X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=t%2Fsearch.t;h=54f2db906a04cd04d7cddda3bf175feed2e82c4f;hb=f826a7830bce67c0f8c653baf97f7769bb2c57d9;hp=02189ac4c9c79df3107049fc8d21fa9112ee9983;hpb=8ea217f4452e34776ba294b5090827c99656cada;p=public-inbox.git
diff --git a/t/search.t b/t/search.t
index 02189ac4..54f2db90 100644
--- a/t/search.t
+++ b/t/search.t
@@ -1,5 +1,5 @@
-# Copyright (C) 2015, all contributors
-# License: AGPLv3 or later (https://www.gnu.org/licenses/agpl-3.0.txt)
+# Copyright (C) 2015-2018 all contributors
+# License: AGPL-3.0+
use strict;
use warnings;
use Test::More;
@@ -7,8 +7,7 @@ eval { require PublicInbox::SearchIdx; };
plan skip_all => "Xapian missing for search" if $@;
use File::Temp qw/tempdir/;
use Email::MIME;
-use Data::Dumper;
-my $tmpdir = tempdir(CLEANUP => 1);
+my $tmpdir = tempdir('pi-search-XXXXXX', TMPDIR => 1, CLEANUP => 1);
my $git_dir = "$tmpdir/a.git";
my ($root_id, $last_id);
@@ -16,56 +15,45 @@ is(0, system(qw(git init -q --bare), $git_dir), "git init (main)");
eval { PublicInbox::Search->new($git_dir) };
ok($@, "exception raised on non-existent DB");
-{
- my $orig = "FOO " x 30;
- my $summ = PublicInbox::Search::subject_summary($orig);
-
- $summ = length($summ);
- $orig = length($orig);
- ok($summ < $orig && $summ > 0, "summary shortened ($orig => $summ)");
-
- $orig = "FOO" x 30;
- $summ = PublicInbox::Search::subject_summary($orig);
-
- $summ = length($summ);
- $orig = length($orig);
- ok($summ < $orig && $summ > 0,
- "summary shortened but not empty: $summ");
-}
-
my $rw = PublicInbox::SearchIdx->new($git_dir, 1);
+my $ibx = $rw->{-inbox};
+$ibx->with_umask(sub {
+ $rw->_xdb_acquire;
+ $rw->_xdb_release;
+});
+$rw = undef;
my $ro = PublicInbox::Search->new($git_dir);
my $rw_commit = sub {
- $rw = undef;
+ $rw->commit_txn_lazy if $rw;
$rw = PublicInbox::SearchIdx->new($git_dir, 1);
+ $rw->begin_txn_lazy;
};
{
# git repository perms
- is(PublicInbox::SearchIdx->_git_config_perm(undef),
- &PublicInbox::SearchIdx::PERM_GROUP,
+ is($ibx->_git_config_perm(), &PublicInbox::InboxWritable::PERM_GROUP,
"undefined permission is group");
- is(PublicInbox::SearchIdx::_umask_for(
- PublicInbox::SearchIdx->_git_config_perm('0644')),
+ is(PublicInbox::InboxWritable::_umask_for(
+ PublicInbox::InboxWritable->_git_config_perm('0644')),
0022, "644 => umask(0022)");
- is(PublicInbox::SearchIdx::_umask_for(
- PublicInbox::SearchIdx->_git_config_perm('0600')),
+ is(PublicInbox::InboxWritable::_umask_for(
+ PublicInbox::InboxWritable->_git_config_perm('0600')),
0077, "600 => umask(0077)");
- is(PublicInbox::SearchIdx::_umask_for(
- PublicInbox::SearchIdx->_git_config_perm('0640')),
+ is(PublicInbox::InboxWritable::_umask_for(
+ PublicInbox::InboxWritable->_git_config_perm('0640')),
0027, "640 => umask(0027)");
- is(PublicInbox::SearchIdx::_umask_for(
- PublicInbox::SearchIdx->_git_config_perm('group')),
+ is(PublicInbox::InboxWritable::_umask_for(
+ PublicInbox::InboxWritable->_git_config_perm('group')),
