X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearch.pm;h=67837f4f77300e6ef76333c623cb07eeca24ea30;hb=7eeadcb62729b0efbcb53cd9b7b181897c92cf9a;hp=d4f6f77a548f79acedfaf0b0de2a5b640dcaf017;hpb=478d03688600a4c7b50e205d15d76113e019f3cd;p=public-inbox.git

diff --git a/lib/PublicInbox/Search.pm b/lib/PublicInbox/Search.pm
index d4f6f77a..67837f4f 100644
--- a/lib/PublicInbox/Search.pm
+++ b/lib/PublicInbox/Search.pm
@@ -16,7 +16,7 @@ use constant YYYYMMDD => 4; # for searching in the WWW UI
 
 use Search::Xapian qw/:standard/;
 use PublicInbox::SearchMsg;
-use Email::MIME;
+use PublicInbox::MIME;
 use PublicInbox::MID qw/mid_clean id_compress/;
 
 # This is English-only, everything else is non-standard and may be confused as
@@ -39,7 +39,10 @@ use constant {
 	# 10 - optimize doc for NNTP overviews
 	# 11 - merge threads when vivifying ghosts
 	# 12 - change YYYYMMDD value column to numeric
-	SCHEMA_VERSION => 12,
+	# 13 - fix threading for empty References/In-Reply-To
+	#      (commit 83425ef12e4b65cdcecd11ddcb38175d4a91d5a0)
+	# 14 - fix ghost root vivification
+	SCHEMA_VERSION => 14,
 
 	# n.b. FLAG_PURE_NOT is expensive not suitable for a public website
 	# as it could become a denial-of-service vector
@@ -53,8 +56,6 @@ my %bool_pfx_internal = (
 );
 
 my %bool_pfx_external = (
-	# do we still need these? probably not..
-	path => 'XPATH',
 	mid => 'Q', # uniQue id (Message-ID)
 );
 
@@ -89,7 +90,7 @@ are also supported
 EOF
 	'b:' => 'match within message body, including text attachments',
 	'nq:' => 'match non-quoted text within message body',
-	'quot:' => 'match quoted text within message body',
+	'q:' => 'match quoted text within message body',
 	'n:' => 'match filename of attachment(s)',
 	't:' => 'match within the To header',
 	'c:' => 'match within the Cc header',
@@ -104,11 +105,7 @@ chomp @HELP;
 # da (diff a/ removed lines)
 # db (diff b/ added lines)
 
-my %all_pfx = (%bool_pfx_internal, %bool_pfx_external, %prob_prefix);
-
-sub xpfx { $all_pfx{$_[0]} }
-
-my $mail_query = Search::Xapian::Query->new(xpfx('type') . 'mail');
+my $mail_query = Search::Xapian::Query->new('T' . 'mail');
 
 sub xdir {
 	my (undef, $git_dir) = @_;
@@ -143,11 +140,11 @@ sub get_thread {
 	my $smsg = eval { $self->lookup_message($mid) };
 
 	return { total => 0, msgs => [] } unless $smsg;
-	my $qtid = Search::Xapian::Query->new(xpfx('thread').$smsg->thread_id);
+	my $qtid = Search::Xapian::Query->new('G' . $smsg->thread_id);
 	my $path = $smsg->path;
 	if (defined $path && $path ne '') {
 		my $path = id_compress($smsg->path);
-		my $qsub = Search::Xapian::Query->new(xpfx('path').$path);
+		my $qsub = Search::Xapian::Query->new('XPATH' . $path);
 		$qtid = Search::Xapian::Query->new(OP_OR, $qtid, $qsub);
 	}
 	$opts ||= {};
@@ -276,27 +273,29 @@ sub lookup_message {
 	my ($self, $mid) = @_;
 	$mid = mid_clean($mid);
 
-	my $doc_id = $self->find_unique_doc_id('mid', $mid);
+	my $doc_id = $self->find_unique_doc_id('Q' . $mid);
 	my $smsg;
 	if (defined $doc_id) {
 		# raises on error:
 		my $doc = $self->{xdb}->get_document($doc_id);
 		$smsg = PublicInbox::SearchMsg->wrap($doc, $mid);
-		$smsg->doc_id($doc_id);
+		$smsg->{doc_id} = $doc_id;
 	}
 	$smsg;
 }
 
 sub lookup_mail { # no ghosts!
 	my ($self, $mid) = @_;
-	my $smsg = lookup_message($self, $mid) or return;
-	PublicInbox::SearchMsg->load_doc($smsg->{doc});
+	retry_reopen($self, sub {
+		my $smsg = lookup_message($self, $mid) or return;
+		PublicInbox::SearchMsg->load_doc($smsg->{doc});
+	});
 }
 
 sub find_unique_doc_id {
-	my ($self, $term, $value) = @_;
+	my ($self, $termval) = @_;
 
-	my ($begin, $end) = $self->find_doc_ids($term, $value);
+	my ($begin, $end) = $self->find_doc_ids($termval);
 
 	return undef if $begin->equal($end); # not found
 
@@ -304,26 +303,20 @@ sub find_unique_doc_id {
 
 	# sanity check
 	$begin->inc;
-	$begin->equal($end) or die "Term '$term:$value' is not unique\n";
+	$begin->equal($end) or die "Term '$termval' is not unique\n";
 	$rv;
 }
 
 # returns begin and end PostingIterator
 sub find_doc_ids {
-	my ($self, $term, $value) = @_;
-
-	$self->find_doc_ids_for_term(xpfx($term) . $value);
-}
-
-# returns begin and end PostingIterator
-sub find_doc_ids_for_term {
-	my ($self, $term) = @_;
+	my ($self, $termval) = @_;
 	my $db = $self->{xdb};
 
-	($db->postlist_begin($term), $db->postlist_end($term));
+	($db->postlist_begin($termval), $db->postlist_end($termval));
 }
 
 # normalize subjects so they are suitable as pathnames for URLs
+# XXX: consider for removal
 sub subject_path {
 	my $subj = pop;
 	$subj = subject_normalized($subj);
@@ -341,32 +334,6 @@ sub subject_normalized {
 	$subj;
 }
 
-# for doc data
-sub subject_summary {
-	my $subj = pop;
-	my $max = 68;
-	if (length($subj) > $max) {
-		my @subj = split(/\s+/, $subj);
-		$subj = '';
-		my $l;
-
-		while ($l = shift @subj) {
-			my $new = $subj . $l . ' ';
-			last if length($new) >= $max;
-			$subj = $new;
-		}
-		if ($subj ne '') {
-			my $r = scalar @subj ? ' ...' : '';
-			$subj =~ s/ \z/$r/s;
-		} else {
-			# subject has one REALLY long word, and NOT spam? wtf
-			@subj = ($l =~ /\A(.{1,72})/);
-			$subj = $subj[0] . ' ...';
-		}
-	}
-	$subj;
-}
-
 sub enquire {
 	my ($self) = @_;
 	$self->{enquire} ||= Search::Xapian::Enquire->new($self->{xdb});