# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
# This interface wraps and mimics PublicInbox::Import
+# Used to write to V2 inboxes (see L<public-inbox-v2-format(5)>).
package PublicInbox::V2Writable;
use strict;
use warnings;
use PublicInbox::Inbox;
use PublicInbox::OverIdx;
use PublicInbox::Msgmap;
-use PublicInbox::Spawn;
+use PublicInbox::Spawn qw(spawn);
+use PublicInbox::SearchIdx;
use IO::Handle;
# an estimate of the post-packed size to the raw uncompressed size
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 ($) {
+ my ($creat_opt) = @_;
+ if (ref($creat_opt) eq 'HASH') {
+ if (defined(my $n = $creat_opt->{nproc})) {
+ return $n
+ }
+ }
+
+ my $n = int($ENV{NPROC} || `nproc 2>/dev/null` || 2);
+ # subtract for the main process and git-fast-import
+ $n -= 1;
+ $n < 1 ? 1 : $n;
}
sub count_partitions ($) {
}
sub new {
+ # $creat may be any true value, or 0/undef. A hashref is true,
+ # and $creat->{nproc} may be set to an integer
my ($class, $v2ibx, $creat) = @_;
my $dir = $v2ibx->{mainrepo} or die "no mainrepo in inbox\n";
unless (-d $dir) {
}
$v2ibx = PublicInbox::InboxWritable->new($v2ibx);
+ $v2ibx->umask_prepare;
my $xpfx = "$dir/xap" . PublicInbox::Search::SCHEMA_VERSION;
my $self = {
im => undef, # PublicInbox::Import
parallel => 1,
transact_bytes => 0,
+ current_info => '',
xpfx => $xpfx,
over => PublicInbox::OverIdx->new("$xpfx/over.sqlite3", 1),
lock_path => "$dir/inbox.lock",
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($creat);
bless $self, $class;
}
sub init_inbox {
- my ($self, $parallel) = @_;
+ my ($self, $parallel, $skip) = @_;
$self->{parallel} = $parallel;
$self->idx_init;
my $epoch_max = -1;
git_dir_latest($self, \$epoch_max);
+ if (defined $skip && $epoch_max == -1) {
+ $epoch_max = $skip;
+ }
$self->git_init($epoch_max >= 0 ? $epoch_max : 0);
$self->done;
}
# mimics Import::add and wraps it for v2
sub add {
my ($self, $mime, $check_cb) = @_;
+ $self->{-inbox}->with_umask(sub {
+ _add($self, $mime, $check_cb)
+ });
+}
+
+sub _add {
+ my ($self, $mime, $check_cb) = @_;
# spam check:
if ($check_cb) {
return if $existing;
}
+ # AltId may pre-populate article numbers (e.g. X-Mail-Count
+ # or NNTP article number), use that article number if it's
+ # not in Over.
+ my $altid = $self->{-inbox}->{altid};
+ if ($altid && grep(/:file=msgmap\.sqlite3\z/, @$altid)) {
+ my $num = $self->{mm}->num_for($mid);
+
+ if (defined $num && !$self->{over}->get_art($num)) {
+ $$mid0 = $mid;
+ return $num;
+ }
+ }
+
# very unlikely:
warn "<$mid> reused for mismatched content\n";
my $hdr = $mime->header_obj;
my $dig = content_digest($mime);
- $$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);
