lei: relax rules for "new" in Maildir

mbsync and offlineimap both use ":2," suffixes for filenames in
"new/", however my interpretation of the Maildir spec at
<https://cr.yp.to/proto/maildir.html> is that ":2," is only for
files in "cur/".  My interpretation also matches that of
doveecot, but we'll allow what mbsync and offlineimap do given
their popularity.

diff --git a/lib/PublicInbox/LeiMailSync.pm b/lib/PublicInbox/LeiMailSync.pm
index 803de48c176cc3a7d625838b0282a52a24492d13..3bada42dd035df425cc5964cabb7aaa058915ad0 100644 (file)
--- a/lib/PublicInbox/LeiMailSync.pm
+++ b/lib/PublicInbox/LeiMailSync.pm
@@ -233,9 +233,16 @@ WHERE b.oidbin = ?
        $b2n->execute(pack('H*', $oidhex));
        while (my ($d, $n) = $b2n->fetchrow_array) {
                substr($d, 0, length('maildir:')) = '';
-               my $f = "$d/" . ($n =~ /:2,[a-zA-Z]*\z/ ? "cur/$n" : "new/$n");
-               open my $fh, '<', $f or next;
-               if (-s $fh) {
+               # n.b. both mbsync and offlineimap use ":2," as a suffix
+               # in "new/", despite (from what I understand of reading
+               # <https://cr.yp.to/proto/maildir.html>), the ":2," only
+               # applies to files in "cur/".
+               my @try = $n =~ /:2,[a-zA-Z]+\z/ ? qw(cur new) : qw(new cur);
+               for my $x (@try) {
+                       my $f = "$d/$x/$n";
+                       open my $fh, '<', $f or next;
+                       # some (buggy) Maildir writers are non-atomic:
+                       next unless -s $fh;
                        local $/;
                        my $raw = <$fh>;
                        if ($vrfy && git_sha(1, \$raw)->hexdigest ne $oidhex) {

diff --git a/lib/PublicInbox/MdirReader.pm b/lib/PublicInbox/MdirReader.pm
index b49c8ceb86eec87d1f56da3e20b5fe3f858d34c2..7a0641fb07f287fe8a9ff17127c3041b5428b3fe 100644 (file)
--- a/lib/PublicInbox/MdirReader.pm
+++ b/lib/PublicInbox/MdirReader.pm
@@ -61,7 +61,10 @@ sub maildir_each_eml {
                while (defined(my $bn = readdir($dh))) {
                        next if substr($bn, 0, 1) eq '.';
                        my @f = split(/:/, $bn, -1);
-                       next if scalar(@f) != 1;
+
+                       # mbsync and offlineimap both use "2," in "new/"
+                       next if ($f[1] // '2,') ne '2,' || defined($f[2]);
+
                        next if defined($mod) && !shard_ok($bn, $mod, $shard);
                        my $f = $pfx.$bn;
                        my $eml = eml_from_path($f) or next;

diff --git a/t/lei-index.t b/t/lei-index.t
index 3382d42bfce96f5c7e6038cc81aeb3ea9e7c9625..b7dafb711f7eced8f31a0e118e3897490d666fcf 100644 (file)
--- a/t/lei-index.t
+++ b/t/lei-index.t
@@ -27,6 +27,11 @@ my $expect = do {
        local $/;
        <$fh>;
 };
+
+# mbsync and offlineimap both put ":2," in "new/" files:
+symlink(File::Spec->rel2abs('t/utf8.eml'), "$tmpdir/md/new/u:2,") or
+       xbail "symlink $!";
+
 test_lei({ tmpdir => $tmpdir }, sub {
        my $store_path = "$ENV{HOME}/.local/share/lei/store/";
 
@@ -40,13 +45,16 @@ test_lei({ tmpdir => $tmpdir }, sub {
        lei_ok(qw(q mid:qp@example.com -f text));
        like($lei_out, qr/^hi = bye/sm, 'lei2mail fallback');
 
+       lei_ok(qw(q mid:testmessage@example.com -f text));
+       lei_ok(qw(-C / blob --mail 9bf1002c49eb075df47247b74d69bcd555e23422));
+
        my $all_obj = ['git', "--git-dir=$store_path/ALL.git",
                        qw(cat-file --batch-check --batch-all-objects)];
        is_deeply([xqx($all_obj)], [], 'no git objects');
        lei_ok('import', 't/plack-qp.eml');
        ok(grep(/\A$blob blob /, my @objs = xqx($all_obj)),
                'imported blob');
-       lei_ok(qw(q z:0.. --dedupe=none));
+       lei_ok(qw(q m:qp@example.com --dedupe=none));
        my $res_b = json_utf8->decode($lei_out);
        is_deeply($res_b, $res_a, 'no extra DB entries');