use v5.10.1;
use parent qw(PublicInbox::Lock);
use PublicInbox::SearchIdxShard;
+use PublicInbox::IdxStack;
use PublicInbox::Eml;
use PublicInbox::Git;
use PublicInbox::Import;
use PublicInbox::Msgmap;
use PublicInbox::Spawn qw(spawn popen_rd);
use PublicInbox::SearchIdx;
-use PublicInbox::MultiMidQueue;
use IO::Handle; # ->autoflush
use File::Temp qw(tempfile);
+my $OID = qr/[a-f0-9]{40,}/;
# an estimate of the post-packed size to the raw uncompressed size
my $PACKING_FACTOR = 0.4;
my $xpfx = "$dir/xap" . PublicInbox::Search::SCHEMA_VERSION;
my $self = {
- -inbox => $v2ibx,
+ ibx => $v2ibx,
im => undef, # PublicInbox::Import
parallel => 1,
transact_bytes => 0,
lock_path => "$dir/inbox.lock",
# limit each git repo (epoch) to 1GB or so
rotate_bytes => int((1024 * 1024 * 1024) / $PACKING_FACTOR),
- last_commit => [], # git repo -> commit
+ last_commit => [], # git epoch -> commit
};
$self->{shards} = count_shards($self) || nproc_shards($creat);
$self->{index_max_size} = $v2ibx->{index_max_size};
# mimics Import::add and wraps it for v2
sub add {
my ($self, $eml, $check_cb) = @_;
- $self->{-inbox}->with_umask(\&_add, $self, $eml, $check_cb);
+ $self->{ibx}->with_umask(\&_add, $self, $eml, $check_cb);
}
# indexes a message, returns true if checkpointing is needed
# spam check:
if ($check_cb) {
- $mime = $check_cb->($mime, $self->{-inbox}) or return;
+ $mime = $check_cb->($mime, $self->{ibx}) or return;
}
# All pipes (> $^F) known to Perl 5.6+ have FD_CLOEXEC set,
# 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};
+ my $altid = $self->{ibx}->{altid};
if ($altid && grep(/:file=msgmap\.sqlite3\z/, @$altid)) {
my $num = $self->{mm}->num_for($mid);
# Now that all subprocesses are up, we can open the FDs
# for SQLite:
my $mm = $self->{mm} = PublicInbox::Msgmap->new_file(
- "$self->{-inbox}->{inboxdir}/msgmap.sqlite3", 1);
+ "$self->{ibx}->{inboxdir}/msgmap.sqlite3", 1);
$mm->{dbh}->begin_work;
}
sub idx_init {
my ($self, $opt) = @_;
return if $self->{idx_shards};
- my $ibx = $self->{-inbox};
+ my $ibx = $self->{ibx};
# do not leak read-only FDs to child processes, we only have these
# FDs for duplicate detection so they should not be
sub _replace_oids ($$$) {
my ($self, $mime, $replace_map) = @_;
$self->done;
- my $pfx = "$self->{-inbox}->{inboxdir}/git";
+ my $pfx = "$self->{ibx}->{inboxdir}/git";
my $rewrites = []; # epoch => commit
my $max = $self->{epoch_max};
# (retval[2]) is not part of the stable API shared with Import->remove
sub remove {
my ($self, $eml, $cmt_msg) = @_;
- my $r = $self->{-inbox}->with_umask(\&rewrite_internal,
+ my $r = $self->{ibx}->with_umask(\&rewrite_internal,
$self, $eml, $cmt_msg);
defined($r) && defined($r->[0]) ? @$r: undef;
}
sub _replace ($$;$$) {
my ($self, $old_eml, $new_eml, $sref) = @_;
my $arg = [ $self, $old_eml, undef, $new_eml, $sref ];
- my $rewritten = $self->{-inbox}->with_umask(\&rewrite_internal,
+ my $rewritten = $self->{ibx}->with_umask(\&rewrite_internal,
$self, $old_eml, undef, $new_eml, $sref) or return;
my $rewrites = $rewritten->{rewrites};
sub git_hash_raw ($$) {
my ($self, $raw) = @_;
