#!perl -w
-# Copyright (C) 2020 all contributors <meta@public-inbox.org>
+# Copyright (C) 2020-2021 all contributors <meta@public-inbox.org>
# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
use strict;
use Test::More;
use PublicInbox::TestCommon;
+use PublicInbox::Config;
+use PublicInbox::Search;
+use PublicInbox::InboxWritable;
use Fcntl qw(:seek);
require_git(2.6);
-require_mods(qw(DBD::SQLite Search::Xapian));
+require_mods(qw(json DBD::SQLite Search::Xapian));
use_ok 'PublicInbox::ExtSearch';
use_ok 'PublicInbox::ExtSearchIdx';
+use_ok 'PublicInbox::OverIdx';
+my $sock = tcp_server();
+my $host_port = $sock->sockhost . ':' . $sock->sockport;
my ($home, $for_destroy) = tmpdir();
local $ENV{HOME} = $home;
mkdir "$home/.public-inbox" or BAIL_OUT $!;
-open my $fh, '>', "$home/.public-inbox/config" or BAIL_OUT $!;
+my $cfg_path = "$home/.public-inbox/config";
+open my $fh, '>', $cfg_path or BAIL_OUT $!;
print $fh <<EOF or BAIL_OUT $!;
[publicinboxMda]
spamcheck = none
close $fh or BAIL_OUT $!;
my $v2addr = 'v2test@example.com';
my $v1addr = 'v1test@example.com';
-ok(run_script([qw(-init -V2 v2test --newsgroup v2.example), "$home/v2test",
- 'http://example.com/v2test', $v2addr ]), 'v2test init');
+ok(run_script([qw(-init -Lbasic -V2 v2test --newsgroup v2.example),
+ "$home/v2test", 'http://example.com/v2test', $v2addr ]), 'v2test init');
my $env = { ORIGINAL_RECIPIENT => $v2addr };
-open($fh, '<', 't/utf8.eml') or BAIL_OUT("open t/utf8.eml: $!");
+my $eml = eml_load('t/utf8.eml');
+
+$eml->header_set('List-Id', '<v2.example.com>');
+open($fh, '+>', undef) or BAIL_OUT $!;
+$fh->autoflush(1);
+print $fh $eml->as_string or BAIL_OUT $!;
+seek($fh, 0, SEEK_SET) or BAIL_OUT $!;
+
run_script(['-mda', '--no-precheck'], $env, { 0 => $fh }) or BAIL_OUT '-mda';
-ok(run_script([qw(-init -V1 v1test), "$home/v1test",
+ok(run_script([qw(-init -V1 v1test --newsgroup v1.example), "$home/v1test",
'http://example.com/v1test', $v1addr ]), 'v1test init');
-$env = { ORIGINAL_RECIPIENT => $v1addr };
+
+$eml->header_set('List-Id', '<v1.example.com>');
+seek($fh, 0, SEEK_SET) or BAIL_OUT $!;
+truncate($fh, 0) or BAIL_OUT $!;
+print $fh $eml->as_string or BAIL_OUT $!;
seek($fh, 0, SEEK_SET) or BAIL_OUT $!;
+
+$env = { ORIGINAL_RECIPIENT => $v1addr };
run_script(['-mda', '--no-precheck'], $env, { 0 => $fh }) or BAIL_OUT '-mda';
-run_script(['-index', "$home/v1test"]) or BAIL_OUT "index $?";
-ok(run_script([qw(-eindex --all), "$home/eindex"]), 'eindex init');
+run_script([qw(-index -Lbasic), "$home/v1test"]) or BAIL_OUT "index $?";
-my $es = PublicInbox::ExtSearch->new("$home/eindex");
+ok(run_script([qw(-extindex --all), "$home/extindex"]), 'extindex init');
+{
+ my $es = PublicInbox::ExtSearch->new("$home/extindex");
+ ok($es->has_threadid, '->has_threadid');
+}
+
+{ # TODO: -extindex should write this to config
