X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=t%2Fv2reindex.t;h=2a6fc555793d686119892c1a0082d1f07e39fbdd;hb=0b1de991a099b5e8b9a9e3e85b5eaaacc9362dbb;hp=5bc307f1cac11ed34c3c3ec6e4c8258544c85652;hpb=ca62a98e2d8f56767bedc09d1cc29ce49d427d64;p=public-inbox.git diff --git a/t/v2reindex.t b/t/v2reindex.t index 5bc307f1..2a6fc555 100644 --- a/t/v2reindex.t +++ b/t/v2reindex.t @@ -7,6 +7,8 @@ use PublicInbox::MIME; use PublicInbox::ContentId qw(content_digest); use File::Temp qw/tempdir/; use File::Path qw(remove_tree); +require './t/common.perl'; +require_git(2.6); foreach my $mod (qw(DBD::SQLite Search::Xapian)) { eval "require $mod"; @@ -19,8 +21,15 @@ my $ibx_config = { name => 'test-v2writable', version => 2, -primary_address => 'test@example.com', + indexlevel => 'full', }; -my $ibx = PublicInbox::Inbox->new($ibx_config); +my $agpl = eval { + open my $fh, '<', 'COPYING' or die "can't open COPYING: $!"; + local $/; + <$fh>; +}; +$agpl or die "AGPL or die :P\n"; +my $phrase = q("defending all users' freedom"); my $mime = PublicInbox::MIME->create( header => [ From => 'a@example.com', @@ -28,42 +37,88 @@ my $mime = PublicInbox::MIME->create( Subject => 'this is a subject', Date => 'Fri, 02 Oct 1993 00:00:00 +0000', ], - body => "hello world\n", + body => $agpl, ); -local $ENV{NPROC} = 2; -my $im = PublicInbox::V2Writable->new($ibx, 1); -foreach my $i (1..10) { - $mime->header_set('Message-Id', "<$i\@example.com>"); - ok($im->add($mime), "message $i added"); - if ($i == 4) { +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, {nproc => 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}); } -} -if ('test remove later') { - $mime->header_set('Message-Id', "<5\@example.com>"); - $im->remove($mime); + $im->done; + $minmax = [ $ibx->mm->minmax ]; + ok(defined $minmax->[0] && defined $minmax->[1], 'minmax defined'); + is_deeply($minmax, [ 1, 10 ], 'minmax as expected'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); + + my ($min, $max) = @$minmax; + $msgmap = $ibx->mm->msg_range(\$min, $max); + is_deeply($msgmap, [ + [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'); } -$im->done; -my $minmax = [ $ibx->mm->minmax ]; -ok(defined $minmax->[0] && defined $minmax->[1], 'minmax defined'); -is_deeply($minmax, [ 1, 10 ], 'minmax as expected'); +{ + my %config = %$ibx_config; + my $ibx = PublicInbox::Inbox->new(\%config); + my $im = PublicInbox::V2Writable->new($ibx, 1); + eval { $im->index_sync({reindex => 1}) }; + is($@, '', 'no error from reindexing'); + $im->done; -eval { $im->index_sync({reindex => 1}) }; -is($@, '', 'no error from reindexing'); -$im->done; + delete $ibx->{mm}; + is_deeply([ $ibx->mm->minmax ], $minmax, 'minmax unchanged'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); + + my ($min, $max) = $ibx->mm->minmax; + is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged'); +} my $xap = "$mainrepo/xap".PublicInbox::Search::SCHEMA_VERSION(); remove_tree($xap); ok(!-d $xap, 'Xapian directories removed'); -eval { $im->index_sync({reindex => 1}) }; -is($@, '', 'no error from reindexing'); -$im->done; -ok(-d $xap, 'Xapian directories recreated'); +{ + my %config = %$ibx_config; + my $ibx = PublicInbox::Inbox->new(\%config); + my $im = PublicInbox::V2Writable->new($ibx, 1); + eval { $im->index_sync({reindex => 1}) }; + is($@, '', 'no error from reindexing'); + $im->done; + ok(-d $xap, 'Xapian directories recreated'); + + delete $ibx->{mm}; + is_deeply([ $ibx->mm->minmax ], $minmax, 'minmax unchanged'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); -delete $ibx->{mm}; -is_deeply($minmax, [ $ibx->mm->minmax ], 'minmax unchanged'); + my ($min, $max) = $ibx->mm->minmax; + is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged'); +} ok(unlink "$mainrepo/msgmap.sqlite3", 'remove msgmap'); remove_tree($xap); @@ -71,28 +126,309 @@ 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 $im = PublicInbox::V2Writable->new($ibx, 1); eval { $im->index_sync({reindex => 1}) }; is($@, '', 'no error from reindexing without msgmap'); is(scalar(@warn), 0, 'no warnings from reindexing'); $im->done; ok(-d $xap, 'Xapian directories