);
my %bool_pfx_external = (
- mid => 'XMID', # uniQue id (Message-ID)
+ mid => 'Q', # Message-ID (full/exact), this is mostly uniQue
);
+my $non_quoted_body = 'XNQ XDFN XDFA XDFB XDFHH XDFCTX XDFPRE XDFPOST';
my %prob_prefix = (
# for mairix compatibility
s => 'S',
c => 'XCC',
tcf => 'XTO XCC A',
a => 'XTO XCC A',
- b => 'XNQ XQUOT',
- bs => 'XNQ XQUOT S',
+ b => $non_quoted_body . ' XQUOT',
+ bs => $non_quoted_body . ' XQUOT S',
n => 'XFN',
q => 'XQUOT',
- nq => 'XNQ',
+ nq => $non_quoted_body,
dfn => 'XDFN',
dfa => 'XDFA',
dfb => 'XDFB',
dfblob => 'XDFPRE XDFPOST',
# default:
- '' => 'XM S A XNQ XQUOT XFN',
+ '' => 'XM S A XQUOT XFN ' . $non_quoted_body,
);
# not documenting m: and mid: for now, the using the URLs works w/o Xapian
my $mail_query = Search::Xapian::Query->new('T' . 'mail');
sub xdir {
- my (undef, $git_dir) = @_;
- "$git_dir/public-inbox/xapian" . SCHEMA_VERSION;
+ my ($self) = @_;
+ if ($self->{version} == 1) {
+ "$self->{mainrepo}/public-inbox/xapian" . SCHEMA_VERSION;
+ } else {
+ my $dir = "$self->{mainrepo}/xap" . SCHEMA_VERSION;
+ my $part = $self->{partition};
+ defined $part or die "partition not given";
+ $dir .= "/$part";
+ }
}
sub new {
- my ($class, $git_dir, $altid) = @_;
- my $dir = $class->xdir($git_dir);
- my $db = Search::Xapian::Database->new($dir);
- bless { xdb => $db, git_dir => $git_dir, altid => $altid }, $class;
+ my ($class, $mainrepo, $altid) = @_;
+ my $version = 1;
+ my $ibx = $mainrepo;
+ if (ref $ibx) {
+ $version = $ibx->{version} || 1;
+ $mainrepo = $ibx->{mainrepo};
+ }
+ my $self = bless {
+ mainrepo => $mainrepo,
+ altid => $altid,
+ version => $version,
+ }, $class;
+ if ($version >= 2) {
+ my $dir = "$self->{mainrepo}/xap" . SCHEMA_VERSION;
+ my $xdb;
+ my $parts = 0;
+ foreach my $part (<$dir/*>) {
+ -d $part && $part =~ m!/\d+\z! or next;
+ $parts++;
+ my $sub = Search::Xapian::Database->new($part);
+ if ($xdb) {
+ $xdb->add_database($sub);
+ } else {
+ $xdb = $sub;
+ }
+ }
+ $self->{xdb} = $xdb;
+ $self->{skel} = Search::Xapian::Database->new("$dir/skel");
+ } else {
+ $self->{xdb} = Search::Xapian::Database->new($self->xdir);
+ }
+ $self;
}
-sub reopen { $_[0]->{xdb}->reopen }
+sub reopen {
+ my ($self) = @_;
+ $self->{xdb}->reopen;
+ if (my $skel = $self->{skel}) {
+ $skel->reopen;
+ }
+}
# read-only
sub query {
sub get_thread {
my ($self, $mid, $opts) = @_;
- my $smsg = eval { $self->lookup_message($mid) };
+ my $smsg = retry_reopen($self, sub { lookup_skeleton($self, $mid) });
return { total => 0, msgs => [] } unless $smsg;
my $qtid = Search::Xapian::Query->new('G' . $smsg->thread_id);
# always sort threads by timestamp, this makes life easier
# for the threading algorithm (in SearchThread.pm)
$opts->{asc} = 1;
-
+ $opts->{enquire} = enquire_skel($self);
_do_enquire($self, $qtid, $opts);
}
if (ref($@) eq 'Search::Xapian::DatabaseModifiedError') {
reopen($self);
} else {
+ warn "ref: ", ref($@), "\n";
die;
}
}
sub _enquire_once {
my ($self, $query, $opts) = @_;
- my $enquire = $self->enquire;
+ my $enquire = $opts->{enquire} || enquire($self);
if (defined $query) {
$query = Search::Xapian::Query->new(OP_AND,$query,$mail_query);
} else {
sub query_xover {
my ($self, $beg, $end, $offset) = @_;
my $qp = Search::Xapian::QueryParser->new;
