treewide: run update-copyrights from gnulib for 2019

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

diff --git a/lib/PublicInbox/Over.pm b/lib/PublicInbox/Over.pm
index b2f68835a42738cce7613c6b544e6d93a9b0123d..b9b02f967b324173fe66caf5e02fbb9df864e892 100644 (file)
--- a/lib/PublicInbox/Over.pm
+++ b/lib/PublicInbox/Over.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2018 all contributors <meta@public-inbox.org>
+# Copyright (C) 2018-2020 all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # for XOVER, OVER in NNTP, and feeds/homepage/threads in PSGI
@@ -11,19 +11,19 @@ use DBI;
 use DBD::SQLite;
 use PublicInbox::SearchMsg;
 use Compress::Zlib qw(uncompress);
+use constant DEFAULT_LIMIT => 1000;
 
 sub dbh_new {
-       my ($self) = @_;
-       my $ro = ref($self) eq 'PublicInbox::Over';
+       my ($self, $rw) = @_;
        my $f = $self->{filename};
-       if (!$ro && !-f $f) { # SQLite defaults mode to 0644, we want 0666
+       if ($rw && !-f $f) { # SQLite defaults mode to 0644, we want 0666
                open my $fh, '+>>', $f or die "failed to open $f: $!";
        }
        my $dbh = DBI->connect("dbi:SQLite:dbname=$f",'','', {
                AutoCommit => 1,
                RaiseError => 1,
                PrintError => 0,
-               ReadOnly => $ro,
+               ReadOnly => !$rw,
                sqlite_use_immediate_transaction => 1,
        });
        $dbh->{sqlite_unicode} = 1;
@@ -39,13 +39,16 @@ sub disconnect { $_[0]->{dbh} = undef }
 
 sub connect { $_[0]->{dbh} ||= $_[0]->dbh_new }
 
-sub load_from_row {
-       my ($smsg) = @_;
+sub load_from_row ($;$) {
+       my ($smsg, $cull) = @_;
        bless $smsg, 'PublicInbox::SearchMsg';
        if (defined(my $data = delete $smsg->{ddd})) {
                $data = uncompress($data);
                utf8::decode($data);
-               $smsg->load_from_data($data);
+               PublicInbox::SearchMsg::load_from_data($smsg, $data);
+
+               # saves over 600K for 1000+ message threads
+               PublicInbox::SearchMsg::psgi_cull($smsg) if $cull;
        }
        $smsg
 }
@@ -53,10 +56,11 @@ sub load_from_row {
 sub do_get {
        my ($self, $sql, $opts, @args) = @_;
        my $dbh = $self->connect;
-       my $lim = (($opts->{limit} || 0) + 0) || 1000;
+       my $lim = (($opts->{limit} || 0) + 0) || DEFAULT_LIMIT;
        $sql .= "LIMIT $lim";
        my $msgs = $dbh->selectall_arrayref($sql, { Slice => {} }, @args);
-       load_from_row($_) for @$msgs;
+       my $cull = $opts->{cull};
+       load_from_row($_, $cull) for @$msgs;
        $msgs
 }
 
@@ -81,6 +85,7 @@ sub nothing () { wantarray ? (0, []) : [] };
 sub get_thread {
        my ($self, $mid, $prev) = @_;
        my $dbh = $self->connect;
+       my $opts = { cull => 1 };
 
        my $id = $dbh->selectrow_array(<<'', undef, $mid);
 SELECT id FROM msgid WHERE mid = ? LIMIT 1
@@ -97,21 +102,47 @@ ORDER BY num ASC LIMIT 1
 SELECT tid,sid FROM over WHERE num = ? LIMIT 1
 
        defined $tid or return nothing; # $sid may be undef
+
+       my $cond_all = '(tid = ? OR sid = ?) AND num > ?';
        my $sort_col = 'ds';
        $num = 0;
-       if ($prev) {
+       if ($prev) { # mboxrd stream, only
                $num = $prev->{num} || 0;
                $sort_col = 'num';
        }
-       my $cond = '(tid = ? OR sid = ?) AND num > ?';
-       my $msgs = do_get($self, <<"", {}, $tid, $sid, $num);
-SELECT num,ts,ds,ddd FROM over WHERE $cond ORDER BY $sort_col ASC
 
-       return $msgs unless wantarray;
+       my $cols = 'num,ts,ds,ddd';
+       unless (wantarray) {
+               return do_get($self, <<"", $opts, $tid, $sid, $num);
+SELECT $cols FROM over WHERE $cond_all
+ORDER BY $sort_col ASC
 
-       my $nr = $dbh->selectrow_array(<<"", undef, $tid, $sid, $num);
-SELECT COUNT(num) FROM over WHERE $cond
+       }
 
+       # HTML view always wants an array and never uses $prev,
+       # but the mbox stream never wants an array and always has $prev
+       die '$prev not supported with wantarray' if $prev;
+       my $nr = $dbh->selectrow_array(<<"", undef, $tid, $sid, $num);
+SELECT COUNT(num) FROM over WHERE $cond_all
+
+       # giant thread, prioritize strict (tid) matches and throw
+       # in the loose (sid) matches at the end
+       my $msgs = do_get($self, <<"", $opts, $tid, $num);
+SELECT $cols FROM over WHERE tid = ? AND num > ?
+ORDER BY $sort_col ASC
+
+       # do we have room for loose matches? get the most recent ones, first:
+       my $lim = DEFAULT_LIMIT - scalar(@$msgs);
+       if ($lim > 0) {
+               $opts->{limit} = $lim;
+               my $loose = do_get($self, <<"", $opts, $tid, $sid, $num);
+SELECT $cols FROM over WHERE tid != ? AND sid = ? AND num > ?
+ORDER BY $sort_col DESC
+
+               # TODO separate strict and loose matches here once --reindex
+               # is fixed to preserve `tid' properly
+               push @$msgs, @$loose;
+       }
        ($nr, $msgs);
 }