t/v[12]reindex.t: Test incremental indexing works

Capture interesting commits of the test repository in mark variables.

Use those marks to build interesting scenarios where index_sync proceeds
as if those marks are the heads of the repositor.  Use this capability to
test what happens when adds and deletes are mixed within a repository.

Be sad because things don't yet work as they should.

Signed-off-by: "Eric W. Biederman" <ebiederm@xmission.com>

diff --git a/t/v1reindex.t b/t/v1reindex.t
index 877e70390910d20fb05ba446eaddffdff78fc7d3..8e78aa761333942678dc83abcd4af0a5363948bc 100644 (file)
--- a/t/v1reindex.t
+++ b/t/v1reindex.t
@@ -33,6 +33,7 @@ my $mime = PublicInbox::MIME->create(
 );
 my $minmax;
 my $msgmap;
+my ($mark1, $mark2, $mark3, $mark4);
 {
        my %config = %$ibx_config;
        my $ibx = PublicInbox::Inbox->new(\%config);
@@ -41,13 +42,17 @@ my $msgmap;
                $mime->header_set('Message-Id', "<$i\@example.com>");
                ok($im->add($mime), "message $i added");
                if ($i == 4) {
+                       $mark1 = $im->get_mark($im->{tip});
                        $im->remove($mime);
+                       $mark2 = $im->get_mark($im->{tip});
                }
        }
 
        if ('test remove later') {
+               $mark3 = $im->get_mark($im->{tip});
                $mime->header_set('Message-Id', "<5\@example.com>");
                $im->remove($mime);
+               $mark4 = $im->get_mark($im->{tip});
        }
 
        $im->done;
@@ -222,4 +227,193 @@ ok(!-d $xap, 'Xapian directories removed again');
        is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged');
 }
 
+# An incremental indexing test
+ok(unlink "$mainrepo/public-inbox/msgmap.sqlite3", 'remove msgmap');
+remove_tree($xap);
+ok(!-d $xap, 'Xapian directories removed again');
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark1 4 simple additions in the same index_sync
+       eval { $rw->index_sync({ref => $mark1}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 4, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [4, '4@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark2 A delete separated form and add in the same index_sync
+       eval { $rw->index_sync({ref => $mark2}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 3, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark3 adds following the delete at mark2
+       eval { $rw->index_sync({ref => $mark3}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [5, '5@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark4 A delete of an older message
+       eval { $rw->index_sync({ref => $mark4}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+
+
+# Another incremental indexing test
+ok(unlink "$mainrepo/public-inbox/msgmap.sqlite3", 'remove msgmap');
+remove_tree($xap);
+ok(!-d $xap, 'Xapian directories removed again');
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark2 an add and it's delete in the same index_sync
+       eval { $rw->index_sync({ref => $mark2}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 3, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark3 adds following the delete at mark2
+       eval { $rw->index_sync({ref => $mark3}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [5, '5@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       my $im = PublicInbox::Import->new($ibx->git, undef, undef, $ibx);
+       my $rw = PublicInbox::SearchIdx->new($ibx, 1);
+       # mark4 A delete of an older message
+       eval { $rw->index_sync({ref => $mark4}) };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+
+
 done_testing();

diff --git a/t/v2reindex.t b/t/v2reindex.t
index 31e61d5b2578947b25e8fe41f56053c96c9818e5..ce13a07c39c08472bfb475dcdd9d801e406c088d 100644 (file)
--- a/t/v2reindex.t
+++ b/t/v2reindex.t
@@ -33,21 +33,27 @@ my $mime = PublicInbox::MIME->create(
 local $ENV{NPROC} = 2;
 my $minmax;
 my $msgmap;
+my ($mark1, $mark2, $mark3, $mark4);
 {
        my %config = %$ibx_config;
        my $ibx = PublicInbox::Inbox->new(\%config);
        my $im = PublicInbox::V2Writable->new($ibx, 1);
+       my $im0 = $im->importer();
        foreach my $i (1..10) {
                $mime->header_set('Message-Id', "<$i\@example.com>");
                ok($im->add($mime), "message $i added");
                if ($i == 4) {
+                       $mark1 = $im0->get_mark($im0->{tip});
                        $im->remove($mime);
+                       $mark2 = $im0->get_mark($im0->{tip});
                }
        }
 
        if ('test remove later') {
+               $mark3 = $im0->get_mark($im0->{tip});
                $mime->header_set('Message-Id', "<5\@example.com>");
                $im->remove($mime);
+               $mark4 = $im0->get_mark($im0->{tip});
        }
 
        $im->done;
@@ -211,4 +217,193 @@ ok(!-d $xap, 'Xapian directories removed again');
        is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged');
 }
 
+
+# An incremental indexing test
+ok(unlink "$mainrepo/msgmap.sqlite3", 'remove msgmap');
+remove_tree($xap);
+ok(!-d $xap, 'Xapian directories removed again');
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark1 4 simple additions in the same index_sync
+       $ibx->{ref_head} = $mark1;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 4, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [4, '4@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark2 A delete separated from an add in the same index_sync
+       $ibx->{ref_head} = $mark2;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 3, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark3 adds following the delete at mark2
+       $ibx->{ref_head} = $mark3;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [5, '5@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark4 A delete of an older message
+       $ibx->{ref_head} = $mark4;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+
+
+# Another incremental indexing test
+ok(unlink "$mainrepo/msgmap.sqlite3", 'remove msgmap');
+remove_tree($xap);
+ok(!-d $xap, 'Xapian directories removed again');
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark2 an add and it's delete in the same index_sync
+       $ibx->{ref_head} = $mark2;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 3, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark3 adds following the delete at mark2
+       $ibx->{ref_head} = $mark3;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [5, '5@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+{
+       my @warn;
+       local $SIG{__WARN__} = sub { push @warn, @_ };
+       my %config = %$ibx_config;
+       my $ibx = PublicInbox::Inbox->new(\%config);
+       # mark4 A delete of an older message
+       $ibx->{ref_head} = $mark4;
+       my $im = PublicInbox::V2Writable->new($ibx);
+       eval { $im->index_sync() };
+       is($@, '', 'no error from reindexing without msgmap');
+       is_deeply(\@warn, [], 'no warnings');
+       $im->done;
+       my ($min, $max) = $ibx->mm->minmax;
+       is($min, 1, 'min as expected');
+       is($max, 10, 'max as expected');
+       is_deeply($ibx->mm->msg_range(\$min, $max),
+                 [
+                  [1, '1@example.com' ],
+                  [2, '2@example.com' ],
+                  [3, '3@example.com' ],
+                  [6, '6@example.com' ],
+                  [7, '7@example.com' ],
+                  [8, '8@example.com' ],
+                  [9, '9@example.com' ],
+                  [10, '10@example.com' ],
+                 ], 'msgmap as expected' );
+}
+
 done_testing();