# grab the expected OID we have to reindex:
- open my $tmp_fh, '+>', undef or die "failed to open tmp: $!";
- $tmp_fh->autoflush(1);
- print $tmp_fh $$raw or die "print \$tmp_fh: $!";
- sysseek($tmp_fh, 0, 0) or die "seek failed: $!";
-
- my $git_dir = $self->{-inbox}->git->{git_dir};
+ pipe(my($in, $w)) or die "pipe: $!";
+ my $git_dir = $self->{ibx}->git->{git_dir};
my $cmd = ['git', "--git-dir=$git_dir", qw(hash-object --stdin)];
- my $r = popen_rd($cmd, undef, { 0 => $tmp_fh });
+ my $r = popen_rd($cmd, undef, { 0 => $in });
+ print $w $$raw or die "print \$w: $!";
+ close $w or die "close \$w: $!";
local $/ = "\n";
chomp(my $oid = <$r>);
close $r or die "git hash-object failed: $?";
- $oid =~ /\A[a-f0-9]{40}\z/ or die "OID not expected: $oid";
+ $oid =~ /\A$OID\z/ or die "OID not expected: $oid";
$oid;
}
}
# make sure we really got the OID:
- my ($blob, $type, $bytes) = $self->{-inbox}->git->check($expect_oid);
+ my ($blob, $type, $bytes) = $self->{ibx}->git->check($expect_oid);
$blob eq $expect_oid or die "BUG: $expect_oid not found after replace";
# don't leak FDs to Xapian:
- $self->{-inbox}->git->cleanup;
+ $self->{ibx}->git->cleanup;
# reindex modified messages:
for my $smsg (@$need_reindex) {
num => $smsg->{num},
mid => $smsg->{mid},
}, 'PublicInbox::Smsg';
- my $v2w = { autime => $smsg->{ds}, cotime => $smsg->{ts} };
- $new_smsg->populate($new_mime, $v2w);
+ my $sync = { autime => $smsg->{ds}, cotime => $smsg->{ts} };
+ $new_smsg->populate($new_mime, $sync);
do_idx($self, \$raw, $new_mime, $new_smsg);
}
$rewritten->{rewrites};
my $nbytes = $self->{total_bytes};
$self->{total_bytes} = 0;
$self->lock_release(!!$nbytes) if $shards;
- $self->{-inbox}->git->cleanup;
+ $self->{ibx}->git->cleanup;
}
sub fill_alternates ($$) {
my ($self, $epoch) = @_;
- my $pfx = "$self->{-inbox}->{inboxdir}/git";
- my $all = "$self->{-inbox}->{inboxdir}/all.git";
-
- unless (-d $all) {
- PublicInbox::Import::init_bare($all);
- }
+ my $pfx = "$self->{ibx}->{inboxdir}/git";
+ my $all = "$self->{ibx}->{inboxdir}/all.git";
+ PublicInbox::Import::init_bare($all) unless -d $all;
my $info_dir = "$all/objects/info";
my $alt = "$info_dir/alternates";
my (%alt, $new);
$mode = (stat($fh))[2] & 07777;
# we assign a sort score to every alternate and favor
- # the newest (highest numbered) one when we
+ # the newest (highest numbered) one because loose objects
+ # require scanning epochs and only the latest epoch is
+ # expected to see loose objects
my $score;
my $other = 0; # in case admin adds non-epoch repos
%alt = map {;
sub git_init {
my ($self, $epoch) = @_;
- my $git_dir = "$self->{-inbox}->{inboxdir}/git/$epoch.git";
+ my $git_dir = "$self->{ibx}->{inboxdir}/git/$epoch.git";
PublicInbox::Import::init_bare($git_dir);
my @cmd = (qw/git config/, "--file=$git_dir/config",
'include.path', '../../all.git/config');
sub git_dir_latest {
my ($self, $max) = @_;
$$max = -1;
- my $pfx = "$self->{-inbox}->{inboxdir}/git";
+ my $pfx = "$self->{ibx}->{inboxdir}/git";
return unless -d $pfx;
my $latest;
opendir my $dh, $pfx or die "opendir $pfx: $!\n";
sub import_init {
