treewide: run update-copyrights from gnulib for 2019

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

diff --git a/lib/PublicInbox/SearchMsg.pm b/lib/PublicInbox/SearchMsg.pm
index ab971e002d8a52846eaed9e83dacbab474e4f2d2..42384936b923364bb17fc515bd6b39c0d83bc35e 100644 (file)
--- 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);
@@ -13,25 +17,18 @@ use Time::Local qw(timegm);
 
 sub new {
        my ($class, $mime) = @_;
-       my $doc = Search::Xapian::Document->new;
-       bless { doc => $doc, mime => $mime }, $class;
+       bless { mime => $mime }, $class;
 }
 
 sub wrap {
-       my ($class, $doc, $mid) = @_;
-       bless { doc => $doc, mime => undef, mid => $mid }, $class;
-}
-
-sub get {
-       my ($class, $head, $db, $mid) = @_;
-       my $doc_id = $head->get_docid;
-       my $doc = $db->get_document($doc_id);
-       load_expand(wrap($class, $doc, $mid))
+       my ($class, $mid) = @_;
+       bless { mid => $mid }, $class;
 }
 
 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 {
@@ -59,19 +56,21 @@ 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]);
 }
 
 sub load_expand {
-       my ($self) = @_;
-       my $doc = $self->{doc};
+       my ($self, $doc) = @_;
        my $data = $doc->get_data or return;
        $self->{ts} = get_val($doc, PublicInbox::Search::TS());
        my $dt = get_val($doc, PublicInbox::Search::DT());
@@ -82,10 +81,22 @@ sub load_expand {
        $self;
 }
 
-sub load_doc {
-       my ($class, $doc) = @_;
-       my $self = bless { doc => $doc }, $class;
-       $self->load_expand;
+sub psgi_cull ($) {
+       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
@@ -98,9 +109,9 @@ sub __hdr ($$) {
        return $val if defined $val;
 
        my $mime = $self->{mime} or return;
-       $val = $mime->header($field);
-       $val = '' unless defined $val;
-       $val =~ tr/\n/ /;
+       my @raw = $mime->header($field);
+       $val = join(', ', @raw);
+       $val =~ tr/\t\n/  /;
        $val =~ tr/\r//d;
        $self->{$field} = $val;
 }
@@ -157,33 +168,31 @@ sub references {
        defined $x ? $x : '';
 }
 
-sub _get_term_val ($$$) {
-       my ($self, $pfx, $re) = @_;
-       my $doc = $self->{doc};
-       my $end = $doc->termlist_end;
-       my $i = $doc->termlist_begin;
-       $i->skip_to($pfx);
-       if ($i != $end) {
-               my $val = $i->get_termname;
-               $val =~ s/$re// and return $val;
-       }
-       undef;
-}
-
 sub mid ($;$) {
        my ($self, $mid) = @_;
 
        if (defined $mid) {
                $self->{mid} = $mid;
-       } elsif (my $rv = $self->{mid}) {
+       } elsif (defined(my $rv = $self->{mid})) {
                $rv;
-       } elsif ($self->{doc}) {
-               $self->{mid} = _get_term_val($self, 'Q', qr/\AQ/);
        } else {
+               die "NO {mime} for mid\n" unless $self->{mime};
                $self->_extract_mid; # v1 w/o Xapian
        }
 }
 
 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;