X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FContentHash.pm;h=d3ff146aa96e91ef3ada23c29689de0a33f6b2e0;hb=e6aa13bccb7ea5d5b3246b3a944621515905e360;hp=bacc9cdda12498abbb0ada5d2a2e2faec10190f2;hpb=f28fdcd6d8d6ac36c7b6adf6670238426f3cc067;p=public-inbox.git

diff --git a/lib/PublicInbox/ContentHash.pm b/lib/PublicInbox/ContentHash.pm
index bacc9cdd..d3ff146a 100644
--- a/lib/PublicInbox/ContentHash.pm
+++ b/lib/PublicInbox/ContentHash.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2018-2021 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>
 
 # Unstable internal API.
@@ -15,7 +15,8 @@ use PublicInbox::MID qw(mids references);
 use PublicInbox::MsgIter;
 
 # not sure if less-widely supported hash families are worth bothering with
-use Digest::SHA;
+use PublicInbox::SHA; # faster, but no ->clone
+use Digest::SHA; # we still need this for ->clone
 
 sub digest_addr ($$$) {
 	my ($dig, $h, $v) = @_;
@@ -63,8 +64,9 @@ sub content_digest ($;$) {
 	# 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)};
-	foreach my $mid (@{references($eml)}) {
-		$dig->add("ref\0$mid\0") unless $seen{$mid}++;
+	for (grep { !$seen{$_}++ } @{references($eml)}) {
+		utf8::encode($_);
+		$dig->add("ref\0$_\0");
 	}
 
 	# Only use Sender: if From is not present
@@ -92,15 +94,15 @@ sub content_digest ($;$) {
 }
 
 sub content_hash ($) {
-	content_digest($_[0])->digest;
+	content_digest($_[0], PublicInbox::SHA->new(256))->digest;
 }
 
+# don't clone the result of this
 sub git_sha ($$) {
 	my ($n, $eml) = @_;
-	my $dig = Digest::SHA->new($n);
+	my $dig = PublicInbox::SHA->new($n);
 	my $bref = ref($eml) eq 'SCALAR' ? $eml : \($eml->as_string);
-	$dig->add('blob '.length($$bref)."\0");
-	$dig->add($$bref);
+	$dig->add('blob '.length($$bref)."\0", $$bref);
 	$dig;
 }