- $qp->set_database($self->{xdb});
+ $qp->set_database($self->{skel} || $self->{xdb});
$qp->add_valuerangeprocessor($self->num_range_processor);
my $query = $qp->parse_query("$beg..$end", QP_FLAGS);
- _do_enquire($self, $query, {num => 1, limit => 200, offset => $offset});
+ my $opts = {
+ enquire => enquire_skel($self),
+ num => 1,
+ limit => 200,
+ offset => $offset,
+ };
+ _do_enquire($self, $query, $opts);
+}
+
+sub query_ts {
+ my ($self, $ts, $opts) = @_;
+ my $qp = $self->{qp_ts} ||= eval {
+ my $q = Search::Xapian::QueryParser->new;
+ $q->set_database($self->{skel} || $self->{xdb});
+ $q->add_valuerangeprocessor(
+ Search::Xapian::NumberValueRangeProcessor->new(TS));
+ $q
+ };
+ my $query = $qp->parse_query($ts, QP_FLAGS);
+ $opts->{enquire} = enquire_skel($self);
+ _do_enquire($self, $query, $opts);
+}
+
+sub lookup_skeleton {
+ my ($self, $mid) = @_;
+ my $skel = $self->{skel} or return lookup_message($self, $mid);
+ $mid = mid_clean($mid);
+ my $term = 'Q' . $mid;
+ my $smsg;
+ my $beg = $skel->postlist_begin($term);
+ if ($beg != $skel->postlist_end($term)) {
+ my $doc_id = $beg->get_docid;
+ if (defined $doc_id) {
+ # raises on error:
+ my $doc = $skel->get_document($doc_id);
+ $smsg = PublicInbox::SearchMsg->wrap($doc, $mid);
+ $smsg->{doc_id} = $doc_id;
+ }
+ }
+ $smsg;
}
sub lookup_message {
my ($self, $mid) = @_;
$mid = mid_clean($mid);
- my $doc_id = $self->find_unique_doc_id('XMID' . $mid);
+ my $doc_id = $self->find_first_doc_id('Q' . $mid);
my $smsg;
if (defined $doc_id) {
# raises on error:
});
}
+sub lookup_article {
+ my ($self, $num) = @_;
+ my $term = 'XNUM'.$num;
+ my $smsg;
+ eval {
+ retry_reopen($self, sub {
+ my $db = $self->{skel} || $self->{xdb};
+ my $head = $db->postlist_begin($term);
+ return if $head == $db->postlist_end($term);
+ my $doc_id = $head->get_docid;
+ return unless defined $doc_id;
+ # raises on error:
+ my $doc = $db->get_document($doc_id);
+ $smsg = PublicInbox::SearchMsg->wrap($doc);
+ $smsg->load_expand;
+ $smsg->{doc_id} = $doc_id;
+ });
+ };
+ $smsg;
+}
+
+sub each_smsg_by_mid {
+ my ($self, $mid, $cb) = @_;
+ my $xdb = $self->{xdb};
+ # XXX retry_reopen isn't necessary for V2Writable, but the PSGI
+ # interface will need it...
+ my ($head, $tail) = $self->find_doc_ids('Q' . $mid);
+ for (; $head->nequal($tail); $head->inc) {
+ my $doc_id = $head->get_docid;
+ my $doc = $xdb->get_document($doc_id);
+ my $smsg = PublicInbox::SearchMsg->wrap($doc, $mid);
+ $smsg->{doc_id} = $doc_id;
+ $cb->($smsg) or return;
+ }
+}
+
sub find_unique_doc_id {
my ($self, $termval) = @_;
($db->postlist_begin($termval), $db->postlist_end($termval));
}
+sub find_first_doc_id {
+ my ($self, $termval) = @_;
+
+ my ($begin, $end) = $self->find_doc_ids($termval);
+
+ return undef if $begin->equal($end); # not found
+
+ $begin->get_docid;
+}
+
# normalize subjects so they are suitable as pathnames for URLs
# XXX: consider for removal
sub subject_path {
$self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb});
}
+sub enquire_skel {
+ my ($self) = @_;
+ if (my $skel = $self->{skel}) {
+ $self->{enquire_skel} ||= Search::Xapian::Enquire->new($skel);
+ } else {
+ enquire($self);
+ }
+}
+
sub help {
my ($self) = @_;
$self->qp; # parse altids