}
# frequently activated.
delete $ibx->{$_} foreach (qw(git mm search));
- if ($self->{parallel}) {
- pipe(my ($r, $w)) or die "pipe failed: $!";
- $self->{bnote} = [ $r, $w ];
- $w->autoflush(1);
- }
+ my $indexlevel = $ibx->{indexlevel};
+ if ($indexlevel && $indexlevel eq 'basic') {
+ $self->{parallel} = 0;
+ }
+
+ if ($self->{parallel}) {
+ pipe(my ($r, $w)) or die "pipe failed: $!";
+ # pipe for barrier notifications doesn't need to be big,
+ # 1031: F_SETPIPE_SZ
+ fcntl($w, 1031, 4096) if $^O eq 'linux';
+ $self->{bnote} = [ $r, $w ];
+ $w->autoflush(1);
+ }
my $over = $self->{over};
$ibx->umask_prepare;
$self->done;
my $pfx = "$self->{-inbox}->{mainrepo}/git";
my $purges = [];
- foreach my $i (0..$self->{epoch_max}) {
- my $git = PublicInbox::Git->new("$pfx/$i.git");
+ my $max = $self->{epoch_max};
+
+ unless (defined($max)) {
+ defined(my $latest = git_dir_latest($self, \$max)) or return;
+ $self->{epoch_max} = $max;
+ }
+ foreach my $i (0..$max) {
+ my $git_dir = "$pfx/$i.git";
+ -d $git_dir or next;
+ my $git = PublicInbox::Git->new($git_dir);
my $im = $self->import_init($git, 0, 1);
$purges->[$i] = $im->purge_oids($purge);
+ $im->done;
}
$purges;
}
+sub content_ids ($) {
+ my ($mime) = @_;
+ my @cids = ( content_id($mime) );
+
+ # Email::MIME->as_string doesn't always round-trip, so we may
+ # use a second content_id
+ my $rt = content_id(PublicInbox::MIME->new(\($mime->as_string)));
+ push @cids, $rt if $cids[0] ne $rt;
+ \@cids;
+}
+
+sub content_matches ($$) {
+ my ($cids, $existing) = @_;
+ my $cid = content_id($existing);
+ foreach (@$cids) {
+ return 1 if $_ eq $cid
+ }
+ 0
+}
+
sub remove_internal {
my ($self, $mime, $cmt_msg, $purge) = @_;
$self->idx_init;
my $im = $self->importer unless $purge;
my $over = $self->{over};
- my $cid = content_id($mime);
+ my $cids = content_ids($mime);
my $parts = $self->{idx_parts};
my $mm = $self->{mm};
my $removed;
}
my $orig = $$msg;
my $cur = PublicInbox::MIME->new($msg);
- if (content_id($cur) eq $cid) {
+ if (content_matches($cids, $cur)) {
$smsg->{mime} = $cur;
$gone{$smsg->{num}} = [ $smsg, \$orig ];
}
sub remove {
my ($self, $mime, $cmt_msg) = @_;
- remove_internal($self, $mime, $cmt_msg);
+ $self->{-inbox}->with_umask(sub {
+ remove_internal($self, $mime, $cmt_msg);
+ });
}
sub purge {
my ($self, $mime) = @_;
- my $purges = remove_internal($self, $mime, undef, {});
+ my $purges = $self->{-inbox}->with_umask(sub {
+ remove_internal($self, $mime, undef, {});
+ }) or return;
$self->idx_init if @$purges; # ->done is called on purges
for my $i (0..$#$purges) {
defined(my $cmt = $purges->[$i]) or next;
delete $self->{bnote};
$self->{transact_bytes} = 0;
$self->lock_release if $parts;
+ $self->{-inbox}->git->cleanup;
}
-sub git_init {
+sub fill_alternates ($$) {
my ($self, $epoch) = @_;
- my $pfx = "$self->{-inbox}->{mainrepo}/git";
- my $git_dir = "$pfx/$epoch.git";
- my @cmd = (qw(git init --bare -q), $git_dir);
- PublicInbox::Import::run_die(\@cmd);
+ my $pfx = "$self->{-inbox}->{mainrepo}/git";
my $all = "$self->{-inbox}->{mainrepo}/all.git";
+ my @cmd;
unless (-d $all) {
- @cmd = (qw(git init --bare -q), $all);
- PublicInbox::Import::run_die(\@cmd);
- @cmd = (qw/git config/, "--file=$all/config",
- 'repack.writeBitmaps', 'true');
- PublicInbox::Import::run_die(\@cmd);