my ($self, $git, $packed_bytes, $tmp) = @_;
- my $im = PublicInbox::Import->new($git, undef, undef, $self->{-inbox});
+ my $im = PublicInbox::Import->new($git, undef, undef, $self->{ibx});
$im->{bytes_added} = int($packed_bytes / $PACKING_FACTOR);
$im->{lock_path} = undef;
$im->{path_type} = 'v2';
return $msg if $msg;
}
# older message, should be in alternates
- my $ibx = $self->{-inbox};
- $ibx->msg_by_smsg($smsg);
+ $self->{ibx}->msg_by_smsg($smsg);
}
sub content_exists ($$$) {
sub atfork_child {
my ($self) = @_;
- my $fh = delete $self->{reindex_pipe};
- close $fh if $fh;
if (my $shards = $self->{idx_shards}) {
$_->atfork_child foreach @$shards;
}
$self->{bnote}->[1];
}
-sub mark_deleted ($$$$) {
- my ($self, $sync, $git, $oid) = @_;
- return if PublicInbox::SearchIdx::too_big($self, $git, $oid);
- my $msgref = $git->cat_file($oid);
- my $mime = PublicInbox::Eml->new($$msgref);
- my $mids = mids($mime->header_obj);
- my $chash = content_hash($mime);
- foreach my $mid (@$mids) {
- $sync->{D}->{"$mid\0$chash"} = $oid;
- }
-}
-
-sub reindex_checkpoint ($$$) {
- my ($self, $sync, $git) = @_;
+sub reindex_checkpoint ($$) {
+ my ($self, $sync) = @_;
- $git->cleanup;
$sync->{mm_tmp}->atfork_prepare;
$self->done; # release lock
if (my $pr = $sync->{-opt}->{-progress}) {
- my ($bn) = (split('/', $git->{git_dir}))[-1];
- $pr->("$bn ".sprintf($sync->{-regen_fmt}, $sync->{nr}));
+ $pr->(sprintf($sync->{-regen_fmt}, $sync->{nr}));
}
# allow -watch or -mda to write...
$sync->{mm_tmp}->atfork_parent;
}
-# only for a few odd messages with multiple Message-IDs
-sub reindex_oid_m ($$$$;$) {
- my ($self, $sync, $git, $oid, $regen_num) = @_;
- $self->{current_info} = "multi_mid $oid";
- my ($num, $mid0, $len);
- my $msgref = $git->cat_file($oid, \$len);
- my $mime = PublicInbox::Eml->new($$msgref);
- my $mids = mids($mime->header_obj);
- my $chash = content_hash($mime);
- die "BUG: reindex_oid_m called for <=1 mids" if scalar(@$mids) <= 1;
-
- for my $mid (reverse @$mids) {
- delete($sync->{D}->{"$mid\0$chash"}) and
- die "BUG: reindex_oid should handle <$mid> delete";
- }
- my $over = $self->{over};
- for my $mid (reverse @$mids) {
- ($num, $mid0) = $over->num_mid0_for_oid($oid, $mid);
- next unless defined $num;
- if (defined($regen_num) && $regen_num != $num) {
- die "BUG: regen(#$regen_num) != over(#$num)";
- }
- }
- unless (defined($num)) {
- for my $mid (reverse @$mids) {
- # is this a number we got before?
- my $n = $sync->{mm_tmp}->num_for($mid);
- next unless defined $n;
- next if defined($regen_num) && $regen_num != $n;
- ($num, $mid0) = ($n, $mid);
- last;
- }
- }
- if (defined($num)) {
- $sync->{mm_tmp}->num_delete($num);
- } elsif (defined $regen_num) {
- $num = $regen_num;
- for my $mid (reverse @$mids) {
- $self->{mm}->mid_set($num, $mid) == 1 or next;
- $mid0 = $mid;
- last;
- }
- unless (defined $mid0) {
- warn "E: cannot regen #$num\n";
- return;
- }
- } else { # fixup bugs in old mirrors on reindex
- for my $mid (reverse @$mids) {
- $num = $self->{mm}->mid_insert($mid);
- next unless defined $num;
- $mid0 = $mid;
- last;
- }
- if (defined $mid0) {
