use warnings;
use Fcntl qw(:flock :DEFAULT);
use PublicInbox::MIME;
-use Email::MIME::ContentType;
-$Email::MIME::ContentType::STRICT_PARAMS = 0;
use base qw(PublicInbox::Search);
-use PublicInbox::MID qw/mid_clean id_compress mid_mime/;
+use PublicInbox::MID qw/mid_clean id_compress mid_mime mids references/;
use PublicInbox::MsgIter;
use Carp qw(croak);
use POSIX qw(strftime);
$doc->add_value($col, $num);
}
-sub add_values ($$$$) {
- my ($smsg, $bytes, $num, $lines) = @_;
+sub add_values ($$) {
+ my ($doc, $values) = @_;
- my $ts = $smsg->ts;
- my $doc = $smsg->{doc};
- add_val($doc, &PublicInbox::Search::TS, $ts);
+ my $ts = $values->[PublicInbox::Search::TS];
+ add_val($doc, PublicInbox::Search::TS, $ts);
- defined($num) and add_val($doc, &PublicInbox::Search::NUM, $num);
+ my $num = $values->[PublicInbox::Search::NUM];
+ defined($num) and add_val($doc, PublicInbox::Search::NUM, $num);
- defined($bytes) and add_val($doc, &PublicInbox::Search::BYTES, $bytes);
+ my $bytes = $values->[PublicInbox::Search::BYTES];
+ defined($bytes) and add_val($doc, PublicInbox::Search::BYTES, $bytes);
- add_val($doc, &PublicInbox::Search::LINES, $lines);
+ my $lines = $values->[PublicInbox::Search::LINES];
+ add_val($doc, PublicInbox::Search::LINES, $lines);
my $yyyymmdd = strftime('%Y%m%d', gmtime($ts));
add_val($doc, PublicInbox::Search::YYYYMMDD, $yyyymmdd);
my $db = $self->{xdb};
my ($doc_id, $old_tid);
- my $mid = mid_clean(mid_mime($mime));
- my $threader = $self->{threader};
+ my @mids = mid_mime($mime);
+ if (@mids > 1) {
+ warn "Multi-MID: ( ",join(' | ', @mids)," )\n";
+ }
+ my $mid = mid_clean($mids[0]);
+ my $skel = $self->{skeleton};
eval {
die 'Message-ID too long' if length($mid) > MAX_MID_SIZE;
# convert a ghost to a regular message
# it will also clobber any existing regular message
$doc_id = $smsg->{doc_id};
- $old_tid = $smsg->thread_id unless $threader;
+ $old_tid = $smsg->thread_id unless $skel;
}
$smsg = PublicInbox::SearchMsg->new($mime);
my $doc = $smsg->{doc};
}
my $lines = $mime->body_raw =~ tr!\n!\n!;
- add_values($smsg, $bytes, $num, $lines);
+ my @values = ($smsg->ts, $num, $bytes, $lines);
+ add_values($doc, \@values);
my $tg = $self->term_generator;
# populates smsg->references for smsg->to_doc_data
my $refs = parse_references($smsg);
my $data = $smsg->to_doc_data($blob);
- if ($threader) {
- $threader->thread_msg($mid, $smsg->ts, $xpath, $data);
+ if ($skel) {
+ push @values, $mid, $xpath, $data;
+ $skel->index_skeleton(\@values);
} else {
link_message($self, $smsg, $refs, $old_tid);
}
sub parse_references ($) {
my ($smsg) = @_;
- my $doc = $smsg->{doc};
- my $mid = $smsg->mid;
my $mime = $smsg->{mime};
my $hdr = $mime->header_obj;
-
- # last References should be IRT, but some mail clients do things
- # out of order, so trust IRT over References iff IRT exists
- my @refs = (($hdr->header_raw('References') || '') =~ /<([^>]+)>/g);
- push(@refs, (($hdr->header_raw('In-Reply-To') || '') =~ /<([^>]+)>/g));
-
- if (@refs) {
- my %uniq = ($mid => 1);
- my @orig_refs = @refs;
- @refs = ();
-
- # prevent circular references via References: here:
- foreach my $ref (@orig_refs) {
- if (length($ref) > MAX_MID_SIZE) {
- warn "References: <$ref> too long, ignoring\n";
- }