recreated'); delete $ibx->{mm}; - is_deeply($minmax, [ $ibx->mm->minmax ], 'minmax unchanged'); + is_deeply([ $ibx->mm->minmax ], $minmax, 'minmax unchanged'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); + + my ($min, $max) = $ibx->mm->minmax; + is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged'); } +my %sizes; ok(unlink "$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); + my $im = PublicInbox::V2Writable->new($ibx, 1); eval { $im->index_sync({reindex => 1}) }; is($@, '', 'no error from reindexing without msgmap'); is_deeply(\@warn, [], 'no warnings'); $im->done; ok(-d $xap, 'Xapian directories recreated'); delete $ibx->{mm}; - is_deeply($minmax, [ $ibx->mm->minmax ], 'minmax unchanged'); + is_deeply([ $ibx->mm->minmax ], $minmax, 'minmax unchanged'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); + my $mset = $ibx->search->query($phrase, {mset=>1}); + isnt($mset->size, 0, "phrase search succeeds on indexlevel=full"); + for (<"$xap/*/*">) { $sizes{$ibx->{indexlevel}} += -s _ if -f $_ } + + my ($min, $max) = $ibx->mm->minmax; + is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged'); +} + +ok(unlink "$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; + $config{indexlevel} = 'medium'; + my $ibx = PublicInbox::Inbox->new(\%config); + my $im = PublicInbox::V2Writable->new($ibx); + eval { $im->index_sync({reindex => 1}) }; + is($@, '', 'no error from reindexing without msgmap'); + is_deeply(\@warn, [], 'no warnings'); + $im->done; + ok(-d $xap, 'Xapian directories recreated'); + delete $ibx->{mm}; + is_deeply([ $ibx->mm->minmax ], $minmax, 'minmax unchanged'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); + + if (0) { + # not sure why, but Xapian seems to fallback to terms and + # phrase searches still work + delete $ibx->{search}; + my $mset = $ibx->search->query($phrase, {mset=>1}); + is($mset->size, 0, 'phrase search does not work on medium'); + } + my $words = $phrase; + $words =~ tr/"'//d; + my $mset = $ibx->search->query($words, {mset=>1}); + isnt($mset->size, 0, "normal search works on indexlevel=medium"); + for (<"$xap/*/*">) { $sizes{$ibx->{indexlevel}} += -s _ if -f $_ } + + ok($sizes{full} > $sizes{medium}, 'medium is smaller than full'); + + + my ($min, $max) = $ibx->mm->minmax; + is_deeply($ibx->mm->msg_range(\$min, $max), $msgmap, 'msgmap unchanged'); +} + +ok(unlink "$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; + $config{indexlevel} = 'basic'; + my $ibx = PublicInbox::Inbox->new(\%config); + my $im = PublicInbox::V2Writable->new($ibx); + eval { $im->index_sync({reindex => 1}) }; + is($@, '', 'no error from reindexing without msgmap'); + is_deeply(\@warn, [], 'no warnings'); + $im->done; + ok(-d $xap, 'Xapian directories recreated'); + delete $ibx->{mm}; + is_deeply([ $ibx->mm->minmax ], $minmax, 'minmax unchanged'); + is($ibx->mm->num_highwater, 10, 'num_highwater as expected'); + + isnt($ibx->search, 'no search for basic'); + + for (<"$xap/*/*">) { $sizes{$ibx->{indexlevel}} += -s _ if -f $_ } + ok($sizes{medium} > $sizes{basic}, 'basic is smaller than medium'); + + my ($min, $max) = $ibx->mm->minmax; + is_deeply($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($max, 4, 'max as expected'); + is($ibx->mm->num_highwater, 4, 'num_highwater 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($max, 3, 'max as expected'); + is($ibx->mm->num_highwater, 4, 'num_highwater 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($max, 10, 'max as expected'); + is($ibx->mm->num_highwater, 10, 'num_highwater 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($max, 10, 'max as expected'); + is($ibx->mm->num_highwater, 10, 'num_highwater 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($max, 3, 'max as expected'); + is($ibx->mm->num_highwater, 4, 'num_highwater 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($max, 10, 'max as expected'); + is($ibx->mm->num_highwater, 10, 'num_highwater 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($max, 10, 'max as expected'); + is($ibx->mm->num_highwater, 10, 'num_highwater 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();