my $self = {
-inbox => $v2ibx,
im => undef, # PublicInbox::Import
- xap_rw => undef, # PublicInbox::V2SearchIdx
- xap_ro => undef,
partitions => $nparts,
parallel => 1,
transact_bytes => 0,
$num = $self->{skel}->{mm}->mid_insert($$mid0);
}
}
- my @cur = $hdr->header_raw('Message-Id');
- $hdr->header_set('Message-Id', "<$$mid0>", @cur);
+ PublicInbox::Import::append_mid($hdr, $$mid0);
$num;
}
$skel->_msgmap_init->{dbh}->begin_work;
}
-sub remove {
- my ($self, $mime, $cmt_msg) = @_;
+sub purge_oids {
+ my ($self, $purge) = @_; # $purge = { $object_id => 1, ... }
+ $self->done;
+ my $pfx = "$self->{-inbox}->{mainrepo}/git";
+ foreach my $i (0..$self->{max_git}) {
+ my $git = PublicInbox::Git->new("$pfx/$i.git");
+ my $im = $self->import_init($git, 0);
+ $im->purge_oids($purge);
+ }
+}
+
+sub remove_internal {
+ my ($self, $mime, $cmt_msg, $purge) = @_;
$self->barrier;
$self->idx_init;
- my $im = $self->importer;
+ my $im = $self->importer unless $purge;
my $ibx = $self->{-inbox};
my $srch = $ibx->search;
my $cid = content_id($mime);
# no bugs in our deduplication code:
$removed = $smsg;
$removed->{mime} = $cur;
- $im->remove(\$orig, $cmt_msg);
+ my $oid = $smsg->{blob};
+ if ($purge) {
+ $purge->{$oid} = 1;
+ } else {
+ $im->remove(\$orig, $cmt_msg);
+ }
$orig = undef;
$removed->num; # memoize this for callers
- my $oid = $smsg->{blob};
foreach my $idx (@$parts, $skel) {
$idx->remote_remove($oid, $mid);
}
});
$self->barrier;
}
+ if ($purge && scalar keys %$purge) {
+ purge_oids($self, $purge);
+ }
$removed;
}
+sub remove {
+ my ($self, $mime, $cmt_msg) = @_;
+ remove_internal($self, $mime, $cmt_msg);
+}
+
+sub purge {
+ my ($self, $mime) = @_;
+ remove_internal($self, $mime, undef, {});
+}
+
+
sub done {
my ($self) = @_;
my $locked = defined $self->{idx_parts};
use File::Temp qw(tempfile);
use PublicInbox::Spawn qw(spawn);
- my ($ah, $an) = tempfile('email-cur-XXXXXXXX');
+ my ($ah, $an) = tempfile('email-cur-XXXXXXXX', TMPDIR => 1);
print $ah $cur->as_string or die "print: $!";
close $ah or die "close: $!";
- my ($bh, $bn) = tempfile('email-new-XXXXXXXX');
+ my ($bh, $bn) = tempfile('email-new-XXXXXXXX', TMPDIR => 1);
PublicInbox::Import::drop_unwanted_headers($new);
print $bh $new->as_string or die "print: $!";
close $bh or die "close: $!";
my $mids = mids($mime->header_obj);
my $cid = content_id($mime);
foreach my $mid (@$mids) {
- $D->{$mid.$cid} = 1;
+ $D->{"$mid\0$cid"} = 1;
}
}
my $num = -1;
my $del = 0;
foreach my $mid (@$mids) {
- $del += (delete $D->{$mid.$cid} || 0);
+ $del += (delete $D->{"$mid\0$cid"} || 0);
my $n = $mm_tmp->num_for($mid);
if (defined $n && $n > $num) {
$mid0 = $mid;
}
my ($min, $max) = $mm_tmp->minmax;
defined $max and die "leftover article numbers at $min..$max\n";
+ my @d = sort keys %$D;
+ if (@d) {
+ warn "BUG: ", scalar(@d)," unseen deleted messages marked\n";
+ foreach (@d) {
+ my ($mid, undef) = split(/\0/, $_, 2);
+ warn "<$mid>\n";
+ }
+ }
}
1;