- next if $uniq{$ref};
- $uniq{$ref} = 1;
- push @refs, $ref;
+ my $refs = references($hdr);
+ return $refs if scalar(@$refs) == 0;
+
+ # prevent circular references via References here:
+ my %mids = map { $_ => 1 } @{mids($hdr)};
+ my @keep;
+ foreach my $ref (@$refs) {
+ # FIXME: this is an archive-prevention vector like X-No-Archive
+ if (length($ref) > MAX_MID_SIZE) {
+ warn "References: <$ref> too long, ignoring\n";
}
+ next if $mids{$ref};
+ push @keep, $ref;
}
- $smsg->{references} = '<'.join('> <', @refs).'>' if @refs;
- \@refs
+ $smsg->{references} = '<'.join('> <', @keep).'>' if @keep;
+ \@keep;
}
sub link_message {
$smsg->{doc}->add_term('G' . $tid);
}
-sub index_blob {
- my ($self, $mime, $bytes, $num, $blob) = @_;
- $self->add_message($mime, $bytes, $num, $blob);
-}
-
sub index_git_blob_id {
my ($doc, $pfx, $objid) = @_;
}
sub index_mm {
- my ($self, $mime, $warn_existing) = @_;
+ my ($self, $mime) = @_;
my $mid = mid_clean(mid_mime($mime));
my $mm = $self->{mm};
my $num = $mm->mid_insert($mid);
return $num if defined $num;
- warn "<$mid> reused\n" if $warn_existing;
# fallback to num_for since filters like RubyLang set the number
$mm->num_for($mid);
}
sub index_mm2 {
my ($self, $mime, $bytes, $blob) = @_;
my $num = $self->{mm}->num_for(mid_clean(mid_mime($mime)));
- index_blob($self, $mime, $bytes, $num, $blob);
+ add_message($self, $mime, $bytes, $num, $blob);
}
sub unindex_mm2 {
sub index_both {
my ($self, $mime, $bytes, $blob) = @_;
my $num = index_mm($self, $mime);
- index_blob($self, $mime, $bytes, $num, $blob);
+ add_message($self, $mime, $bytes, $num, $blob);
}
sub unindex_both {
}
} else {
# user didn't install DBD::SQLite and DBI
- rlog($self, $xlog, *index_blob, *unindex_blob, $cb);
+ rlog($self, $xlog, *add_message, *unindex_blob, $cb);
}
}
sub merge_threads {
my ($self, $winner_tid, $loser_tid) = @_;
return if $winner_tid == $loser_tid;
- my ($head, $tail) = $self->find_doc_ids('G' . $loser_tid);
my $db = $self->{xdb};
- for (; $head != $tail; $head->inc) {
- my $docid = $head->get_docid;
- my $doc = $db->get_document($docid);
- $doc->remove_term('G' . $loser_tid);
- $doc->add_term('G' . $winner_tid);
- $db->replace_document($docid, $doc);
+ my $batch_size = 1000; # don't let @ids grow too large to avoid OOM
+ while (1) {
+ my ($head, $tail) = $self->find_doc_ids('G' . $loser_tid);
+ return if $head == $tail;
+ my @ids;
+ for (; $head != $tail && @ids < $batch_size; $head->inc) {
+ push @ids, $head->get_docid;
+ }
+ foreach my $docid (@ids) {
+ my $doc = $db->get_document($docid);
+ $doc->remove_term('G' . $loser_tid);
+ $doc->add_term('G' . $winner_tid);
+ $db->replace_document($docid, $doc);
+ }
}
}
$_[0]->{lockfh} = undef;
}
-# remote_* subs are only used by SearchIdxPart and SearchIdxThread:
+# remote_* subs are only used by SearchIdxPart and SearchIdxSkeleton
sub remote_commit {
my ($self) = @_;
print { $self->{w} } "commit\n" or die "failed to write commit: $!";
print $w "close\n" or die "failed to write to pid:$pid: $!\n";
close $w or die "failed to close pipe for pid:$pid: $!\n";
waitpid($pid, 0) == $pid or die "remote process did not finish";
- $? == 0 or die ref($self)." exited with: $?";
+ $? == 0 or die ref($self)." pid:$pid exited with: $?";
}
1;