thread: prevent hidden threads in /$INBOX/ landing page

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

diff --git a/lib/PublicInbox/SearchThread.pm b/lib/PublicInbox/SearchThread.pm
index 5774a9562b55f27956fd8e8e1baad11389e95e08..450a06f43bc86f6266d32811e6efd507ffe94c7a 100644 (file)
--- a/lib/PublicInbox/SearchThread.pm
+++ b/lib/PublicInbox/SearchThread.pm
@@ -7,7 +7,7 @@
 # Mail::Thread is unmaintained and unavailable on some distros.
 # We also do not want pruning or subject grouping, since we want
 # to encourage strict threading and hopefully encourage people
-# to use proper In-Reply-To.
+# to use proper In-Reply-To/References.
 #
 # This includes fixes from several open bugs for Mail::Thread
 #
@@ -21,40 +21,42 @@ package PublicInbox::SearchThread;
 use strict;
 use warnings;
 
-sub new {
-       return bless {
-               messages => $_[1],
-               id_table => {},
-               rootset  => []
-       }, $_[0];
-}
-
 sub thread {
-       my $self = shift;
-       _add_message($self, $_) foreach @{$self->{messages}};
-       my $id_table = delete $self->{id_table};
-       $self->{rootset} = [ grep { !delete $_->{parent} } values %$id_table ];
+       my ($messages, $ordersub, $ibx) = @_;
+       my $id_table = {};
+       _add_message($id_table, $_) foreach @$messages;
+       my $rootset = [ grep {
+                       !delete($_->{parent}) && $_->visible($ibx)
+               } values %$id_table ];
+       $id_table = undef;
+       $rootset = $ordersub->($rootset);
+       $_->order_children($ordersub, $ibx) for @$rootset;
+       $rootset;
 }
 
 sub _get_cont_for_id ($$) {
-       my ($self, $mid) = @_;
-       $self->{id_table}{$mid} ||= PublicInbox::SearchThread::Msg->new($mid);
+       my ($id_table, $mid) = @_;
+       $id_table->{$mid} ||= PublicInbox::SearchThread::Msg->new($mid);
 }
 
 sub _add_message ($$) {
-       my ($self, $smsg) = @_;
+       my ($id_table, $smsg) = @_;
 
        # A. if id_table...
-       my $this = _get_cont_for_id($self, $smsg->{mid});
+       my $this = _get_cont_for_id($id_table, $smsg->{mid});
        $this->{smsg} = $smsg;
 
        # B. For each element in the message's References field:
        defined(my $refs = $smsg->{references}) or return;
 
+       # This loop exists to help fill in gaps left from missing
+       # messages.  It is not needed in a perfect world where
+       # everything is perfectly referenced, only the last ref
+       # matters.
        my $prev;
        foreach my $ref ($refs =~ m/<([^>]+)>/g) {
                # Find a Container object for the given Message-ID
-               my $cont = _get_cont_for_id($self, $ref);
+               my $cont = _get_cont_for_id($id_table, $ref);
 
                # Link the References field's Containers together in
                # the order implied by the References header
@@ -73,19 +75,12 @@ sub _add_message ($$) {
        }
 
        # C. Set the parent of this message to be the last element in
-       # References...
-       if ($prev && !$this->has_descendent($prev)) { # would loop
-               $prev->add_child($this)
+       # References.
+       if (defined $prev && !$this->has_descendent($prev)) { # would loop
+               $prev->add_child($this);
        }
 }
 
-sub order {
-       my ($self, $ordersub) = @_;
-       my $rootset = $ordersub->($self->{rootset});
-       $self->{rootset} = $rootset;
-       $_->order_children($ordersub) for @$rootset;
-}
-
 package PublicInbox::SearchThread::Msg;
 use strict;
 use warnings;
@@ -126,22 +121,32 @@ sub add_child {
 
 sub has_descendent {
        my ($self, $child) = @_;
+       my %seen; # loop prevention
        while ($child) {
-               return 1 if $self == $child;
+               return 1 if $self == $child || $seen{$child}++;
                $child = $child->{parent};
        }
        0;
 }
 
+# Do not show/keep ghosts iff they have no children.  Sometimes
+# a ghost Message-ID is the result of a long header line
+# being folded/mangled by a MUA, and not a missing message.
+sub visible ($$) {
+       my ($self, $ibx) = @_;
+       ($self->{smsg} ||= eval { $ibx->smsg_by_mid($self->{id}) }) ||
+        (scalar values %{$self->{children}});
+}
+
 sub order_children {
-       my ($cur, $ordersub) = @_;
+       my ($cur, $ordersub, $ibx) = @_;
 
-       my %seen = ($cur => 1);
+       my %seen = ($cur => 1); # self-referential loop prevention
        my @q = ($cur);
        while (defined($cur = shift @q)) {
                my $c = $cur->{children}; # The hashref here...
 
-               $c = [ grep { !$seen{$_}++ } values %$c ]; # spot/break loops
+               $c = [ grep { !$seen{$_}++ && visible($_, $ibx) } values %$c ];
                $c = $ordersub->($c) if scalar @$c > 1;
                $cur->{children} = $c; # ...becomes an arrayref
                push @q, @$c;