- if ($sync->{reindex}) {
- warn "reindex added #$num <$mid0>\n";
- }
- } else {
- warn "E: cannot find article #\n";
- return;
- }
- }
- $sync->{nr}++;
- my $smsg = bless {
- raw_bytes => $len,
- num => $num,
- blob => $oid,
- mid => $mid0,
- }, 'PublicInbox::Smsg';
- $smsg->populate($mime, $self);
- if (do_idx($self, $msgref, $mime, $smsg)) {
- reindex_checkpoint($self, $sync, $git);
- }
-}
-
-sub check_unindexed ($$$) {
- my ($self, $num, $mid0) = @_;
- my $unindexed = $self->{unindexed} // {};
- my $n = delete($unindexed->{$mid0});
- defined $n or return;
- if ($n != $num) {
- die "BUG: unindexed $n != $num <$mid0>\n";
- } else {
- $self->{mm}->mid_set($num, $mid0);
- }
-}
-
-sub multi_mid_q_push ($$$) {
+sub reindex_oid ($$$) {
my ($self, $sync, $oid) = @_;
- my $multi_mid = $sync->{multi_mid} //= PublicInbox::MultiMidQueue->new;
- if ($sync->{reindex}) { # no regen on reindex
- $multi_mid->push_oid($oid, $self);
- } else {
- my $num = $sync->{regen}--;
- die "BUG: ran out of article numbers" if $num <= 0;
- $multi_mid->set_oid($num, $oid, $self);
- }
-}
-
-sub reindex_oid ($$$$) {
- my ($self, $sync, $git, $oid) = @_;
- return if PublicInbox::SearchIdx::too_big($self, $git, $oid);
+ return if PublicInbox::SearchIdx::too_big($self, $oid);
my ($num, $mid0, $len);
- my $msgref = $git->cat_file($oid, \$len);
+ my $msgref = $self->{ibx}->git->cat_file($oid, \$len);
return if $len == 0; # purged
my $mime = PublicInbox::Eml->new($$msgref);
my $mids = mids($mime->header_obj);
if (scalar(@$mids) == 0) {
warn "E: $oid has no Message-ID, skipping\n";
return;
- } elsif (scalar(@$mids) == 1) {
- my $mid = $mids->[0];
-
- # was the file previously marked as deleted?, skip if so
- if (delete($sync->{D}->{"$mid\0$chash"})) {
- if (!$sync->{reindex}) {
- $num = $sync->{regen}--;
- $self->{mm}->num_highwater($num);
- }
- return;
- }
+ }
- # is this a number we got before?
- $num = $sync->{mm_tmp}->num_for($mid);
+ # {unindexed} is unlikely
+ if ((my $unindexed = $self->{unindexed}) && scalar(@$mids) == 1) {
+ $num = delete($unindexed->{$mids->[0]});
if (defined $num) {
- $mid0 = $mid;
- check_unindexed($self, $num, $mid0);
- } else {
- $num = $sync->{regen}--;
- die "BUG: ran out of article numbers" if $num <= 0;
- if ($self->{mm}->mid_set($num, $mid) != 1) {
- warn "E: unable to assign $num => <$mid>\n";
- return;
- }
- $mid0 = $mid;
+ $mid0 = $mids->[0];
+ $self->{mm}->mid_set($num, $mid0);
+ delete($self->{unindexed}) if !keys(%$unindexed);
}
- } else { # multiple MIDs are a weird case:
- my $del = 0;
- for (@$mids) {
- $del += delete($sync->{D}->{"$_\0$chash"}) // 0;
+ }
+ if (!defined($num)) { # reuse if reindexing (or duplicates)
+ my $over = $self->{over};
+ for my $mid (@$mids) {
+ ($num, $mid0) = $over->num_mid0_for_oid($oid, $mid);
+ last if defined $num;
}
- if ($del) {
- unindex_oid_remote($self, $oid, $_) for @$mids;
- # do not delete from {mm_tmp}, since another
- # single-MID message may use it.
- } else { # handle them at the end:
- multi_mid_q_push($self, $sync, $oid);
+ }
+ $mid0 //= do { # is this a number we got before?