0007, 'group => umask(0007)');
- is(PublicInbox::SearchIdx::_umask_for(
- PublicInbox::SearchIdx->_git_config_perm('everybody')),
+ is(PublicInbox::InboxWritable::_umask_for(
+ PublicInbox::InboxWritable->_git_config_perm('everybody')),
0002, 'everybody => umask(0002)');
- is(PublicInbox::SearchIdx::_umask_for(
- PublicInbox::SearchIdx->_git_config_perm('umask')),
+ is(PublicInbox::InboxWritable::_umask_for(
+ PublicInbox::InboxWritable->_git_config_perm('umask')),
umask, 'umask => existing umask');
}
-{
+$ibx->with_umask(sub {
my $root = Email::MIME->create(
header_str => [
Date => 'Fri, 02 Oct 1993 00:00:00 +0000',
@@ -83,62 +71,54 @@ my $rw_commit = sub {
'Message-ID' => '',
From => 'John Smith ',
To => 'list@example.com',
+ Cc => 'foo@example.com',
],
body => "goodbye forever :<\n");
my $rv;
+ $rw_commit->();
$root_id = $rw->add_message($root);
is($root_id, int($root_id), "root_id is an integer: $root_id");
$last_id = $rw->add_message($last);
is($last_id, int($last_id), "last_id is an integer: $last_id");
-}
+});
sub filter_mids {
- my ($res) = @_;
- sort(map { $_->mid } @{$res->{msgs}});
+ my ($msgs) = @_;
+ sort(map { $_->mid } @$msgs);
}
{
$rw_commit->();
$ro->reopen;
- my $found = $ro->lookup_message('');
- ok($found, "message found");
- is($root_id, $found->{doc_id}, 'doc_id set correctly');
- $found->ensure_metadata;
- is($found->mid, 'root@s', 'mid set correctly');
- ok(int($found->thread_id) > 0, 'thread_id is an integer');
+ my $found = $ro->query('m:root@s');
+ is(scalar(@$found), 1, "message found");
+ is($found->[0]->mid, 'root@s', 'mid set correctly');
+ my ($res, @res);
my @exp = sort qw(root@s last@s);
- my $res = $ro->query("path:hello_world");
- my @res = filter_mids($res);
- is_deeply(\@res, \@exp, 'got expected results for path: match');
- foreach my $p (qw(hello hello_ hello_world2 hello_world_)) {
- $res = $ro->query("path:$p");
- is($res->{total}, 0, "path variant `$p' does not match");
- }
-
- $res = $ro->query('subject:(Hello world)');
+ $res = $ro->query('s:(Hello world)');
@res = filter_mids($res);
- is_deeply(\@res, \@exp, 'got expected results for subject:() match');
+ is_deeply(\@res, \@exp, 'got expected results for s:() match');
- $res = $ro->query('subject:"Hello world"');
+ $res = $ro->query('s:"Hello world"');
@res = filter_mids($res);
- is_deeply(\@res, \@exp, 'got expected results for subject:"" match');
+ is_deeply(\@res, \@exp, 'got expected results for s:"" match');
- $res = $ro->query('subject:"Hello world"', {limit => 1});
- is(scalar @{$res->{msgs}}, 1, "limit works");
- my $first = $res->{msgs}->[0];
+ $res = $ro->query('s:"Hello world"', {limit => 1});
+ is(scalar @$res, 1, "limit works");
+ my $first = $res->[0];
- $res = $ro->query('subject:"Hello world"', {offset => 1});
- is(scalar @{$res->{msgs}}, 1, "offset works");
- my $second = $res->{msgs}->[0];
+ $res = $ro->query('s:"Hello world"', {offset => 1});
+ is(scalar @$res, 1, "offset works");
+ my $second = $res->[0];
