content_id: use `mids' and `references' for MID extraction

[public-inbox.git] / lib / PublicInbox / ContentId.pm

diff --git a/lib/PublicInbox/ContentId.pm b/lib/PublicInbox/ContentId.pm
index 65d5a76c09b43f4dec2edb08998c22ffe6d6e95e..d1a009e77f2ed9e9e8b4de30a79f85a706ef37e8 100644 (file)
--- a/lib/PublicInbox/ContentId.pm
+++ b/lib/PublicInbox/ContentId.pm
@@ -6,12 +6,13 @@ use strict;
 use warnings;
 use base qw/Exporter/;
 our @EXPORT_OK = qw/content_id/;
+use PublicInbox::MID qw(mids references);
 
 # not sure if less-widely supported hash families are worth bothering with
 use Digest::SHA;
 
 # Content-* headers are often no-ops, so maybe we don't need them
-my @ID_HEADERS = qw(Subject From Date Message-ID References To Cc In-Reply-To);
+my @ID_HEADERS = qw(Subject From Date To Cc);
 
 sub content_id ($;$) {
        my ($mime, $alg) = @_;
@@ -19,9 +20,21 @@ sub content_id ($;$) {
        my $dig = Digest::SHA->new($alg);
        my $hdr = $mime->header_obj;
 
+       # 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:
+       my %seen;
+       foreach my $mid (@{mids($hdr)}) {
+               $dig->add('mid: '.$mid);
+               $seen{$mid} = 1;
+       }
+       foreach my $mid (@{references($hdr)}) {
+               next if $seen{$mid};
+               $dig->add('ref: '.$mid);
+       }
        foreach my $h (@ID_HEADERS) {
                my @v = $hdr->header_raw($h);
-               $dig->add($_) foreach @v;
+               $dig->add("$h: $_") foreach @v;
        }
        $dig->add($mime->body_raw);
        'SHA-' . $dig->algorithm . ':' . $dig->hexdigest;