+ $num = $sync->{mm_tmp}->num_for($mids->[0]);
+ defined($num) ? $mids->[0] : undef;
+ };
+ if (!defined($num)) {
+ for (my $i = $#$mids; $i >= 1; $i--) {
+ $num = $sync->{mm_tmp}->num_for($mids->[$i]);
+ if (defined($num)) {
+ $mid0 = $mids->[$i];
+ last;
+ }
}
+ }
+ if (defined($num)) {
+ $sync->{mm_tmp}->num_delete($num);
+ } else { # never seen
+ $num = $self->{mm}->mid_insert($mids->[0]);
+ if (defined($num)) {
+ $mid0 = $mids->[0];
+ } else { # rare, try the rest of them, backwards
+ for (my $i = $#$mids; $i >= 1; $i--) {
+ $num = $self->{mm}->mid_insert($mids->[$i]);
+ if (defined($num)) {
+ $mid0 = $mids->[$i];
+ last;
+ }
+ }
+ }
+ }
+ if (!defined($num)) {
+ warn "E: $oid <", join('> <', @$mids), "> is a duplicate\n";
return;
}
- $sync->{mm_tmp}->mid_delete($mid0) or
- die "failed to delete <$mid0> for article #$num\n";
$sync->{nr}++;
my $smsg = bless {
raw_bytes => $len,
blob => $oid,
mid => $mid0,
}, 'PublicInbox::Smsg';
- $smsg->populate($mime, $self);
+ $smsg->populate($mime, $sync);
if (do_idx($self, $msgref, $mime, $smsg)) {
- reindex_checkpoint($self, $sync, $git);
+ reindex_checkpoint($self, $sync);
}
}
last_epoch_commit($self, $i, $cmt);
}
-sub git_dir_n ($$) { "$_[0]->{-inbox}->{inboxdir}/git/$_[1].git" }
+sub git_dir_n ($$) { "$_[0]->{ibx}->{inboxdir}/git/$_[1].git" }
sub last_commits ($$) {
my ($self, $epoch_max) = @_;
$range;
}
+sub prepare_range_stack {
+ my ($git, $sync, $range) = @_;
+ # Don't bump num_highwater on --reindex by using {D}.
+ # We intentionally do NOT use {D} in the non-reindex case because
+ # we want NNTP article number gaps from unindexed messages to
+ # show up in mirrors, too.
+ my $D = $sync->{D} //= $sync->{reindex} ? {} : undef; # OID_BIN => NR
+
+ my $fh = $git->popen(qw(log --raw -r --pretty=tformat:%at-%ct-%H
+ --no-notes --no-color --no-renames --no-abbrev),
+ $range);
+ my ($at, $ct, $stk);
+ while (<$fh>) {
+ if (/\A([0-9]+)-([0-9]+)-($OID)$/o) {
+ ($at, $ct) = ($1 + 0, $2 + 0);
+ $stk //= PublicInbox::IdxStack->new($3);
+ } elsif (/\A:\d{6} 100644 $OID ($OID) [AM]\td$/o) {
+ my $oid = $1;
+ if ($D) { # reindex case
+ $D->{pack('H*', $oid)}++;
+ } else { # non-reindex case:
+ $stk->push_rec('d', $at, $ct, $oid);
+ }
+ } elsif (/\A:\d{6} 100644 $OID ($OID) [AM]\tm$/o) {
+ my $oid = $1;
+ if ($D) {
+ my $oid_bin = pack('H*', $oid);
+ my $nr = --$D->{$oid_bin};
+ delete($D->{$oid_bin}) if $nr <= 0;
+
+ # nr < 0 (-1) means it never existed
+ $stk->push_rec('m', $at, $ct, $oid) if $nr < 0;
+ } else {
+ $stk->push_rec('m', $at, $ct, $oid);
+ }
+ }
+ }
+ close $fh or die "git log failed: \$?=$?";
+ $stk ? $stk->read_prepare : undef;
+}
+
sub sync_prepare ($$$) {
my ($self, $sync, $epoch_max) = @_;
my $pr = $sync->{-opt}->{-progress};
my $regen_max = 0;
- my $head = $self->{-inbox}->{ref_head} || 'refs/heads/master';
+ my $head = $self->{ibx}->{ref_head} || 'refs/heads/master';
# reindex stops at the current heads and we later rerun index_sync
# without {reindex}
my $reindex_heads = last_commits($self, $epoch_max) if $sync->{reindex};
for (my $i = $epoch_max; $i >= 0; $i--) {
- die 'BUG: already indexing!' if $self->{reindex_pipe};
my $git_dir = git_dir_n($self, $i);
-d $git_dir or next; # missing epochs are fine
my $git = PublicInbox::Git->new($git_dir);
next if $?; # new repo
my $range = log_range($self, $sync, $git, $i, $tip) or next;