+ open $fh, '>>', $cfg_path or BAIL_OUT $!;
+ print $fh <<EOF or BAIL_OUT $!;
+; for ->ALL
+[extindex "all"]
+ topdir = $home/extindex
+EOF
+ close $fh or BAIL_OUT $!;
+
+ my $pi_cfg = PublicInbox::Config->new;
+ $pi_cfg->fill_all;
+ ok($pi_cfg->ALL, '->ALL');
+ my $ibx = $pi_cfg->{-by_newsgroup}->{'v2.example'};
+ my $ret = $pi_cfg->ALL->nntp_xref_for($ibx, $ibx->over->get_art(1));
+ is_deeply($ret, { 'v1.example' => 1, 'v2.example' => 1 },
+ '->nntp_xref_for');
+}
+
+SKIP: {
+ require_mods(qw(Net::NNTP), 1);
+ my ($out, $err) = ("$home/nntpd.out.log", "$home/nntpd.err.log");
+ my $cmd = [ '-nntpd', '-W0', "--stdout=$out", "--stderr=$err" ];
+ my $td = start_script($cmd, undef, { 3 => $sock });
+ my $n = Net::NNTP->new($host_port);
+ my @xp = $n->xpath('<testmessage@example.com>');
+ is_deeply(\@xp, [ qw(v1.example/1 v2.example/1) ]);
+ $n->group('v1.example');
+ my $res = $n->head(1);
+ @$res = grep(/^Xref: /, @$res);
+ like($res->[0], qr/ v1\.example:1 v2\.example:1/, 'nntp_xref works');
+}
+
+my $es = PublicInbox::ExtSearch->new("$home/extindex");
{
my $smsg = $es->over->get_art(1);
ok($smsg, 'got first article');
is($es->over->get_art(2), undef, 'only one added');
my $xref3 = $es->over->get_xref3(1);
like($xref3->[0], qr/\A\Qv2.example\E:1:/, 'order preserved 1');
- like($xref3->[1], qr!\A\Q$home/v1test\E:1:!, 'order preserved 2');
+ like($xref3->[1], qr/\A\Qv1.example\E:1:/, 'order preserved 2');
is(scalar(@$xref3), 2, 'only to entries');
}
-{
+if ('inbox edited') {
my ($in, $out, $err);
$in = $out = $err = '';
my $opt = { 0 => \$in, 1 => \$out, 2 => \$err };
my $env = { MAIL_EDITOR => "$^X -i -p -e 's/test message/BEST MSG/'" };
my $cmd = [ qw(-edit -Ft/utf8.eml), "$home/v2test" ];
ok(run_script($cmd, $env, $opt), '-edit');
- ok(run_script([qw(-eindex --all), "$home/eindex"], undef, $opt),
- 'eindex again');
+ ok(run_script([qw(-extindex --all), "$home/extindex"], undef, $opt),
+ 'extindex again');
like($err, qr/discontiguous range/, 'warned about discontiguous range');
my $msg1 = $es->over->get_art(1) or BAIL_OUT 'msg1 missing';
my $msg2 = $es->over->get_art(2) or BAIL_OUT 'msg2 missing';
is(scalar(@$x1), 1, 'original only has one xref3');
is(scalar(@$x2), 1, 'new message has one xref3');
isnt($x1->[0], $x2->[0], 'xref3 differs');
+
+ my $mset = $es->mset('b:"BEST MSG"');
+ is($mset->size, 1, 'new message found');
+ $mset = $es->mset('b:"test message"');
+ is($mset->size, 1, 'old message found');
+ delete @$es{qw(git over xdb)}; # fork preparation
+
+ my $pi_cfg = PublicInbox::Config->new;
+ $pi_cfg->fill_all;
+ is(scalar($pi_cfg->ALL->mset('s:Testing')->items), 2,
+ '2 results in ->ALL');
+ my $res = {};
+ my $nr = 0;
+ $pi_cfg->each_inbox(sub {
+ $nr++;
+ my ($ibx) = @_;
+ local $SIG{__WARN__} = sub {}; # FIXME support --reindex
+ my $mset = $ibx->isrch->mset('s:Testing');