isnt($first, $second, "offset returned different result from limit");
}
# ghost vivication
-{
+$ibx->with_umask(sub {
$rw_commit->();
my $rmid = '';
my $reply_to_ghost = Email::MIME->create(
@@ -168,15 +148,21 @@ sub filter_mids {
my $ghost_id = $rw->add_message($was_ghost);
is($ghost_id, int($ghost_id), "ghost_id is an integer: $ghost_id");
- ok($ghost_id < $reply_id, "ghost vivified from earlier message");
-}
+ my $msgs = $rw->{over}->get_thread('ghost-message@s');
+ is(scalar(@$msgs), 2, 'got both messages in ghost thread');
+ foreach (qw(sid tid)) {
+ is($msgs->[0]->{$_}, $msgs->[1]->{$_}, "{$_} match");
+ }
+ isnt($msgs->[0]->{num}, $msgs->[1]->{num}, "num do not match");
+ ok($_->{num} > 0, 'positive art num') foreach @$msgs
+});
# search thread on ghost
{
$rw_commit->();
$ro->reopen;
- # Subject:
+ # subject
my $res = $ro->query('ghost');
my @exp = sort qw(ghost-message@s ghost-reply@s);
my @res = filter_mids($res);
@@ -184,15 +170,21 @@ sub filter_mids {
# body
$res = $ro->query('goodbye');
- is($res->{msgs}->[0]->mid, 'last@s', 'got goodbye message body');
+ is($res->[0]->mid, 'last@s', 'got goodbye message body');
+
+ # datestamp
+ $res = $ro->query('dt:20101002000001..20101002000001');
+ @res = filter_mids($res);
+ is_deeply(\@res, ['ghost-message@s'], 'exact Date: match works');
+ $res = $ro->query('dt:20101002000002..20101002000002');
+ is_deeply($res, [], 'exact Date: match down to the second');
}
# long message-id
-{
+$ibx->with_umask(sub {
$rw_commit->();
$ro->reopen;
my $long_mid = 'last' . ('x' x 60). '@s';
- my $long_midc = Digest::SHA::sha1_hex($long_mid);
my $long = Email::MIME->create(
header_str => [
@@ -231,14 +223,14 @@ sub filter_mids {
$rw_commit->();
$ro->reopen;
my $t = $ro->get_thread('root@s');
- is($t->{total}, 4, "got all 4 mesages in thread");
- my @exp = sort($long_reply_mid, 'root@s', 'last@s', $long_midc);
+ is(scalar(@$t), 4, "got all 4 mesages in thread");
+ my @exp = sort($long_reply_mid, 'root@s', 'last@s', $long_mid);
@res = filter_mids($t);
is_deeply(\@res, \@exp, "get_thread works");
-}
+});
# quote prioritization
-{
+$ibx->with_umask(sub {
$rw_commit->();
$rw->add_message(Email::MIME->create(
header_str => [
@@ -260,22 +252,23 @@ sub filter_mids {
],
body => "theatre\nfade\n"));
my $res = $rw->query("theatre");
- is($res->{total}, 2, "got both matches");
- is($res->{msgs}->[0]->mid, 'nquote@a', "non-quoted scores higher");
- is($res->{msgs}->[1]->mid, 'quote@a', "quoted result still returned");
+ is(scalar(@$res), 2, "got both matches");
+ is($res->[0]->mid, 'nquote@a', "non-quoted scores higher");
+ is($res->[1]->mid, 'quote@a', "quoted result still returned");
$res = $rw->query("illusions");
- is($res->{total}, 1, "got a match for quoted text");
- is($res->{msgs}->[0]->mid, 'quote@a',
+ is(scalar(@$res), 1, "got a match for quoted text");
+ is($res->[0]->mid, 'quote@a',
"quoted result returned if nothing else");
-}
+});
# circular references
-{
+$ibx->with_umask(sub {
