X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FSearchMsg.pm;h=42384936b923364bb17fc515bd6b39c0d83bc35e;hb=95bdac7f09c69036efed537a4d03d5bdd2ae4eb6;hp=65e085f4fb0ae0e3cbb03ba9c885c8d9d3432eb8;hpb=588111b86a69c28384f4c907fe7b1b87aa484f32;p=public-inbox.git

diff --git a/lib/PublicInbox/SearchMsg.pm b/lib/PublicInbox/SearchMsg.pm
index 65e085f4..42384936 100644
--- a/lib/PublicInbox/SearchMsg.pm
+++ b/lib/PublicInbox/SearchMsg.pm
@@ -1,11 +1,15 @@
-# Copyright (C) 2015-2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2015-2020 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 # based on notmuch, but with no concept of folders, files or flags
 #
 # Wraps a document inside our Xapian search index.
+# There may be many of these objects loaded in memory at once
+# for large threads in our WWW UI.
 package PublicInbox::SearchMsg;
 use strict;
 use warnings;
+use base qw(Exporter);
+our @EXPORT_OK = qw(subject_normalized);
 use PublicInbox::MID qw/mid_clean mid_mime/;
 use PublicInbox::Address;
 use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp);
@@ -21,15 +25,10 @@ sub wrap {
 	bless { mid => $mid }, $class;
 }
 
-sub get {
-	my ($class, $head, $db, $mid) = @_;
-	my $doc_id = $head->get_docid;
-	load_expand(wrap($class, $mid), $db->get_document($doc_id));
-}
-
 sub get_val ($$) {
 	my ($doc, $col) = @_;
-	Search::Xapian::sortable_unserialise($doc->get_value($col));
+	# sortable_unserialise is defined by PublicInbox::Search::load_xapian()
+	sortable_unserialise($doc->get_value($col));
 }
 
 sub to_doc_data {
@@ -57,11 +56,14 @@ sub load_from_data ($$) {
 
 		# To: and Cc: are stored to optimize HDR/XHDR in NNTP since
 		# some NNTP clients will use that for message displays.
+		# NNTP only, and only stored in Over(view), not Xapian
 		$self->{to},
 		$self->{cc},
 
 		$self->{blob},
 		$self->{mid},
+
+		# NNTP only
 		$self->{bytes},
 		$self->{lines}
 	) = split(/\n/, $_[1]);
@@ -79,10 +81,22 @@ sub load_expand {
 	$self;
 }
 
-sub load_doc {
-	my ($class, $doc) = @_;
-	my $self = bless {}, $class;
-	load_expand($self, $doc);
+sub psgi_cull ($) {
+	my ($self) = @_;
+	from_name($self); # fill in {from_name} so we can delete {from}
+
+	# drop NNTP-only fields which aren't relevant to PSGI results:
+	# saves ~80K on a 200 item search result:
+	delete @$self{qw(from ts to cc bytes lines)};
+	$self;
+}
+
+# Only called by PSGI interface, not NNTP
+sub from_mitem {
+	my ($mitem, $srch) = @_;
+	return $srch->retry_reopen(\&from_mitem, $mitem) if $srch;
+	my $self = bless {}, __PACKAGE__;
+	psgi_cull(load_expand($self, $mitem->get_document));
 }
 
 # :bytes and :lines metadata in RFC 3977
@@ -95,8 +109,8 @@ sub __hdr ($$) {
 	return $val if defined $val;
 
 	my $mime = $self->{mime} or return;
-	$val = $mime->header($field);
-	$val = '' unless defined $val;
+	my @raw = $mime->header($field);
+	$val = join(', ', @raw);
 	$val =~ tr/\t\n/  /;
 	$val =~ tr/\r//d;
 	$self->{$field} = $val;
@@ -169,4 +183,16 @@ sub mid ($;$) {
 
 sub _extract_mid { mid_clean(mid_mime($_[0]->{mime})) }
 
+our $REPLY_RE = qr/^re:\s+/i;
+
+sub subject_normalized ($) {
+	my ($subj) = @_;
+	$subj =~ s/\A\s+//s; # no leading space
+	$subj =~ s/\s+\z//s; # no trailing space
+	$subj =~ s/\s+/ /gs; # no redundant spaces
+	$subj =~ s/\.+\z//; # no trailing '.'
+	$subj =~ s/$REPLY_RE//igo; # remove reply prefix
+	$subj;
+}
+
 1;