diff options
Diffstat (limited to 'lib/PublicInbox/Mbox.pm')
| -rw-r--r-- | lib/PublicInbox/Mbox.pm | 195 |
1 files changed, 130 insertions, 65 deletions
diff --git a/lib/PublicInbox/Mbox.pm b/lib/PublicInbox/Mbox.pm index 04c86cc1..11b23022 100644 --- a/lib/PublicInbox/Mbox.pm +++ b/lib/PublicInbox/Mbox.pm @@ -26,12 +26,51 @@ sub subject_fn ($) { $fn eq '' ? 'no-subject' : $fn; } -sub emit1 { - my ($ctx, $msg) = @_; - $msg = Email::Simple->new($msg); - my $fn = subject_fn($msg); +sub mb_stream { + my ($more) = @_; + bless $more, 'PublicInbox::Mbox'; +} + +# called by PSGI server as body response +sub getline { + my ($more) = @_; # self + my ($ctx, $id, $prev, $next, $cur) = @$more; + if ($cur) { # first + pop @$more; + return msg_str($ctx, $cur); + } + $cur = $next or return; + my $ibx = $ctx->{-inbox}; + $next = $ibx->search->next_by_mid($ctx->{mid}, \$id, \$prev); + @$more = ($ctx, $id, $prev, $next); # $next may be undef, here + my $mref = $ibx->msg_by_smsg($cur) or return; + msg_str($ctx, Email::Simple->new($mref)); +} + +sub close {} # noop + +sub emit_raw { + my ($ctx) = @_; + my $mid = $ctx->{mid}; + my $ibx = $ctx->{-inbox}; + my $first; + my $more; + if (my $srch = $ibx->search) { + my ($id, $prev); + my $smsg = $srch->next_by_mid($mid, \$id, \$prev) or return; + my $mref = $ibx->msg_by_smsg($smsg) or return; + $first = Email::Simple->new($mref); + my $next = $srch->next_by_mid($mid, \$id, \$prev); + # $more is for ->getline + $more = [ $ctx, $id, $prev, $next, $first ] if $next; + } else { + my $mref = $ibx->msg_by_mid($mid) or return; + $first = Email::Simple->new($mref); + } + return unless defined $first; + my $fn = subject_fn($first); my @hdr = ('Content-Type'); - if ($ctx->{-inbox}->{obfuscate}) { + if ($ibx->{obfuscate}) { # obfuscation is stupid, but maybe scrapers are, too... push @hdr, 'application/mbox'; $fn .= '.mbox'; @@ -40,14 +79,11 @@ sub emit1 { $fn .= '.txt'; } push @hdr, 'Content-Disposition', "inline; filename=$fn"; - - # single message should be easily renderable in browsers, - # unless obfuscation is enabled :< - [ 200, \@hdr, [ msg_str($ctx, $msg) ] ] + [ 200, \@hdr, $more ? mb_stream($more) : [ msg_str($ctx, $first) ] ]; } sub msg_str { - my ($ctx, $simple) = @_; # Email::Simple object + my ($ctx, $simple, $mid) = @_; # Email::Simple object my $header_obj = $simple->header_obj; # drop potentially confusing headers, ssoma already should've dropped @@ -57,7 +93,7 @@ sub msg_str { } my $ibx = $ctx->{-inbox}; my $base = $ibx->base_url($ctx->{env}); - my $mid = mid_clean($header_obj->header('Message-ID')); + $mid = $ctx->{mid} unless defined $mid; $mid = mid_escape($mid); my @append = ( 'Archived-At', "<$base$mid/>", @@ -93,9 +129,24 @@ 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); + my $mid = $ctx->{mid}; + my $msgs = $srch->get_thread($mid, {}); + return [404, [qw(Content-Type text/plain)], []] if !@$msgs; + my $prev = $msgs->[-1]; + my $i = 0; + my $cb = sub { + while (1) { + if (my $smsg = $msgs->[$i++]) { + return $smsg; + } + # refill result set + $msgs = $srch->get_thread($mid, $prev); + return unless @$msgs; + $prev = $msgs->[-1]; + $i = 0; + } + }; + PublicInbox::MboxGz->response($ctx, $cb, $msgs->[0]->subject); } sub emit_range { @@ -110,12 +161,56 @@ sub emit_range { mbox_all($ctx, $query); } +sub mbox_all_ids { + my ($ctx) = @_; + my $prev = 0; + my $ids = $ctx->{-inbox}->mm->ids_after(\$prev) or return + [404, [qw(Content-Type text/plain)], ["No results found\n"]]; + my $i = 0; + my $over = $ctx->{srch}->{over_ro}; + my $cb = sub { + do { + while ((my $num = $ids->[$i++])) { + my $smsg = $over->get_art($num) or next; + return $smsg; + } + $ids = $ctx->{-inbox}->mm->ids_after(\$prev); + $i = 0; + } while (@$ids); + undef; + }; + return PublicInbox::MboxGz->response($ctx, $cb, 'all'); +} + sub mbox_all { my ($ctx, $query) = @_; eval { require IO::Compress::Gzip }; return sub { need_gzip(@_) } if $@; - my $cb = sub { $ctx->{srch}->query($query, @_) }; + return mbox_all_ids($ctx) if $query eq ''; + my $opts = { mset => 2 }; + my $srch = $ctx->{srch}; + my $mset = $srch->query($query, $opts); + $opts->{offset} = $mset->size or + return [404, [qw(Content-Type text/plain)], + ["No results found\n"]]; + my $i = 0; + my $cb = sub { # called by MboxGz->getline + while (1) { + while (my $mi = (($mset->items)[$i++])) { + my $doc = $mi->get_document; + my $smsg = $srch->retry_reopen(sub { + PublicInbox::SearchMsg->load_doc($doc); + }) or next; + return $smsg; + } + # refill result set + $mset = $srch->query($query, $opts); + my $size = $mset->size or return; + $opts->{offset} += $size; + $i = 0; + } + }; PublicInbox::MboxGz->response($ctx, $cb, 'results-'.$query); } @@ -146,8 +241,6 @@ sub new { gz => IO::Compress::Gzip->new(\$buf, Time => 0), cb => $cb, ctx => $ctx, - msgs => [], - opts => { offset => 0 }, }, $class; } @@ -155,63 +248,35 @@ sub response { my ($class, $ctx, $cb, $fn) = @_; my $body = $class->new($ctx, $cb); # http://www.iana.org/assignments/media-types/application/gzip - $body->{hdr} = [ 'Content-Type', 'application/gzip' ]; - $body->{fn} = $fn; - # fill in Content-Disposition filename - my $hdr = $body->getline; - if ($body->{hdr}) { - return [ 404, ['Content-Type','text/plain'], - [ "No results found\n" ] ]; + my @h = qw(Content-Type application/gzip); + if ($fn) { + $fn = to_filename($fn); + push @h, 'Content-Disposition', "inline; filename=$fn.mbox.gz"; } - [ 200, $hdr, $body ]; -} - -sub set_filename ($$) { - my ($fn, $msg) = @_; - return to_filename($fn) if defined($fn); - - PublicInbox::Mbox::subject_fn($msg); + [ 200, \@h, $body ]; } # called by Plack::Util::foreach or similar sub getline { my ($self) = @_; my $ctx = $self->{ctx} or return; - my $res; - my $ibx = $ctx->{-inbox}; - my $gz = $self->{gz}; - do { - # work on existing result set - while (defined(my $smsg = shift @{$self->{msgs}})) { - my $msg = eval { $ibx->msg_by_smsg($smsg) } or next; - $msg = Email::Simple->new($msg); - $gz->write(PublicInbox::Mbox::msg_str($ctx, $msg)); - - # use subject of first message as subject - if (my $hdr = delete $self->{hdr}) { - my $fn = set_filename($self->{fn}, $msg); - push @$hdr, 'Content-Disposition', - "inline; filename=$fn.mbox.gz"; - return $hdr; - } - my $bref = $self->{buf}; - if (length($$bref) >= 8192) { - my $ret = $$bref; # copy :< - ${$self->{buf}} = ''; - return $ret; - } - - # be fair to other clients on public-inbox-httpd: - return ''; + while (my $smsg = $self->{cb}->()) { + my $msg = $ctx->{-inbox}->msg_by_smsg($smsg) or next; + $msg = Email::Simple->new($msg); + $self->{gz}->write(PublicInbox::Mbox::msg_str($ctx, $msg, + $smsg->{mid})); + my $bref = $self->{buf}; + if (length($$bref) >= 8192) { + my $ret = $$bref; # copy :< + ${$self->{buf}} = ''; + return $ret; } - # refill result set - $res = $self->{cb}->($self->{opts}); - $self->{msgs} = $res->{msgs}; - $res = scalar @{$self->{msgs}}; - $self->{opts}->{offset} += $res; - } while ($res); - $gz->close; + # be fair to other clients on public-inbox-httpd: + return ''; + } + delete($self->{gz})->close; + # signal that we're done and can return undef next call: delete $self->{ctx}; ${delete $self->{buf}}; } |