+ PublicInbox::Import::init_bare($all);
}
-
- @cmd = (qw/git config/, "--file=$git_dir/config",
+ @cmd = (qw/git config/, "--file=$pfx/$epoch.git/config",
'include.path', '../../all.git/config');
PublicInbox::Import::run_die(\@cmd);
my $alt = "$all/objects/info/alternates";
- my $new_obj_dir = "../../git/$epoch.git/objects";
my %alts;
+ my @add;
if (-e $alt) {
open(my $fh, '<', $alt) or die "open < $alt: $!\n";
%alts = map { chomp; $_ => 1 } (<$fh>);
}
- return $git_dir if $alts{$new_obj_dir};
+ foreach my $i (0..$epoch) {
+ my $dir = "../../git/$i.git/objects";
+ push @add, $dir if !$alts{$dir} && -d "$pfx/$i.git";
+ }
+ return unless @add;
open my $fh, '>>', $alt or die "open >> $alt: $!\n";
- print $fh "$new_obj_dir\n" or die "print >> $alt: $!\n";
+ foreach my $dir (@add) {
+ print $fh "$dir\n" or die "print >> $alt: $!\n";
+ }
close $fh or die "close $alt: $!\n";
+}
+
+sub git_init {
+ my ($self, $epoch) = @_;
+ my $git_dir = "$self->{-inbox}->{mainrepo}/git/$epoch.git";
+ my @cmd = (qw(git init --bare -q), $git_dir);
+ PublicInbox::Import::run_die(\@cmd);
+ fill_alternates($self, $epoch);
$git_dir
}
if (defined $latest) {
my $git = PublicInbox::Git->new($latest);
my $packed_bytes = $git->packed_bytes;
- if ($packed_bytes >= $self->{rotate_bytes}) {
+ my $unpacked_bytes = $packed_bytes / $PACKING_FACTOR;
+
+ if ($unpacked_bytes >= $self->{rotate_bytes}) {
$epoch = $max + 1;
} else {
$self->{epoch_max} = $max;
sub diff ($$$) {
my ($mid, $cur, $new) = @_;
use File::Temp qw(tempfile);
- use PublicInbox::Spawn qw(spawn);
my ($ah, $an) = tempfile('email-cur-XXXXXXXX', TMPDIR => 1);
print $ah $cur->as_string or die "print: $!";
sub lookup_content {
my ($self, $mime, $mid) = @_;
my $over = $self->{over};
- my $cid = content_id($mime);
- my $found;
+ my $cids = content_ids($mime);
my ($id, $prev);
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
my $msg = get_blob($self, $smsg);
next;
}
my $cur = PublicInbox::MIME->new($msg);
- if (content_id($cur) eq $cid) {
+ if (content_matches($cids, $cur)) {
$smsg->{mime} = $cur;
- $found = $smsg;
- last;
+ return $smsg;
}
+
# XXX DEBUG_DIFF is experimental and may be removed
diff($mid, $cur, $mime) if $ENV{DEBUG_DIFF};
}
- $found;
+ undef;
}
sub atfork_child {
my $mids = mids($mime->header_obj);
my $cid = content_id($mime);
foreach my $mid (@$mids) {
- $D->{"$mid\0$cid"} = 1;
+ $D->{"$mid\0$cid"} = $oid;
}
}
sub reindex_oid {
- my ($self, $mm_tmp, $D, $git, $oid, $regen) = @_;
+ my ($self, $mm_tmp, $D, $git, $oid, $regen, $reindex) = @_;
my $len;
my $msgref = $git->cat_file($oid, \$len);
my $mime = PublicInbox::MIME->new($$msgref);
my $num = -1;
my $del = 0;
foreach my $mid (@$mids) {
- $del += (delete $D->{"$mid\0$cid"} || 0);
+ $del += delete($D->{"$mid\0$cid"}) ? 1 : 0;
my $n = $mm_tmp->num_for($mid);
if (defined $n && $n > $num) {
$mid0 = $mid;
$num = $n;
+ $self->{mm}->mid_set($num, $mid0);
}
}
if (!defined($mid0) && $regen && !$del) {
if (!defined($mid0) || $del) {
if (!defined($mid0) && $del) { # expected for deletes
- $$regen--;
+ $num = $$regen--;
+ $self->{mm}->num_highwater($num) unless $reindex;
return
}
$heads;
}
-sub is_ancestor ($$$) {
- my ($git, $cur, $tip) = @_;
- return 0 unless $git->check($cur);
- my $cmd = [ 'git', "--git-dir=$git->{git_dir}",