+ my $s = 'foo://'. ('Circle' x 15).'/foo';
my $doc_id = $rw->add_message(Email::MIME->create(
+ header => [ Subject => $s ],
header_str => [
Date => 'Sat, 02 Oct 2010 00:00:01 +0000',
- Subject => 'Circle',
'Message-ID' => '',
'References' => '',
'In-Reply-To' => '',
@@ -284,11 +277,189 @@ sub filter_mids {
],
body => "LOOP!\n"));
ok($doc_id > 0, "doc_id defined with circular reference");
- my $smsg = $rw->lookup_message('circle@a');
- $smsg->ensure_metadata;
- is($smsg->references_sorted, '', "no references created");
+ my $smsg = $rw->query('m:circle@a', {limit=>1})->[0];
+ is($smsg->references, '', "no references created");
+ is($s, $smsg->subject, 'long subject not rewritten');
+});
+
+$ibx->with_umask(sub {
+ my $str = eval {
+ my $mbox = 't/utf8.mbox';
+ open(my $fh, '<', $mbox) or die "failed to open mbox: $mbox\n";
+ local $/;
+ <$fh>
+ };
+ $str =~ s/\AFrom [^\n]+\n//s;
+ my $mime = Email::MIME->new($str);
+ my $doc_id = $rw->add_message($mime);
+ ok($doc_id > 0, 'message indexed doc_id with UTF-8');
+ my $msg = $rw->query('m:testmessage@example.com', {limit => 1})->[0];
+ is($mime->header('Subject'), $msg->subject, 'UTF-8 subject preserved');
+});
+
+{
+ my $msgs = $ro->query('d:19931002..20101002');
+ ok(scalar(@$msgs) > 0, 'got results within range');
+ $msgs = $ro->query('d:20101003..');
+ is(scalar(@$msgs), 0, 'nothing after 20101003');
+ $msgs = $ro->query('d:..19931001');
+ is(scalar(@$msgs), 0, 'nothing before 19931001');
+}
+
+# names and addresses
+{
+ my $mset = $ro->query('t:list@example.com', {mset => 1});
+ is($mset->size, 6, 'searched To: successfully');
+ foreach my $m ($mset->items) {
+ my $smsg = $ro->lookup_article($m->get_docid);
+ like($smsg->to, qr/\blist\@example\.com\b/, 'to appears');
+ }
+
+ $mset = $ro->query('tc:list@example.com', {mset => 1});
+ is($mset->size, 6, 'searched To+Cc: successfully');
+ foreach my $m ($mset->items) {
+ my $smsg = $ro->lookup_article($m->get_docid);
+ my $tocc = join("\n", $smsg->to, $smsg->cc);
+ like($tocc, qr/\blist\@example\.com\b/, 'tocc appears');
+ }
+
+ foreach my $pfx ('tcf:', 'c:') {
+ my $mset = $ro->query($pfx . 'foo@example.com', { mset => 1 });
+ is($mset->items, 1, "searched $pfx successfully for Cc:");
+ foreach my $m ($mset->items) {
+ my $smsg = $ro->lookup_article($m->get_docid);
+ like($smsg->cc, qr/\bfoo\@example\.com\b/,
+ 'cc appears');
+ }
+ }
+
+ foreach my $pfx ('', 'tcf:', 'f:') {
+ my $res = $ro->query($pfx . 'Laggy');
+ is(scalar(@$res), 1,
+ "searched $pfx successfully for From:");
+ foreach my $smsg (@$res) {
+ like($smsg->from, qr/Laggy Sender/,
+ "From appears with $pfx");
+ }
+ }
}
+{
+ $rw_commit->();
+ $ro->reopen;
+ my $res = $ro->query('b:hello');
+ is(scalar(@$res), 0, 'no match on body search only');
+ $res = $ro->query('bs:smith');
+ is(scalar(@$res), 0,
+ 'no match on body+subject search for From');
+
+ $res = $ro->query('q:theatre');
+ is(scalar(@$res), 1, 'only one quoted body');
+ like($res->[0]->from, qr/\AQuoter/, 'got quoted body');
