v2writable: go backwards through alternate Message-IDs

[public-inbox.git] / lib / PublicInbox / V2Writable.pm

diff --git a/lib/PublicInbox/V2Writable.pm b/lib/PublicInbox/V2Writable.pm
index 9b280c686b547e303dde7245a34d6d9ed93b6a6d..34f13e26cb2531ebda1da78266357f5c00aa01d2 100644 (file)
--- a/lib/PublicInbox/V2Writable.pm
+++ b/lib/PublicInbox/V2Writable.pm
@@ -53,8 +53,6 @@ sub new {
        my $self = {
                -inbox => $v2ibx,
                im => undef, #  PublicInbox::Import
-               xap_rw => undef, # PublicInbox::V2SearchIdx
-               xap_ro => undef,
                partitions => $nparts,
                parallel => 1,
                transact_bytes => 0,
@@ -69,7 +67,9 @@ sub init_inbox {
        my ($self, $parallel) = @_;
        $self->{parallel} = $parallel;
        $self->idx_init;
-       $self->git_init(0);
+       my $max_git = -1;
+       git_dir_latest($self, \$max_git);
+       $self->git_init($max_git >= 0 ? $max_git : 0);
        $self->done;
 }
 
@@ -137,7 +137,7 @@ sub num_for {
                warn "<$mid> reused for mismatched content\n";
 
                # try the rest of the mids
-               foreach my $i (1..$#$mids) {
+               for(my $i = $#$mids; $i >= 1; $i--) {
                        my $m = $mids->[$i];
                        $num = $self->{skel}->{mm}->mid_insert($m);
                        if (defined $num) {
@@ -209,11 +209,22 @@ sub idx_init {
        $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);
@@ -245,11 +256,15 @@ sub remove {
                                # 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);
                                }
@@ -258,9 +273,23 @@ sub remove {
                });
                $self->barrier;
        }
+       if ($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 $self->{idx_parts};
@@ -333,7 +362,6 @@ sub git_init {
        my ($self, $new) = @_;
        my $pfx = "$self->{-inbox}->{mainrepo}/git";
        my $git_dir = "$pfx/$new.git";
-       die "$git_dir exists\n" if -e $git_dir;
        my @cmd = (qw(git init --bare -q), $git_dir);
        PublicInbox::Import::run_die(\@cmd);
 
@@ -524,7 +552,7 @@ sub reindex_oid {
                $num = $$regen--;
                die "BUG: ran out of article numbers\n" if $num <= 0;
                my $mm = $self->{skel}->{mm};
-               foreach my $mid (@$mids) {
+               foreach my $mid (reverse @$mids) {
                        if ($mm->mid_set($num, $mid) == 1) {
                                $mid0 = $mid;
                                last;
@@ -532,7 +560,11 @@ sub reindex_oid {
                }
                if (!defined($mid0)) {
                        my $id = '<' . join('> <', @$mids) . '>';
-                       warn "Message-Id $id unusable for $num\n";
+                       warn "Message-ID $id unusable for $num\n";
+                       foreach my $mid (@$mids) {
+                               defined(my $n = $mm->num_for($mid)) or next;
+                               warn "#$n previously mapped for <$mid>\n";
+                       }
                }
        }
 
@@ -595,6 +627,7 @@ sub reindex {
                for (my $cur = $max_git; $cur >= 0; $cur--) {
                        die "already reindexing!\n" if $self->{reindex_pipe};
                        my $git = PublicInbox::Git->new("$pfx/$cur.git");
+                       -d $git->{git_dir} or next; # missing parts are fine
                        chomp($tip = $git->qx('rev-parse', $head)) unless $tip;
                        my $h = $cur == $max_git ? $tip : $head;
                        my @count = ('rev-list', '--count', $h, '--', 'm');
@@ -616,6 +649,7 @@ sub reindex {
                die "already reindexing!\n" if delete $self->{reindex_pipe};
                my $cmt;
                my $git_dir = "$pfx/$cur.git";
+               -d $git_dir or next; # missing parts are fine
                my $git = PublicInbox::Git->new($git_dir);
                my $h = $cur == $max_git ? $tip : $head;
                my $fh = $self->{reindex_pipe} = $git->popen(@cmd, $h);
@@ -631,8 +665,17 @@ sub reindex {
                }
                delete $self->{reindex_pipe};
        }
+       my $gaps;
+       if ($regen && $$regen != 0) {
+               warn "W: leftover article number ($$regen)\n";
+               $gaps = 1;
+       }
        my ($min, $max) = $mm_tmp->minmax;
-       defined $max and die "leftover article numbers at $min..$max\n";
+       if (defined $max) {
+               warn "W: leftover article numbers at $min..$max\n";
+               $gaps = 1;
+       }
+       warn "W: were old git partitions deleted?\n" if $gaps;
        my @d = sort keys %$D;
        if (@d) {
                warn "BUG: ", scalar(@d)," unseen deleted messages marked\n";