my $PACKING_FACTOR = 0.4;
# assume 2 cores if GNU nproc(1) is not available
-sub nproc () {
- int($ENV{NPROC} || `nproc 2>/dev/null` || 2);
+sub nproc_parts () {
+ my $n = int($ENV{NPROC} || `nproc 2>/dev/null` || 2);
+ # subtract for the main process and git-fast-import
+ $n -= 1;
+ $n < 1 ? 1 : $n;
}
sub count_partitions ($) {
rotate_bytes => int((1024 * 1024 * 1024) / $PACKING_FACTOR),
last_commit => [], # git repo -> commit
};
- $self->{partitions} = count_partitions($self) || nproc();
+ $self->{partitions} = count_partitions($self) || nproc_parts();
bless $self, $class;
}
};
# crap, Message-ID is already known, hope somebody just resent:
- $self->barrier;
foreach my $m (@$mids) {
# read-only lookup now safe to do after above barrier
my $existing = $self->lookup_content($mime, $m);
my $hdr = $mime->header_obj;
my $dig = content_digest($mime);
- $$mid0 = PublicInbox::Import::digest2mid($dig);
+ $$mid0 = PublicInbox::Import::digest2mid($dig, $hdr);
my $num = $self->{mm}->mid_insert($$mid0);
unless (defined $num) {
# it's hard to spoof the last Received: header
my @recvd = $hdr->header_raw('Received');
$dig->add("Received: $_") foreach (@recvd);
- $$mid0 = PublicInbox::Import::digest2mid($dig);
+ $$mid0 = PublicInbox::Import::digest2mid($dig, $hdr);
$num = $self->{mm}->mid_insert($$mid0);
# fall back to a random Message-ID and give up determinism:
until (defined($num)) {
$dig->add(rand);
- $$mid0 = PublicInbox::Import::digest2mid($dig);
+ $$mid0 = PublicInbox::Import::digest2mid($dig, $hdr);
warn "using random Message-ID <$$mid0> as fallback\n";
$num = $self->{mm}->mid_insert($$mid0);
}
sub remove_internal {
my ($self, $mime, $cmt_msg, $purge) = @_;
- $self->barrier;
$self->idx_init;
my $im = $self->importer unless $purge;
- my $ibx = $self->{-inbox};
my $over = $self->{over};
my $cid = content_id($mime);
my $parts = $self->{idx_parts};
my %gone;
my ($id, $prev);
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
- my $msg = $ibx->msg_by_smsg($smsg);
+ my $msg = get_blob($self, $smsg);
if (!defined($msg)) {
warn "broken smsg for $mid\n";
next; # continue
$orig = undef;
$self->unindex_oid_remote($oid, $mid);
}
- $self->barrier;
}
if (defined $mark) {
}
}
-sub done {
- my ($self) = @_;
- my $im = delete $self->{im};
- $im->done if $im; # PublicInbox::Import::done
-
- my $mm = $self->{mm};
- $mm->{dbh}->commit if $mm;
-
- # order matters, we can only close {over} after all partitions
- # are done because the partitions also write to {over}
- my $parts = delete $self->{idx_parts};
- if ($parts) {
- $_->remote_commit for @$parts;
- $_->remote_close for @$parts;
- }
-
- my $over = $self->{over};
- $over->commit_lazy;
- $over->disconnect;
-
- if ($mm) {
- $mm->{dbh}->begin_work;
- set_last_commits($self);
- $mm->{dbh}->commit;
- delete $self->{mm};
- }
-
- delete $self->{bnote};
- $self->{transact_bytes} = 0;
- $self->lock_release if $parts;
-}
-
-sub checkpoint {
- my ($self) = @_;
- my $im = $self->{im};
- $im->checkpoint if $im; # PublicInbox::Import::checkpoint
- $self->barrier;
-}
-
sub barrier_init {
my ($self, $n) = @_;
$self->{bnote} or return;
}
}
-# issue a write barrier to ensure all data is visible to other processes
