X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FOver.pm;h=06ea439d484f78bbd51350b6ce490f9b0ec43967;hb=af0b0fb7a454470a32c452119d0392e0dedb3fe1;hp=f32743c05a8a461baf0c4f5a52c983d5fecd25af;hpb=0e68dbad3dc5e3fbc44e8ba8be576b81455d3359;p=public-inbox.git diff --git a/lib/PublicInbox/Over.pm b/lib/PublicInbox/Over.pm index f32743c0..06ea439d 100644 --- a/lib/PublicInbox/Over.pm +++ b/lib/PublicInbox/Over.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2018-2020 all contributors +# Copyright (C) 2018-2021 all contributors # License: AGPL-3.0+ # for XOVER, OVER in NNTP, and feeds/homepage/threads in PSGI @@ -6,7 +6,7 @@ # tweaked/updated over time and rebuilt. package PublicInbox::Over; use strict; -use warnings; +use v5.10.1; use DBI; use DBD::SQLite; use PublicInbox::Smsg; @@ -18,7 +18,9 @@ sub dbh_new { my $f = delete $self->{filename}; if (!-f $f) { # SQLite defaults mode to 0644, we want 0666 if ($rw) { + require PublicInbox::Spawn; open my $fh, '+>>', $f or die "failed to open $f: $!"; + PublicInbox::Spawn::nodatacow_fd(fileno($fh)); } else { $self->{filename} = $f; # die on stat() below: } @@ -40,7 +42,28 @@ sub dbh_new { $st = pack('dd', $st[0], $st[1]); } while ($st ne $self->{st} && $tries++ < 3); warn "W: $f: .st_dev, .st_ino unstable\n" if $st ne $self->{st}; - $dbh->do('PRAGMA synchronous = OFF') if ($rw // 0) > 1; + + if ($rw) { + # TRUNCATE reduces I/O compared to the default (DELETE). + # + # Do not use WAL by default since we expect the case + # where any users may read via read-only daemons + # (-httpd/-imapd/-nntpd); but only a single user has + # write permissions for -watch/-mda. + # + # Read-only WAL support in SQLite 3.22.0 (2018-01-22) + # doesn't do what we need: it is only intended for + # immutable read-only media (e.g. CD-ROM) and not + # usable for our use case described above. + # + # If an admin is willing to give read-only daemons R/W + # permissions; they can enable WAL manually and we will + # respect that by not clobbering it. + my $jm = $dbh->selectrow_array('PRAGMA journal_mode'); + $dbh->do('PRAGMA journal_mode = TRUNCATE') if $jm ne 'wal'; + + $dbh->do('PRAGMA synchronous = OFF') if $rw > 1; + } $dbh; } @@ -49,21 +72,21 @@ sub new { bless { filename => $f }, $class; } -sub disconnect { +sub dbh_close { my ($self) = @_; if (my $dbh = delete $self->{dbh}) { + delete $self->{-get_art}; $self->{filename} = $dbh->sqlite_db_filename; } } -sub connect { $_[0]->{dbh} //= $_[0]->dbh_new } +sub dbh ($) { $_[0]->{dbh} //= $_[0]->dbh_new } # dbh_new may be subclassed sub load_from_row ($;$) { my ($smsg, $cull) = @_; bless $smsg, 'PublicInbox::Smsg'; if (defined(my $data = delete $smsg->{ddd})) { $data = uncompress($data); - utf8::decode($data); PublicInbox::Smsg::load_from_data($smsg, $data); # saves over 600K for 1000+ message threads @@ -74,10 +97,9 @@ sub load_from_row ($;$) { sub do_get { my ($self, $sql, $opts, @args) = @_; - my $dbh = $self->connect; my $lim = (($opts->{limit} || 0) + 0) || DEFAULT_LIMIT; $sql .= "LIMIT $lim"; - my $msgs = $dbh->selectall_arrayref($sql, { Slice => {} }, @args); + my $msgs = dbh($self)->selectall_arrayref($sql, { Slice => {} }, @args); my $cull = $opts->{cull}; load_from_row($_, $cull) for @$msgs; $msgs @@ -99,11 +121,20 @@ ORDER BY num ASC } +sub get_all { + my $self = shift; + my $nr = scalar(@_) or return []; + my $in = '?' . (',?' x ($nr - 1)); + do_get($self, <<"", { cull => 1, limit => $nr }, @_); +SELECT num,ts,ds,ddd FROM over WHERE num IN ($in) + +} + sub nothing () { wantarray ? (0, []) : [] }; sub get_thread { my ($self, $mid, $prev) = @_; - my $dbh = $self->connect; + my $dbh = dbh($self); my $opts = { cull => 1 }; my $id = $dbh->selectrow_array(<<'', undef, $mid); @@ -165,6 +196,35 @@ ORDER BY $sort_col DESC ($nr, $msgs); } +# strict `tid' matches, only, for thread-expanded