support altid mechanism for v2

There's enough gmane links out there in wild that it makes sense
to maintain support for these mappings.

diff --git a/MANIFEST b/MANIFEST
index b17f1bea04e0af0c3634f6683f0ba2d15a79f7ab..82cc67d348aad9684651e5122e5e6780f9d207ff 100644 (file)
--- a/MANIFEST
+++ b/MANIFEST
@@ -135,6 +135,7 @@ scripts/ssoma-replay
 scripts/xhdr-num2mid
 t/address.t
 t/altid.t
+t/altid_v2.t
 t/cgi.t
 t/check-www-inbox.perl
 t/common.perl

diff --git a/lib/PublicInbox/AltId.pm b/lib/PublicInbox/AltId.pm
index d1b2dc24229cc2f2ab5761f982d47193527a5040..f8aa4cb8cf71a244d369fc671bf3639d635acdb4 100644 (file)
--- a/lib/PublicInbox/AltId.pm
+++ b/lib/PublicInbox/AltId.pm
@@ -22,17 +22,31 @@ sub new {
        } split(/[&;]/, $query);
        my $f = $params{file} or die "file: required for $type spec $spec\n";
        unless (index($f, '/') == 0) {
-               $f = "$inbox->{mainrepo}/public-inbox/$f";
+               if (($inbox->{version} || 1) == 1) {
+                       $f = "$inbox->{mainrepo}/public-inbox/$f";
+               } else {
+                       $f = "$inbox->{mainrepo}/$f";
+               }
        }
        bless {
-               mm_alt => PublicInbox::Msgmap->new_file($f, $writable),
+               filename => $f,
+               writable => $writable,
                xprefix => 'X'.uc($prefix),
        }, $class;
 }
 
+sub mm_alt {
+       my ($self) = @_;
+       $self->{mm_alt} ||= eval {
+               my $f = $self->{filename};
+               my $writable = $self->{filename};
+               PublicInbox::Msgmap->new_file($f, $writable);
+       };
+}
+
 sub mid2alt {
        my ($self, $mid) = @_;
-       $self->{mm_alt}->num_for($mid);
+       $self->mm_alt->num_for($mid);
 }
 
 1;

diff --git a/lib/PublicInbox/Filter/RubyLang.pm b/lib/PublicInbox/Filter/RubyLang.pm
index 63e8d4226e8ac2b8a2a2ac86041e14c53fe1542b..cb69e38a68aa01891287edd8b53448e53cd36259 100644 (file)
--- a/lib/PublicInbox/Filter/RubyLang.pm
+++ b/lib/PublicInbox/Filter/RubyLang.pm
@@ -6,6 +6,7 @@ package PublicInbox::Filter::RubyLang;
 use base qw(PublicInbox::Filter::Base);
 use strict;
 use warnings;
+use PublicInbox::MID qw(mids);
 
 my $l1 = qr/Unsubscribe:\s
        <mailto:ruby-\w+-request\@ruby-lang\.org\?subject=unsubscribe>/x;
@@ -44,16 +45,23 @@ sub scrub {
        my $altid = $self->{-altid};
        if ($altid) {
                my $hdr = $mime->header_obj;
-               my $mid = $hdr->header_raw('Message-ID');
-               unless (defined $mid) {
-                       return $self->REJECT('Message-Id missing');
+               my $mids = mids($hdr);
+               return $self->REJECT('Message-ID missing') unless (@$mids);
+               my @v = $hdr->header_raw('X-Mail-Count');
+               my $n;
+               foreach (@v) {
+                       /\A\s*(\d+)\s*\z/ or next;
+                       $n = $1;
+                       last;
                }
-               my $n = $hdr->header_raw('X-Mail-Count');
-               if (!defined($n) || $n !~ /\A\s*\d+\s*\z/) {
+               unless (defined $n) {
                        return $self->REJECT('X-Mail-Count not numeric');
                }
-               $mid = PublicInbox::MID::mid_clean($mid);
-               $altid->{mm_alt}->mid_set($n, $mid);
+               foreach my $mid (@$mids) {
+                       my $r = $altid->mm_alt->mid_set($n, $mid);
+                       next if $r == 0;
+                       last;
+               }
        }
        $self->ACCEPT($mime);
 }

diff --git a/t/altid_v2.t b/t/altid_v2.t
new file mode 100644 (file)
index 0000000..87f1452
--- /dev/null
+++ b/t/altid_v2.t
@@ -0,0 +1,55 @@
+# Copyright (C) 2016-2018 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+use strict;
+use warnings;
+use Test::More;
+use File::Temp qw/tempdir/;
+foreach my $mod (qw(DBD::SQLite Search::Xapian)) {
+       eval "require $mod";
+       plan skip_all => "$mod missing for altid_v2.t" if $@;
+}
+
+use_ok 'PublicInbox::V2Writable';
+use_ok 'PublicInbox::Inbox';
+my $tmpdir = tempdir('pi-altidv2-XXXXXX', TMPDIR => 1, CLEANUP => 1);
+my $mainrepo = "$tmpdir/inbox";
+my $full = "$tmpdir/inbox/another-nntp.sqlite3";
+my $altid = [ 'serial:gmane:file=another-nntp.sqlite3' ];
+
+{
+       ok(mkdir($mainrepo), 'created repo for msgmap');
+       my $mm = PublicInbox::Msgmap->new_file($full, 1);
+       is($mm->mid_set(1234, 'a@example.com'), 1, 'mid_set once OK');
+       ok(0 == $mm->mid_set(1234, 'a@example.com'), 'mid_set not idempotent');
+       ok(0 == $mm->mid_set(1, 'a@example.com'), 'mid_set fails with dup MID');
+}
+
+my $ibx = {
+       mainrepo => $mainrepo,
+       name => 'test-v2writable',
+       version => 2,
+       -primary_address => 'test@example.com',
+       altid => $altid,
+};
+$ibx = PublicInbox::Inbox->new($ibx);
+my $v2w = PublicInbox::V2Writable->new($ibx, 1);
+$v2w->add(Email::MIME->create(
+               header => [
+                       From => 'a@example.com',
+                       To => 'b@example.com',
+                       'Content-Type' => 'text/plain',
+                       Subject => 'boo!',
+                       'Message-ID' => '<a@example.com>',
+               ],
+               body => "hello world gmane:666\n",
+       ));
+$v2w->done;
+
+my $msgs = $ibx->search->reopen->query("gmane:1234");
+is_deeply([map { $_->mid } @$msgs], ['a@example.com'], 'got one match');
+$msgs = $ibx->search->query("gmane:666");
+is_deeply([], $msgs, 'body did NOT match');
+
+done_testing();
+
+1;