-# and read-only ops. Order of data importance is: git > SQLite > Xapian
-sub barrier {
- my ($self) = @_;
+sub checkpoint ($;$) {
+ my ($self, $wait) = @_;
if (my $im = $self->{im}) {
- $im->barrier;
+ if ($wait) {
+ $im->barrier;
+ } else {
+ $im->checkpoint;
+ }
}
my $parts = $self->{idx_parts};
if ($parts) {
$self->{over}->commit_lazy;
# Now deal with Xapian
- my $barrier = $self->barrier_init(scalar @$parts);
+ if ($wait) {
+ my $barrier = $self->barrier_init(scalar @$parts);
+
+ # each partition needs to issue a barrier command
+ $_->remote_barrier for @$parts;
- # each partition needs to issue a barrier command
- $_->remote_barrier for @$parts;
- $self->barrier_wait($barrier); # wait for each Xapian partition
+ # wait for each Xapian partition
+ $self->barrier_wait($barrier);
+ } else {
+ $_->remote_commit for @$parts;
+ }
# last_commit is special, don't commit these until
# remote partitions are done:
$self->{transact_bytes} = 0;
}
+# issue a write barrier to ensure all data is visible to other processes
+# and read-only ops. Order of data importance is: git > SQLite > Xapian
+sub barrier { checkpoint($_[0], 1) };
+
+sub done {
+ my ($self) = @_;
+ my $im = delete $self->{im};
+ $im->done if $im; # PublicInbox::Import::done
+ checkpoint($self);
+ my $mm = delete $self->{mm};
+ $mm->{dbh}->commit if $mm;
+ my $parts = delete $self->{idx_parts};
+ if ($parts) {
+ $_->remote_close for @$parts;
+ }
+ $self->{over}->disconnect;
+ delete $self->{bnote};
+ $self->{transact_bytes} = 0;
+ $self->lock_release if $parts;
+}
+
sub git_init {
my ($self, $epoch) = @_;
my $pfx = "$self->{-inbox}->{mainrepo}/git";
} else {
$self->{im} = undef;
$im->done;
- $self->barrier;
$im = undef;
+ $self->checkpoint;
my $git_dir = $self->git_init(++$self->{epoch_max});
my $git = PublicInbox::Git->new($git_dir);
return $self->import_init($git, 0);
unlink($an, $bn);
}
+sub get_blob ($$) {
+ my ($self, $smsg) = @_;
+ if (my $im = $self->{im}) {
+ my $msg = $im->cat_blob($smsg->{blob});
+ return $msg if $msg;
+ }
+ # older message, should be in alternates
+ my $ibx = $self->{-inbox};
+ $ibx->msg_by_smsg($smsg);
+}
+
sub lookup_content {
my ($self, $mime, $mid) = @_;
- my $ibx = $self->{-inbox};
my $over = $self->{over};
my $cid = content_id($mime);
my $found;
my ($id, $prev);
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
- my $msg = $ibx->msg_by_smsg($smsg);
+ my $msg = get_blob($self, $smsg);
if (!defined($msg)) {
warn "broken smsg for $mid\n";
next;
my %gone;
my ($id, $prev);
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
- $gone{$smsg->num} = 1 if $oid eq $smsg->{blob};
+ $gone{$smsg->{num}} = 1 if $oid eq $smsg->{blob};
1; # continue
}
my $n = scalar keys %gone;
}
$self->{unindexed}->{$_}++ foreach keys %gone;
$self->unindex_oid_remote($oid, $mid);
- $self->barrier;
}
}
sub unindex {
my ($self, $opts, $git, $unindex_range) = @_;
my $un = $self->{unindexed} ||= {}; # num => removal count
- $self->barrier;
my $before = scalar keys %$un;
my @cmd = qw(log --raw -r
--no-notes --no-color --no-abbrev --no-renames);
sub index_sync {
my ($self, $opts) = @_;
$opts ||= {};
- my $ibx = $self->{-inbox};
my $epoch_max;
my $latest = git_dir_latest($self, \$epoch_max);
return unless defined $latest;