+ $res->{$ibx->eidx_key} = $ibx->isrch->mset_to_smsg($ibx, $mset);
+ });
+ is($nr, 2, 'two inboxes');
+ my $exp = {};
+ for my $v (qw(v1 v2)) {
+ my $ibx = $pi_cfg->lookup_newsgroup("$v.example");
+ my $smsg = $ibx->over->get_art(1);
+ $smsg->psgi_cull;
+ $exp->{"$v.example"} = [ $smsg ];
+ }
+ is_deeply($res, $exp, 'isearch limited results');
+ $pi_cfg = $res = $exp = undef;
+
+ open my $rmfh, '+>', undef or BAIL_OUT $!;
+ $rmfh->autoflush(1);
+ print $rmfh $eml2->as_string or BAIL_OUT $!;
+ seek($rmfh, 0, SEEK_SET) or BAIL_OUT $!;
+ $opt->{0} = $rmfh;
+ ok(run_script([qw(-learn rm --all)], undef, $opt), '-learn rm');
+
+ ok(run_script([qw(-extindex --all), "$home/extindex"], undef, undef),
+ 'extindex after rm');
+ is($es->over->get_art(2), undef, 'doc #2 gone');
+ $mset = $es->mset('b:"BEST MSG"');
+ is($mset->size, 0, 'new message gone');
+}
+
+my $misc = $es->misc;
+my @it = $misc->mset('')->items;
+is(scalar(@it), 2, 'two inboxes');
+like($it[0]->get_document->get_data, qr/v2test/, 'docdata matched v2');
+like($it[1]->get_document->get_data, qr/v1test/, 'docdata matched v1');
+
+my $cfg = PublicInbox::Config->new;
+my $schema_version = PublicInbox::Search::SCHEMA_VERSION();
+my $f = "$home/extindex/ei$schema_version/over.sqlite3";
+my $oidx = PublicInbox::OverIdx->new($f);
+if ('inject w/o indexing') {
+ use PublicInbox::Import;
+ my $v1ibx = $cfg->lookup_name('v1test');
+ my $last_v1_commit = $v1ibx->mm->last_commit;
+ my $v2ibx = $cfg->lookup_name('v2test');
+ my $last_v2_commit = $v2ibx->mm->last_commit_xap($schema_version, 0);
+ my $git0 = PublicInbox::Git->new("$v2ibx->{inboxdir}/git/0.git");
+ chomp(my $cmt = $git0->qx(qw(rev-parse HEAD^0)));
+ is($last_v2_commit, $cmt, 'v2 index up-to-date');
+
+ my $v2im = PublicInbox::Import->new($git0, undef, undef, $v2ibx);
+ $v2im->{lock_path} = undef;
+ $v2im->{path_type} = 'v2';
+ $v2im->add(eml_load('t/mda-mime.eml'));
+ $v2im->done;
+ chomp(my $tip = $git0->qx(qw(rev-parse HEAD^0)));
+ isnt($tip, $cmt, '0.git v2 updated');
+
+ # inject a message w/o updating index
+ rename("$home/v1test/public-inbox", "$home/v1test/skip-index") or
+ BAIL_OUT $!;
+ open(my $eh, '<', 't/iso-2202-jp.eml') or BAIL_OUT $!;
+ run_script(['-mda', '--no-precheck'], $env, { 0 => $eh}) or
+ BAIL_OUT '-mda';
+ rename("$home/v1test/skip-index", "$home/v1test/public-inbox") or
+ BAIL_OUT $!;
+
+ my ($in, $out, $err);
+ $in = $out = $err = '';
+ my $opt = { 0 => \$in, 1 => \$out, 2 => \$err };
+ ok(run_script([qw(-extindex -v -v --all), "$home/extindex"],
+ undef, undef), 'extindex noop');
+ $es->{xdb}->reopen;
+ my $mset = $es->mset('mid:199707281508.AAA24167@hoyogw.example');
+ is($mset->size, 0, 'did not attempt to index unindexed v1 message');
+ $mset = $es->mset('mid:multipart-html-sucks@11');
+ is($mset->size, 0, 'did not attempt to index unindexed v2 message');
