searchidx: regenerate and avoid article number gaps on full index

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

diff --git a/lib/PublicInbox/V2Writable.pm b/lib/PublicInbox/V2Writable.pm
index 1cc4b005f86525352bcd1cb81f7ee19139ec9ee5..2cc87305668256709f773415f750e5f52653503c 100644 (file)
--- a/lib/PublicInbox/V2Writable.pm
+++ b/lib/PublicInbox/V2Writable.pm
@@ -15,15 +15,19 @@ use PublicInbox::ContentId qw(content_id content_digest);
 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 $n = int($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 ($) {
@@ -73,7 +77,7 @@ sub new {
                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;
 }
 
@@ -171,19 +175,19 @@ sub num_for_harder {
 
        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);
                }
@@ -256,12 +260,32 @@ sub purge_oids {
        $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 ($$) {
+       my ($cids, $existing) = @_;
+       my $cid = content_id($existing);
+       foreach (@$cids) {
+               return 1 if $_ eq $cid
+       }
+       0
+}
+
 sub remove_internal {
        my ($self, $mime, $cmt_msg, $purge) = @_;
        $self->idx_init;
        my $im = $self->importer unless $purge;
        my $over = $self->{over};
-       my $cid = content_id($mime);
+       my $cids = content_ids($mime);
        my $parts = $self->{idx_parts};
        my $mm = $self->{mm};
        my $removed;
@@ -284,7 +308,7 @@ sub remove_internal {
                        }
                        my $orig = $$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 ];
                        }
@@ -538,7 +562,6 @@ sub import_init {
 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: $!";
@@ -569,8 +592,7 @@ sub get_blob ($$) {
 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);
@@ -579,16 +601,16 @@ sub lookup_content {
                        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 {
@@ -716,16 +738,7 @@ sub last_commits {
        $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) = @_;