+
+ $res = $ro->query('nq:theatre');
+ is(scalar @$res, 1, 'only one non-quoted body');
+ like($res->[0]->from, qr/\ANon-Quoter/, 'got non-quoted body');
+
+ foreach my $pfx (qw(b: bs:)) {
+ $res = $ro->query($pfx . 'theatre');
+ is(scalar @$res, 2, "searched both bodies for $pfx");
+ like($res->[0]->from, qr/\ANon-Quoter/,
+ "non-quoter first for $pfx");
+ }
+}
+
+$ibx->with_umask(sub {
+ my $part1 = Email::MIME->create(
+ attributes => {
+ content_type => 'text/plain',
+ disposition => 'attachment',
+ charset => 'US-ASCII',
+ encoding => 'quoted-printable',
+ filename => 'attached_fart.txt',
+ },
+ body_str => 'inside the attachment',
+ );
+ my $part2 = Email::MIME->create(
+ attributes => {
+ content_type => 'text/plain',
+ disposition => 'attachment',
+ charset => 'US-ASCII',
+ encoding => 'quoted-printable',
+ filename => 'part_deux.txt',
+ },
+ body_str => 'inside another',
+ );
+ my $amsg = Email::MIME->create(
+ header_str => [
+ Subject => 'see attachment',
+ 'Message-ID' => '',
+ From => 'John Smith ',
+ To => 'list@example.com',
+ ],
+ parts => [ $part1, $part2 ],
+ );
+ ok($rw->add_message($amsg), 'added attachment');
+ $rw_commit->();
+ $ro->reopen;
+ my $n = $ro->query('n:attached_fart.txt');
+ is(scalar @$n, 1, 'got result for n:');
+ my $res = $ro->query('part_deux.txt');
+ is(scalar @$res, 1, 'got result without n:');
+ is($n->[0]->mid, $res->[0]->mid,
+ 'same result with and without');
+ my $txt = $ro->query('"inside another"');
+ is($txt->[0]->mid, $res->[0]->mid,
+ 'search inside text attachments works');
+
+ my $mid = $n->[0]->mid;
+ my ($id, $prev);
+ my $art = $ro->next_by_mid($mid, \$id, \$prev);
+ ok($art, 'article exists in OVER DB');
+ $rw->unindex_blob($amsg);
+ $rw->commit_txn_lazy;
+ is($ro->lookup_article($art->{num}), undef, 'gone from OVER DB');
+});
+
+foreach my $f ("$git_dir/public-inbox/msgmap.sqlite3",
+ glob("$git_dir/public-inbox/xapian*/"),
+ glob("$git_dir/public-inbox/xapian*/*")) {
+ my @st = stat($f);
+ my ($bn) = (split(m!/!, $f))[-1];
+ is($st[2] & 07777, -f _ ? 0660 : 0770,
+ "sharedRepository respected for $bn");
+}
+
+$ibx->with_umask(sub {
+ $rw_commit->();
+ my $digits = '10010260936330';
+ my $ua = 'Pine.LNX.4.10';
+ my $mid = "$ua.$digits.2460-100000\@penguin.transmeta.com";
+ is($ro->reopen->query("m:$digits", { mset => 1})->size, 0,
+ 'no results yet');
+ my $pine = Email::MIME->create(
+ header_str => [
+ Subject => 'blah',
+ 'Message-ID' => "<$mid>",
+ From => 'torvalds@transmeta',
+ To => 'list@example.com',
+ ],
+ body => ""
+ );
+ my $x = $rw->add_message($pine);
+ $rw->commit_txn_lazy;
+ is($ro->reopen->query("m:$digits", { mset => 1})->size, 1,
+ 'searching only digit yielded result');
+
+ my $wild = $digits;
+ for my $i (1..6) {
+ chop($wild);
+ is($ro->query("m:$wild*", { mset => 1})->size, 1,
+ "searching chopped($i) digit yielded result $wild ");
+ }
+ is($ro->query("m:Pine m:LNX m:10010260936330", {mset=>1})->size, 1);
+});
+
done_testing();
1;