+ ok(run_script([qw(-index --all)]), 'indexed v1 and v2 inboxes');
+
+ isnt($v1ibx->mm->last_commit, $last_v1_commit, '-index v1 worked');
+ isnt($v2ibx->mm->last_commit_xap($schema_version, 0),
+ $last_v2_commit, '-index v2 worked');
+ ok(run_script([qw(-extindex --all), "$home/extindex"]),
+ 'extindex updates');
+
+ $es->{xdb}->reopen;
+ $mset = $es->mset('mid:199707281508.AAA24167@hoyogw.example');
+ is($mset->size, 1, 'got v1 message');
+ $mset = $es->mset('mid:multipart-html-sucks@11');
+ is($mset->size, 1, 'got v2 message');
+}
+
+if ('reindex catches missed messages') {
+ my $v2ibx = $cfg->lookup_name('v2test');
+ my $im = PublicInbox::InboxWritable->new($v2ibx)->importer(0);
+ my $cmt_a = $v2ibx->mm->last_commit_xap($schema_version, 0);
+ my $eml = eml_load('t/data/0001.patch');
+ $im->add($eml);
+ $im->done;
+ my $cmt_b = $v2ibx->mm->last_commit_xap($schema_version, 0);
+ isnt($cmt_a, $cmt_b, 'v2 0.git HEAD updated');
+ $oidx->dbh;
+ my $uv = $v2ibx->uidvalidity;
+ my $lc_key = "lc-v2:v2.example//$uv;0";
+ is($oidx->eidx_meta($lc_key, $cmt_b), $cmt_a,
+ 'update lc-v2 meta, old is as expected');
+ my $max = $oidx->max;
+ $oidx->dbh_close;
+ ok(run_script([qw(-extindex), "$home/extindex", $v2ibx->{inboxdir}]),
+ '-extindex noop');
+ is($oidx->max, $max, '->max unchanged');
+ is($oidx->eidx_meta($lc_key), $cmt_b, 'lc-v2 unchanged');
+ $oidx->dbh_close;
+ my $opt = { 2 => \(my $err = '') };
+ ok(run_script([qw(-extindex --reindex), "$home/extindex",
+ $v2ibx->{inboxdir}], undef, $opt),
+ '--reindex for unseen');
+ is($oidx->max, $max + 1, '->max bumped');
+ is($oidx->eidx_meta($lc_key), $cmt_b, 'lc-v2 stays unchanged');
+ my @err = split(/^/, $err);
+ is(scalar(@err), 1, 'only one warning') or diag "err=$err";
+ like($err[0], qr/I: reindex_unseen/, 'got reindex_unseen message');
+ my $new = $oidx->get_art($max + 1);
+ is($new->{subject}, $eml->header('Subject'), 'new message added');
+
+ $es->{xdb}->reopen;
+ my $mset = $es->mset("mid:$new->{mid}");
+ is($mset->size, 1, 'previously unseen, now indexed in Xapian');
+
+ ok($im->remove($eml), 'remove new message from v2 inbox');
+ $im->done;
+ my $cmt_c = $v2ibx->mm->last_commit_xap($schema_version, 0);
+ is($oidx->eidx_meta($lc_key, $cmt_c), $cmt_b,
+ 'bump lc-v2 meta again to skip v2 remove');
+ $err = '';
+ $oidx->dbh_close;
+ ok(run_script([qw(-extindex --reindex), "$home/extindex",
+ $v2ibx->{inboxdir}], undef, $opt),
+ '--reindex for stale');
+ @err = split(/^/, $err);
+ is(scalar(@err), 1, 'only one warning') or diag "err=$err";
+ like($err[0], qr/\(#$new->{num}\): stale/, 'got stale message warning');
+ is($oidx->get_art($new->{num}), undef,
+ 'stale message gone from over');
+ is_deeply($oidx->get_xref3($new->{num}), [],
+ 'stale message has no xref3');
+ $es->{xdb}->reopen;
+ $mset = $es->mset("mid:$new->{mid}");