- $sync->{ranges}->[$i] = $range;
-
# can't use 'rev-list --count' if we use --diff-filter
$pr->("$i.git counting $range ... ") if $pr;
- my $n = 0;
- my $fh = $git->popen(qw(log --pretty=tformat:%H
- --no-notes --no-color --no-renames
- --diff-filter=AM), $range, '--', 'm');
- ++$n while <$fh>;
- close $fh or die "git log failed: \$?=$?";
- $pr->("$n\n") if $pr;
- $regen_max += $n;
+ my $stk = prepare_range_stack($git, $sync, $range);
+ my $nr = $stk ? $stk->num_records : 0;
+ $pr->("$nr\n") if $pr;
+ $sync->{stacks}->[$i] = $stk if $stk;
+ $regen_max += $nr;
}
+ # XXX this should not happen unless somebody bypasses checks in
+ # our code and blindly injects "d" file history into git repos
+ if (my @leftovers = keys %{delete($sync->{D}) // {}}) {
+ warn('W: unindexing '.scalar(@leftovers)." leftovers\n");
+ for my $oid (@leftovers) {
+ $oid = unpack('H*', $oid);
+ $self->{current_info} = "leftover $oid";
+ unindex_oid($self, $oid);
+ }
+ }
return 0 if (!$regen_max && !keys(%{$self->{unindex_range}}));
# reindex should NOT see new commits anymore, if we do,
}
}
-sub unindex_oid ($$$;$) {
- my ($self, $git, $oid, $unindexed) = @_;
+sub unindex_oid ($$;$) {
+ my ($self, $oid, $unindexed) = @_;
my $mm = $self->{mm};
- my $msgref = $git->cat_file($oid);
+ my $msgref = $self->{ibx}->git->cat_file($oid);
my $mime = PublicInbox::Eml->new($msgref);
my $mids = mids($mime->header_obj);
$mime = $msgref = undef;
my ($id, $prev);
while (my $smsg = $over->next_by_mid($mid, \$id, \$prev)) {
$gone{$smsg->{num}} = 1 if $oid eq $smsg->{blob};
- 1; # continue
}
- my $n = scalar keys %gone;
- next unless $n;
+ my $n = scalar(keys(%gone)) or next;
if ($n > 1) {
warn "BUG: multiple articles linked to $oid\n",
join(',',sort keys %gone), "\n";
}
}
-my $x40 = qr/[a-f0-9]{40}/;
+# this is rare, it only happens when we get discontiguous history in
+# a mirror because the source used -purge or -edit
sub unindex ($$$$) {
my ($self, $sync, $git, $unindex_range) = @_;
my $unindexed = $self->{unindexed} ||= {}; # $mid0 => $num
# order does not matter, here:
my @cmd = qw(log --raw -r
--no-notes --no-color --no-abbrev --no-renames);
- my $fh = $self->{reindex_pipe} = $git->popen(@cmd, $unindex_range);
+ my $fh = $git->popen(@cmd, $unindex_range);
while (<$fh>) {
- /\A:\d{6} 100644 $x40 ($x40) [AM]\tm$/o or next;
- unindex_oid($self, $git, $1, $unindexed);
+ /\A:\d{6} 100644 $OID ($OID) [AM]\tm$/o or next;
+ unindex_oid($self, $1, $unindexed);
}
- delete $self->{reindex_pipe};
close $fh or die "git log failed: \$?=$?";
return unless $sync->{-opt}->{prune};
my ($self, $sync, $i) = @_;
my $git_dir = git_dir_n($self, $i);
- die 'BUG: already reindexing!' if $self->{reindex_pipe};
-d $git_dir or return; # missing epochs are fine
- fill_alternates($self, $i);
my $git = PublicInbox::Git->new($git_dir);
- if (my $unindex_range = delete $sync->{unindex_range}->{$i}) {
+ if (my $unindex_range = delete $sync->{unindex_range}->{$i}) { # rare
unindex($self, $sync, $git, $unindex_range);
}
- defined(my $range = $sync->{ranges}->[$i]) or return;
- if (my $pr = $sync->{-opt}->{-progress}) {
- $pr->("$i.git indexing $range\n");
- }
-
- my @cmd = qw(log --raw -r --pretty=tformat:%H.%at.%ct
- --no-notes --no-color --no-abbrev --no-renames);
- my $fh = $self->{reindex_pipe} = $git->popen(@cmd, $range);
- my $cmt;
