	my $msgs = delete $res->{msgs};
-	require PublicInbox::GitCatFile;
-	require Email::Simple;
-	my $git = PublicInbox::GitCatFile->new($ctx->{git_dir});
-
-	sub {
-		my ($res) = @_; # Plack callback
-		my $w = $res->([200, [ 'Content-Type' => 'text/plain' ] ]);
-		while (defined(my $smsg = shift @$msgs)) {
-			my $msg = eval {
-				my $path = 'HEAD:' . mid2path($smsg->mid);
-				Email::Simple->new($git->cat_file($path));
-			};
-			emit($w, $msg) if $msg;
-		}
-	}
+sub emit1 {
+	my ($ctx, $msg) = @_;
+	$msg = Email::Simple->new($msg);
+	# single message should be easily renderable in browsers
+	[200, ['Content-Type', 'text/plain'], [ msg_str($ctx, $msg)] ]
 }
 
-sub emit {
-	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->{cgi});
+	my $mid = mid_clean($header_obj->header('Message-ID'));
+	$mid = uri_escape_utf8($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}, @_) };
+	# http://www.iana.org/assignments/media-types/application/gzip
+	[200, ['Content-Type' => 'application/gzip'],
+		PublicInbox::MboxGz->new($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);
+	my $cb = sub { $ctx->{srch}->query($query, @_) };
 
-	$buf = $simple->body;
-	$simple->body_set('');
-	$buf =~ s/^(From )/>$1/gm;
-	$buf .= "\n" unless $buf =~ /\n\z/s;
+	# http://www.iana.org/assignments/media-types/application/gzip
+	[200, [qw(Content-Type application/gzip)],
+		PublicInbox::MboxGz->new($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
+	$fh->close;
+}
+
+1;
+
+package PublicInbox::MboxGz;
+use strict;
+use warnings;
+
+sub new {
+	my ($class, $ctx, $cb) = @_;
+	my $buf = '';
+	bless {
+		buf => \$buf,
+		gz => IO::Compress::Gzip->new(\$buf, Time => 0),
+		cb => $cb,
+		ctx => $ctx,
+		msgs => [],
+		opts => { asc => 1, offset => 0 },
+	}, $class;
+}
 
-	$fh->write($buf);
+# 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 {
+		while (defined(my $smsg = shift @{$self->{msgs}})) {
+			my $msg = eval { $ibx->msg_by_mid($smsg->mid) } or next;
+			$msg = Email::Simple->new($msg);
+			$gz->write(PublicInbox::Mbox::msg_str($ctx, $msg));
+			my $bref = $self->{buf};
+			if (length($$bref) >= 8192) {
+				my $ret = $$bref; # copy :<
+				${$self->{buf}} = '';
+				return $ret;
+			}
+		}
+		$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;