lei: make --dedupe=content always account for Message-IDs

The content dedupe logic was originally designed for v2 public
inboxes as a fallback for when the importer sees identical
Message-IDs.  Thus it did not account for Message-ID(s) in
the message itself.

This change doesn't affect saved searches (the default when
writing to a pathname or IMAP).  It affects --no-save, and
outputs to stdout (even if stdout is redirected to a file).

Prior to this change, lei reused the v2 logic as-is without
accounting for Message-IDs anywhere with `--dedupe=content'
(the default).  This could cause messages to be skipped when
the content matches despite Message-IDs being different.

So with this change, `lei q --dedupe=content' will hash the
Message-ID(s) in the message to ensure messages with different
Message-IDs are NOT deduplicated.

Whether or not this change is a bug fix or introduces regression
is actually debatable.  In my mind, it is better to err on the
side of showing too many messages rather than too few, even if
the actual contents of the message are identical.  Making saved
searches deduplicate without accounting for Message-IDs would be
more difficult, too.

diff --git a/lib/PublicInbox/ContentHash.pm b/lib/PublicInbox/ContentHash.pm
index fc94257c7c243a378b18ae81f04be06b1c3bb0cb..95ca2929139904163192356baed8457133b20280 100644 (file)
--- a/lib/PublicInbox/ContentHash.pm
+++ b/lib/PublicInbox/ContentHash.pm
@@ -54,16 +54,23 @@ sub content_dig_i {
        $dig->add($s);
 }
 
-sub content_digest ($;$) {
-       my ($eml, $dig) = @_;
+sub content_digest ($;$$) {
+       my ($eml, $dig, $hash_mids) = @_;
        $dig //= Digest::SHA->new(256);
 
        # References: and In-Reply-To: get used interchangeably
        # in some "duplicates" in LKML.  We treat them the same
        # in SearchIdx, so treat them the same for this:
        # do NOT consider the Message-ID as part of the content_hash
-       # if we got here, we've already got Message-ID reuse
-       my %seen = map { $_ => 1 } @{mids($eml)};
+       # if we got here, we've already got Message-ID reuse for v2.
+       #
+       # However, `lei q --dedupe=content' does use $hash_mids since
+       # it doesn't have any other dedupe
+       my $mids = mids($eml);
+       if ($hash_mids) {
+               $dig->add("mid\0$_\0") for @$mids;
+       }
+       my %seen = map { $_ => 1 } @$mids;
        for (grep { !$seen{$_}++ } @{references($eml)}) {
                utf8::encode($_);
                $dig->add("ref\0$_\0");

diff --git a/lib/PublicInbox/LeiDedupe.pm b/lib/PublicInbox/LeiDedupe.pm
index 86cd8490d6307729d60fec55f30da81511a82b29..eda54d790084da11653164846b3df5e50d1a597d 100644 (file)
--- a/lib/PublicInbox/LeiDedupe.pm
+++ b/lib/PublicInbox/LeiDedupe.pm
@@ -2,7 +2,7 @@
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 package PublicInbox::LeiDedupe;
 use v5.12;
-use PublicInbox::ContentHash qw(content_hash git_sha);
+use PublicInbox::ContentHash qw(content_hash content_digest git_sha);
 use PublicInbox::SHA qw(sha256);
 
 # n.b. mutt sets most of these headers not sure about Bytes
@@ -69,7 +69,12 @@ sub dedupe_content ($) {
        my ($skv) = @_;
        (sub { # may be called in a child process
                my ($eml) = @_; # $oidhex = $_[1], ignored
-               $skv->set_maybe(content_hash($eml), '');
+
+               # we must account for Message-ID via hash_mids, since
+               # (unlike v2 dedupe) Message-ID is not accounted for elsewhere:
+               $skv->set_maybe(content_digest($eml, PublicInbox::SHA->new(256),
+                               1 # hash_mids
+                               )->digest, '');
        }, sub {
                my ($smsg) = @_;
                $skv->set_maybe(smsg_hash($smsg), '');

diff --git a/t/lei_dedupe.t b/t/lei_dedupe.t
index e1944d025b5fb56f26f845de1af971926039cf19..13fc1f3bdb6a78bc0154255c8e36c4ec6ca25fcc 100644 (file)
--- a/t/lei_dedupe.t
+++ b/t/lei_dedupe.t
@@ -1,5 +1,5 @@
 #!perl -w
-# Copyright (C) 2020-2021 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 use strict;
 use v5.10.1;
@@ -10,6 +10,8 @@ use PublicInbox::Smsg;
 require_mods(qw(DBD::SQLite));
 use_ok 'PublicInbox::LeiDedupe';
 my $eml = eml_load('t/plack-qp.eml');
+my $sameish = eml_load('t/plack-qp.eml');
+$sameish->header_set('Message-ID', '<cuepee@example.com>');
 my $mid = $eml->header_raw('Message-ID');
 my $different = eml_load('t/msg_iter-order.eml');
 $different->header_set('Message-ID', $mid);
@@ -47,6 +49,8 @@ for my $strat (undef, 'content') {
        ok(!$dd->is_dup($different), "different is_dup with $desc dedupe");
        ok(!$dd->is_smsg_dup($smsg), "is_smsg_dup pass w/ $desc dedupe");
        ok($dd->is_smsg_dup($smsg), "is_smsg_dup reject w/ $desc dedupe");
+       ok(!$dd->is_dup($sameish),
+               "Message-ID accounted for w/ same content otherwise");
 }
 $lei->{opt}->{dedupe} = 'bogus';
 eval { PublicInbox::LeiDedupe->new($lei) };