use warnings;
use base qw(PublicInbox::Search PublicInbox::Lock);
use PublicInbox::MIME;
+use PublicInbox::InboxWritable;
use PublicInbox::MID qw/mid_clean id_compress mid_mime mids references/;
use PublicInbox::MsgIter;
use Carp qw(croak);
require PublicInbox::Git;
use constant {
- PERM_UMASK => 0,
- OLD_PERM_GROUP => 1,
- OLD_PERM_EVERYBODY => 2,
- PERM_GROUP => 0660,
- PERM_EVERYBODY => 0664,
BATCH_BYTES => 1_000_000,
DEBUG => !!$ENV{DEBUG},
};
PublicInbox::AltId->new($ibx, $_);
} @$altid ];
}
- $git = $ibx->git;
- } else {
- $git = PublicInbox::Git->new($git_dir); # v1 only
+ } else { # v1
+ $ibx = { mainrepo => $git_dir, version => 1 };
}
+ $ibx = PublicInbox::InboxWritable->new($ibx);
require Search::Xapian::WritableDatabase;
my $self = bless {
mainrepo => $mainrepo,
- git => $git,
+ -inbox => $ibx,
+ git => $ibx->git,
-altid => $altid,
version => $version,
}, $class;
- my $perm = $self->_git_config_perm;
- my $umask = _umask_for($perm);
- $self->{umask} = $umask;
+ $ibx->umask_prepare;
if ($version == 1) {
$self->{lock_path} = "$mainrepo/ssoma.lock";
} elsif ($version == 2) {
my $lines = $values->[PublicInbox::Search::LINES];
add_val($doc, PublicInbox::Search::LINES, $lines);
- my $yyyymmdd = strftime('%Y%m%d', gmtime($ts));
+ my $ds = $values->[PublicInbox::Search::DS];
+ add_val($doc, PublicInbox::Search::DS, $ds);
+ my $yyyymmdd = strftime('%Y%m%d', gmtime($ds));
add_val($doc, PublicInbox::Search::YYYYMMDD, $yyyymmdd);
}
}
my $lines = $mime->body_raw =~ tr!\n!\n!;
- my @values = ($smsg->ts, $num, $bytes, $lines);
+ my @values = ($smsg->ds, $num, $bytes, $lines, $smsg->ts);
add_values($doc, \@values);
my $tg = $self->term_generator;
# populates smsg->references for smsg->to_doc_data
my $refs = parse_references($smsg);
- $mid0 = $mids->[0] unless defined $mid0;
+ $mid0 = $mids->[0] unless defined $mid0; # v1 compatibility
my $data = $smsg->to_doc_data($oid, $mid0);
foreach my $mid (@$mids) {
$tg->index_text($mid, 1, 'XM');
}
}
+ $self->delete_article($num) if defined $num; # for reindexing
if ($skel) {
push @values, $mids, $xpath, $data;
$skel->index_skeleton(\@values);
$doc->add_boolean_term('Q' . $_) foreach @$mids;
+ $doc->add_boolean_term('XNUM' . $num) if defined $num;
$doc_id = $self->{xdb}->add_document($doc);
} else {
$doc_id = link_and_save($self, $doc, $mids, $refs,
$doc_id;
}
+# returns begin and end PostingIterator
+sub find_doc_ids {
+ my ($self, $termval) = @_;
+ my $db = $self->{xdb};
+
+ ($db->postlist_begin($termval), $db->postlist_end($termval));
+}
+
sub batch_do {
my ($self, $termval, $cb) = @_;
my $batch_size = 1000; # don't let @ids grow too large to avoid OOM
}
}
+sub delete_article {
+ my ($self, $num) = @_;
+ my $ndel = 0;
+ batch_do($self, 'XNUM' . $num, sub {
+ my ($ids) = @_;
+ $ndel += scalar @$ids;
+ $self->{xdb}->delete_document($_) for @$ids;
+ });
+}
+
# MID is a hint in V2
sub remove_by_oid {
my ($self, $oid, $mid) = @_;
1;
});
}
+ if ($vivified > 1) {
+ my $id = '<'.join('> <', @$mids).'>';
+ warn "BUG: vivified multiple ($vivified) ghosts for $id\n";
+ }
# not really important, but we return any vivified ghost docid, here:
return $doc_id if defined $doc_id;
link_doc($self, $doc, $refs, $old_tid);
sub index_sync {
my ($self, $opts) = @_;
- with_umask($self, sub { $self->_index_sync($opts) });
+ $self->{-inbox}->with_umask(sub { $self->_index_sync($opts) })
}
sub batch_adjust ($$$$) {
});
}
-sub _read_git_config_perm {
- my ($self) = @_;
- my @cmd = qw(config);
- if ($self->{version} == 2) {
- push @cmd, "--file=$self->{mainrepo}/all.git/config";
- }
- my $fh = $self->{git}->popen(@cmd, 'core.sharedRepository');
- local $/ = "\n";
- my $perm = <$fh>;
- chomp $perm if defined $perm;
- $perm;
-}
-
-sub _git_config_perm {
- my $self = shift;
- my $perm = scalar @_ ? $_[0] : _read_git_config_perm($self);
- return PERM_GROUP if (!defined($perm) || $perm eq '');
- return PERM_UMASK if ($perm eq 'umask');
- return PERM_GROUP if ($perm eq 'group');
- if ($perm =~ /\A(?:all|world|everybody)\z/) {
- return PERM_EVERYBODY;
- }
- return PERM_GROUP if ($perm =~ /\A(?:true|yes|on|1)\z/);
- return PERM_UMASK if ($perm =~ /\A(?:false|no|off|0)\z/);
-
- my $i = oct($perm);
- return PERM_UMASK if ($i == PERM_UMASK);
- return PERM_GROUP if ($i == OLD_PERM_GROUP);
- return PERM_EVERYBODY if ($i == OLD_PERM_EVERYBODY);
-
- if (($i & 0600) != 0600) {
- die "core.sharedRepository mode invalid: ".
- sprintf('%.3o', $i) . "\nOwner must have permissions\n";
- }
- ($i & 0666);
-}
-
-sub _umask_for {
- my ($perm) = @_; # _git_config_perm return value
- my $rv = $perm;
- return umask if $rv == 0;
-
- # set +x bit if +r or +w were set
- $rv |= 0100 if ($rv & 0600);
- $rv |= 0010 if ($rv & 0060);
- $rv |= 0001 if ($rv & 0006);
- (~$rv & 0777);
-}
-
-sub with_umask {
- my ($self, $cb) = @_;
- my $old = umask $self->{umask};
- my $rv = eval { $cb->() };
- my $err = $@;
- umask $old;
- die $err if $err;
- $rv;
-}
-
sub DESTROY {
# order matters for unlocking
$_[0]->{xdb} = undef;