+ is($mset->size, 0, 'stale mid gone Xapian');
+}
+
+if ('reindex catches content bifurcation') {
+ use PublicInbox::MID qw(mids);
+ my $v2ibx = $cfg->lookup_name('v2test');
+ my $im = PublicInbox::InboxWritable->new($v2ibx)->importer(0);
+ my $eml = eml_load('t/data/message_embed.eml');
+ my $cmt_a = $v2ibx->mm->last_commit_xap($schema_version, 0);
+ $im->add($eml);
+ $im->done;
+ my $cmt_b = $v2ibx->mm->last_commit_xap($schema_version, 0);
+ my $uv = $v2ibx->uidvalidity;
+ my $lc_key = "lc-v2:v2.example//$uv;0";
+ $oidx->dbh;
+ is($oidx->eidx_meta($lc_key, $cmt_b), $cmt_a,
+ 'update lc-v2 meta, old is as expected');
+ my $mid = mids($eml)->[0];
+ my $smsg = $v2ibx->over->next_by_mid($mid, \(my $id), \(my $prev));
+ my $oldmax = $oidx->max;
+ my $x3_orig = $oidx->get_xref3(3);
+ is(scalar(@$x3_orig), 1, '#3 has one xref');
+ $oidx->add_xref3(3, $smsg->{num}, $smsg->{blob}, 'v2.example');
+ my $x3 = $oidx->get_xref3(3);
+ is(scalar(@$x3), 2, 'injected xref3');
+ $oidx->commit_lazy;
+ my $opt = { 2 => \(my $err = '') };
+ ok(run_script([qw(-extindex --all), "$home/extindex"], undef, $opt),
+ 'extindex --all is noop');
+ is($err, '', 'no warnings in index');
+ $oidx->dbh;
+ is($oidx->max, $oldmax, 'oidx->max unchanged');
+ $oidx->dbh_close;
+ ok(run_script([qw(-extindex --reindex --all), "$home/extindex"],
+ undef, $opt), 'extindex --reindex');
+ $oidx->dbh;
+ ok($oidx->max > $oldmax, 'oidx->max bumped');
+ like($err, qr/split into 2 due to deduplication change/,
+ 'bifurcation noted');
+ my $added = $oidx->get_art($oidx->max);
+ is($added->{blob}, $smsg->{blob}, 'new blob indexed');
+ is_deeply(["v2.example:$smsg->{num}:$smsg->{blob}"],
+ $oidx->get_xref3($added->{num}),
+ 'xref3 corrected for bifurcated message');
+ is_deeply($oidx->get_xref3(3), $x3_orig, 'xref3 restored for #3');
+}
+
+if ('--reindex --rethread') {
+ my $before = $oidx->dbh->selectrow_array(<<'');
+SELECT MAX(tid) FROM over WHERE num > 0
+
+ my $opt = {};
+ ok(run_script([qw(-extindex --reindex --rethread --all),
+ "$home/extindex"], undef, $opt),
+ '--rethread');
+ my $after = $oidx->dbh->selectrow_array(<<'');
+SELECT MIN(tid) FROM over WHERE num > 0
+
+ # actual rethread logic is identical to v1/v2 and tested elsewhere
+ ok($after > $before, '--rethread updates MIN(tid)');
+}
+
+if ('remove v1test and test gc') {
+ xsys([qw(git config --unset publicinbox.v1test.inboxdir)],
+ { GIT_CONFIG => $cfg_path });
+ my $opt = { 2 => \(my $err = '') };
+ ok(run_script([qw(-extindex --gc), "$home/extindex"], undef, $opt),
+ 'extindex --gc');
+ like($err, qr/^I: remove #1 v1\.example /ms, 'removed v1 message');
+ is(scalar(grep(!/^I:/, split(/^/m, $err))), 0,
+ 'no non-informational messages');
+ $misc->{xdb}->reopen;
+ @it = $misc->mset('')->items;
+ is(scalar(@it), 1, 'only one inbox left');
}
done_testing;