- qw(merge-base --is-ancestor), $cur, $tip ];
- my $pid = spawn($cmd);
- defined $pid or die "spawning ".join(' ', @$cmd)." failed: $!";
- waitpid($pid, 0) == $pid or die join(' ', @$cmd) .' did not finish';
- $? == 0;
-}
+*is_ancestor = *PublicInbox::SearchIdx::is_ancestor;
sub index_prepare {
my ($self, $opts, $epoch_max, $ranges) = @_;
my $git_dir = git_dir_n($self, $i);
-d $git_dir or next; # missing parts are fine
my $git = PublicInbox::Git->new($git_dir);
- chomp(my $tip = $git->qx('rev-parse', $head));
+ chomp(my $tip = $git->qx(qw(rev-parse -q --verify), $head));
+ next if $?; # new repo
my $range;
if (defined(my $cur = $ranges->[$i])) {
$range = "$cur..$tip";
$ranges->[$i] = $range;
# can't use 'rev-list --count' if we use --diff-filter
- my $fh = $git->popen(qw(log --pretty=tformat:%h
+ my $fh = $git->popen(qw(log --pretty=tformat:%H
--no-notes --no-color --no-renames
--diff-filter=AM), $range, '--', 'm');
++$regen_max while <$fh>;
my %gone;
my ($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;
warn "BUG: multiple articles linked to $oid\n",
join(',',sort keys %gone), "\n";
}
- $self->{unindexed}->{$_}++ foreach keys %gone;
+ foreach my $num (keys %gone) {
+ $self->{unindexed}->{$_}++;
+ $self->{mm}->num_delete($num);
+ }
$self->unindex_oid_remote($oid, $mid);
}
}
return unless defined $latest;
$self->idx_init; # acquire lock
my $mm_tmp = $self->{mm}->tmp_clone;
- my $ranges = $opts->{reindex} ? [] : $self->last_commits($epoch_max);
+ my $reindex = $opts->{reindex};
+ my $ranges = $reindex ? [] : $self->last_commits($epoch_max);
- my ($min, $max) = $mm_tmp->minmax;
+ my $high = $self->{mm}->num_highwater();
my $regen = $self->index_prepare($opts, $epoch_max, $ranges);
- $$regen += $max if $max;
- my $D = {};
- my @cmd = qw(log --raw -r --pretty=tformat:%h
+ $$regen += $high if $high;
+ my $D = {}; # "$mid\0$cid" => $oid
+ my @cmd = qw(log --raw -r --pretty=tformat:%H
--no-notes --no-color --no-abbrev --no-renames);
# work backwards through history
my $git_dir = git_dir_n($self, $i);
die "already reindexing!\n" if delete $self->{reindex_pipe};
-d $git_dir or next; # missing parts are fine
+ fill_alternates($self, $i);
my $git = PublicInbox::Git->new($git_dir);
my $unindex = delete $self->{"unindex-range.$i"};
$self->unindex($opts, $git, $unindex) if $unindex;
my $fh = $self->{reindex_pipe} = $git->popen(@cmd, $range);
my $cmt;
while (<$fh>) {
+ chomp;
+ $self->{current_info} = "$i.git $_";
if (/\A$x40$/o && !defined($cmt)) {
- chomp($cmt = $_);
+ $cmt = $_;
} elsif (/\A:\d{6} 100644 $x40 ($x40) [AM]\tm$/o) {
$self->reindex_oid($mm_tmp, $D, $git, $1,
- $regen);
+ $regen, $reindex);
} elsif (/\A:\d{6} 100644 $x40 ($x40) [AM]\td$/o) {
$self->mark_deleted($D, $git, $1);
}
delete $self->{reindex_pipe};
$self->update_last_commit($git, $i, $cmt) if defined $cmt;
}
- my @d = sort keys %$D;
- if (@d) {
- warn "BUG: ", scalar(@d)," unseen deleted messages marked\n";
- foreach (@d) {
- my ($mid, undef) = split(/\0/, $_, 2);
- warn "<$mid>\n";
- }
+
+ # unindex is required for leftovers if "deletes" affect messages
+ # in a previous fetch+index window:
+ if (scalar keys %$D) {
+ my $git = $self->{-inbox}->git;
+ $self->unindex_oid($git, $_) for values %$D;
+ $git->cleanup;
}
$self->done;
}