mbox.gz search results +# and future CLI interface +# returns true if we have IDs, undef if not +sub expand_thread { + my ($self, $ctx) = @_; + my $dbh = dbh($self); + do { + defined(my $num = $ctx->{ids}->[0]) or return; + my ($tid) = $dbh->selectrow_array(<<'', undef, $num); +SELECT tid FROM over WHERE num = ? + + if (defined($tid)) { + my $sql = <<''; +SELECT num FROM over WHERE tid = ? AND num > ? +ORDER BY num ASC LIMIT 1000 + + my $xids = $dbh->selectcol_arrayref($sql, undef, $tid, + $ctx->{prev} // 0); + if (scalar(@$xids)) { + $ctx->{prev} = $xids->[-1]; + $ctx->{xids} = $xids; + return 1; # success + } + } + $ctx->{prev} = 0; + shift @{$ctx->{ids}}; + } while (1); +} + sub recent { my ($self, $opts, $after, $before) = @_; my ($s, @v); @@ -184,31 +244,46 @@ sub recent { $s = '+num > 0 ORDER BY ts DESC'; } } - my $msgs = do_get($self, <<"", $opts, @v); + do_get($self, <<"", $opts, @v); SELECT ts,ds,ddd FROM over WHERE $s - return $msgs unless wantarray; - - my $nr = $self->{dbh}->selectrow_array(<<''); -SELECT COUNT(num) FROM over WHERE num > 0 - - ($nr, $msgs); } sub get_art { my ($self, $num) = @_; - my $dbh = $self->connect; - my $sth = $dbh->prepare_cached(<<'', undef, 1); -SELECT num,ds,ts,ddd FROM over WHERE num = ? LIMIT 1 + # caching $sth ourselves is faster than prepare_cached + my $sth = $self->{-get_art} //= dbh($self)->prepare(<<''); +SELECT num,tid,ds,ts,ddd FROM over WHERE num = ? LIMIT 1 $sth->execute($num); my $smsg = $sth->fetchrow_hashref; $smsg ? load_from_row($smsg) : undef; } +sub get_xref3 { + my ($self, $num, $raw) = @_; + my $dbh = dbh($self); + my $sth = $dbh->prepare_cached(<<'', undef, 1); +SELECT ibx_id,xnum,oidbin FROM xref3 WHERE docid = ? ORDER BY ibx_id,xnum ASC + + $sth->execute($num); + my $rows = $sth->fetchall_arrayref; + return $rows if $raw; + my $eidx_key_sth = $dbh->prepare_cached(<<'', undef, 1); +SELECT eidx_key FROM inboxes WHERE ibx_id = ? + + [ map { + my $r = $_; + $eidx_key_sth->execute($r->[0]); + my $eidx_key = $eidx_key_sth->fetchrow_array; + $eidx_key //= "missing://ibx_id=$r->[0]"; + "$eidx_key:$r->[1]:".unpack('H*', $r->[2]); + } @$rows ]; +} + sub next_by_mid { my ($self, $mid, $id, $prev) = @_; - my $dbh = $self->connect; + my $dbh = dbh($self); unless (defined $$id) { my $sth = $dbh->prepare_cached(<<'', undef, 1); @@ -226,19 +301,13 @@ ORDER BY num ASC LIMIT 1 $sth->execute($$id, $$prev); my $num = $sth->fetchrow_array or return; $$prev = $num; - - $sth = $dbh->prepare_cached(<<"", undef, 1); -SELECT num,ts,ds,ddd FROM over WHERE num = ? LIMIT 1 - - $sth->execute($num); - my $smsg = $sth->fetchrow_hashref or return; - load_from_row($smsg); + get_art($self, $num); } # IMAP search, this is limited by callers to UID_SLICE size (50K) sub uid_range { my ($self, $beg, $end, $sql) = @_; - my $dbh = $self->connect; + my $dbh = dbh($self); my $q = 'SELECT num FROM over WHERE num >= ? AND num <= ?'; # This is read-only, anyways; but caller should verify it's @@ -250,7 +319,7 @@ sub uid_range { sub max { my ($self) = @_; - my $sth = $self->connect->prepare_cached(<<'', undef, 1); + my $sth = dbh($self)->prepare_cached(<<'', undef, 1); SELECT MAX(num) FROM over WHERE num > 0 $sth->execute; @@ -259,7 +328,7 @@ SELECT MAX(num) FROM over WHERE num > 0 sub imap_exists { my ($self, $uid_base, $uid_end) = @_; - my $sth = $self->connect->prepare_cached(<<'', undef, 1); + my $sth = dbh($self)->prepare_cached(<<'', undef, 1); SELECT COUNT(num) FROM over WHERE num > ? AND num <= ? $sth->execute($uid_base, $uid_end); @@ -274,10 +343,7 @@ sub check_inodes { my $st = pack('dd', $st[0], $st[1]); # don't actually reopen, just let {dbh} be recreated later - if ($st ne ($self->{st} // $st)) { - delete($self->{dbh}); - $self->{filename} = $f; - } + dbh_close($self) if $st ne ($self->{st} // $st); } else { warn "W: stat $f: $!\n"; }