- while (<$fh>) {
- chomp;
- $self->{current_info} = "$i.git $_";
- if (/\A($x40)\.([0-9]+)\.([0-9]+)$/o) {
- $cmt //= $1;
- $self->{autime} = $2;
- $self->{cotime} = $3;
- } elsif (/\A:\d{6} 100644 $x40 ($x40) [AM]\tm$/o) {
- reindex_oid($self, $sync, $git, $1);
- } elsif (/\A:\d{6} 100644 $x40 ($x40) [AM]\td$/o) {
- mark_deleted($self, $sync, $git, $1);
+ defined(my $stk = $sync->{stacks}->[$i]) or return;
+ $sync->{stacks}->[$i] = undef;
+ while (my ($f, $at, $ct, $oid) = $stk->pop_rec) {
+ $self->{current_info} = "$i.git $oid";
+ if ($f eq 'm') {
+ $sync->{autime} = $at;
+ $sync->{cotime} = $ct;
+ reindex_oid($self, $sync, $oid);
+ } elsif ($f eq 'd') {
+ unindex_oid($self, $oid);
}
}
- close $fh or die "git log failed: \$?=$?";
- delete @$self{qw(reindex_pipe autime cotime)};
- update_last_commit($self, $git, $i, $cmt) if defined $cmt;
+ delete @$sync{qw(autime cotime)};
+ update_last_commit($self, $git, $i, $stk->{latest_cmt});
}
# public, called by public-inbox-index
my $latest = git_dir_latest($self, \$epoch_max);
return unless defined $latest;
$self->idx_init($opt); # acquire lock
+ fill_alternates($self, $epoch_max);
+ $self->{over}->rethread_prepare($opt);
my $sync = {
- D => {}, # "$mid\0$chash" => $oid
unindex_range => {}, # EPOCH => oid_old..oid_new
reindex => $opt->{reindex},
-opt => $opt
};
$sync->{ranges} = sync_ranges($self, $sync, $epoch_max);
- $sync->{regen} = sync_prepare($self, $sync, $epoch_max);
-
- if ($sync->{regen}) {
+ if (sync_prepare($self, $sync, $epoch_max)) {
# tmp_clone seems to fail if inside a transaction, so
# we rollback here (because we opened {mm} for reading)
# Note: we do NOT rely on DBI transactions for atomicity;
$sync->{mm_tmp} = $self->{mm}->tmp_clone;
}
- # work backwards through history
- for (my $i = $epoch_max; $i >= 0; $i--) {
- index_epoch($self, $sync, $i);
- }
-
- # unindex is required for leftovers if "deletes" affect messages
- # in a previous fetch+index window:
- my $git;
- if (my @leftovers = values %{delete $sync->{D}}) {
- $git = $self->{-inbox}->git;
- for my $oid (@leftovers) {
- $self->{current_info} = "leftover $oid";
- unindex_oid($self, $git, $oid);
- }
- }
- if (my $multi_mid = delete $sync->{multi_mid}) {
- $git //= $self->{-inbox}->git;
- my $min = $multi_mid->{min};
- my $max = $multi_mid->{max};
- if ($sync->{reindex}) {
- # we may need to create new Message-IDs if mirrors
- # were initially indexed with old versions
- for (my $i = $max; $i >= $min; $i--) {
- my $oid;
- $oid = $multi_mid->get_oid($i, $self) or next;
- next unless defined $oid;
- reindex_oid_m($self, $sync, $git, $oid);
- }
- } else { # regen on initial index
- for my $num ($min..$max) {
- my $oid;
- $oid = $multi_mid->get_oid($num, $self) or next;
- reindex_oid_m($self, $sync, $git, $oid, $num);
- }
- }
- }
- $git->cleanup if $git;
+ # work forwards through history
+ index_epoch($self, $sync, $_) for (0..$epoch_max);
$self->done;
if (my $nr = $sync->{nr}) {
my $pr = $sync->{-opt}->{-progress};
$pr->('all.git '.sprintf($sync->{-regen_fmt}, $nr)) if $pr;
}
+ $self->{over}->rethread_done($opt);
# reindex does not pick up new changes, so we rerun w/o it:
if ($opt->{reindex}) {
my %again = %$opt;
$sync = undef;
- delete @again{qw(reindex -skip_lock)};
+ delete @again{qw(rethread reindex -skip_lock)};
index_sync($self, \%again);
}
}