lib/PublicInbox/ExtSearchIdx.pm | 11 ++++-------
 lib/PublicInbox/Import.pm       |  4 +---
 lib/PublicInbox/LeiStore.pm     |  4 ----
 lib/PublicInbox/SearchIdx.pm    | 17 +++--------------
 lib/PublicInbox/Smsg.pm         | 13 +++++++++++++
 lib/PublicInbox/V2Writable.pm   | 23 ++++++++++-------------
 t/import.t                      | 12 ++----------
 t/search.t                      |  2 +-

diff --git a/lib/PublicInbox/ExtSearchIdx.pm b/lib/PublicInbox/ExtSearchIdx.pm
index d55d3db9d31dd935a37eb9762a0d1b32295fcc7f..e6c21866b084fd4aa915cd6ac4c13c1194f8c6d1 100644
--- a/lib/PublicInbox/ExtSearchIdx.pm
+++ b/lib/PublicInbox/ExtSearchIdx.pm
@@ -21,8 +21,7 @@ use Carp qw(croak carp);
 use Sys::Hostname qw(hostname);
 use POSIX qw(strftime);
 use PublicInbox::Search;
-use PublicInbox::SearchIdx qw(crlf_adjust prepare_stack is_ancestor
-	is_bad_blob);
+use PublicInbox::SearchIdx qw(prepare_stack is_ancestor is_bad_blob);
 use PublicInbox::OverIdx;
 use PublicInbox::MiscIdx;
 use PublicInbox::MID qw(mids);
@@ -82,8 +81,6 @@ sub check_batch_limit ($) {
 	my ($req) = @_;
 	my $self = $req->{self};
 	my $new_smsg = $req->{new_smsg};
-
-	# {raw_bytes} may be unset, so just use {bytes}
 	my $n = $self->{transact_bytes} += $new_smsg->{bytes};
 
 	# set flag for PublicInbox::V2Writable::index_todo:
@@ -239,7 +236,7 @@ 	return if is_bad_blob($oid, $type, $size, $req->{oid});
 	my $new_smsg = $req->{new_smsg} = bless {
 		blob => $oid,
 	}, 'PublicInbox::Smsg';
-	$new_smsg->{bytes} = $size + crlf_adjust($$bref);
+	$new_smsg->set_bytes($$bref, $size);
 	defined($req->{xnum} = cur_ibx_xnum($req, $bref)) or return;
 	++${$req->{nr}};
 	do_step($req);
@@ -496,7 +493,7 @@ 	}
 	my $ci = $self->{current_info};
 	local $self->{current_info} = "$ci #$docid $oid";
 	my $re_smsg = bless { blob => $oid }, 'PublicInbox::Smsg';
-	$re_smsg->{bytes} = $size + crlf_adjust($$bref);
+	$re_smsg->set_bytes($$bref, $size);
 	my $eml = PublicInbox::Eml->new($bref);
 	$re_smsg->populate($eml, { autime => $orig_smsg->{ds},
 				cotime => $orig_smsg->{ts} });
@@ -676,7 +673,7 @@ 	return if is_bad_blob($oid, $type, $size, $req->{oid});
 	my $self = $req->{self} // die 'BUG: {self} unset';
 	local $self->{current_info} = "$self->{current_info} $oid";
 	my $new_smsg = bless { blob => $oid, }, 'PublicInbox::Smsg';
-	$new_smsg->{bytes} = $size + crlf_adjust($$bref);
+	$new_smsg->set_bytes($$bref, $size);
 	my $eml = $req->{eml} = PublicInbox::Eml->new($bref);
 	$req->{new_smsg} = $new_smsg;
 	$req->{chash} = content_hash($eml);
diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 052b145b2606123599b0aa451ae971c5ae84d1e2..8a06a661654b861975efbd88cc0df7fdcb010122 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -412,12 +412,10 @@
 	# v2: we need this for Xapian
 	if ($smsg) {
 		$smsg->{blob} = $self->get_mark(":$blob");
-		$smsg->{raw_bytes} = $n;
+		$smsg->set_bytes($raw_email, $n);
 		if (my $oidx = delete $smsg->{-oidx}) { # used by LeiStore
 			return if $oidx->blob_exists($smsg->{blob});
 		}
-		# XXX do we need this? it's in git at this point
-		$smsg->{-raw_email} = \$raw_email;
 	}
 	my $ref = $self->{ref};
 	my $commit = $self->{mark}++;
diff --git a/lib/PublicInbox/LeiStore.pm b/lib/PublicInbox/LeiStore.pm
index 4f77e8fad9d318fe0ce354c3536b5cdd176f8224..7cda7e449e2f8faa69e46c2eee60156631510778 100644
--- a/lib/PublicInbox/LeiStore.pm
+++ b/lib/PublicInbox/LeiStore.pm
@@ -10,7 +10,6 @@ package PublicInbox::LeiStore;
 use strict;
 use v5.10.1;
 use parent qw(PublicInbox::Lock PublicInbox::IPC);
-use PublicInbox::SearchIdx qw(crlf_adjust);
 use PublicInbox::ExtSearchIdx;
 use PublicInbox::Import;
 use PublicInbox::InboxWritable;
@@ -197,9 +196,6 @@ 	my $oidx = $eidx->{oidx};
 	my $smsg = bless { -oidx => $oidx }, 'PublicInbox::Smsg';
 	my $im = $self->importer;
 	$im->add($eml, undef, $smsg) or return; # duplicate returns undef
-	my $msgref = delete $smsg->{-raw_email};
-	$smsg->{bytes} = $smsg->{raw_bytes} + crlf_adjust($$msgref);
-	undef $msgref;
 
 	local $self->{current_info} = $smsg->{blob};
 	if (my @docids = _docids_for($self, $eml)) {
diff --git a/lib/PublicInbox/SearchIdx.pm b/lib/PublicInbox/SearchIdx.pm
index da3ac2e3abbb19e8471feb270b2a5c151b931a49..a7005051d9deef241fab065dd05072348e80d8ce 100644
--- a/lib/PublicInbox/SearchIdx.pm
+++ b/lib/PublicInbox/SearchIdx.pm
@@ -22,7 +22,7 @@ use PublicInbox::OverIdx;
 use PublicInbox::Spawn qw(spawn nodatacow_dir);
 use PublicInbox::Git qw(git_unquote);
 use PublicInbox::MsgTime qw(msg_timestamp msg_datestamp);
-our @EXPORT_OK = qw(crlf_adjust log2stack is_ancestor check_size prepare_stack
+our @EXPORT_OK = qw(log2stack is_ancestor check_size prepare_stack
 	index_text term_generator add_val is_bad_blob);
 my $X = \%PublicInbox::Search::X;
 our ($DB_CREATE_OR_OPEN, $DB_OPEN);
@@ -613,17 +613,6 @@ 		$mm->mid_insert($mids->[0]) // $mm->num_for($mids->[0]);
 	}
 }
 
-# returns the number of bytes to add if given a non-CRLF arg
-sub crlf_adjust ($) {
-	if (index($_[0], "\r\n") < 0) {
-		# common case is LF-only, every \n needs an \r;
-		# so favor a cheap tr// over an expensive m//g
-		$_[0] =~ tr/\n/\n/;
-	} else { # count number of '\n' w/o '\r', expensive:
-		scalar(my @n = ($_[0] =~ m/(?<!\r)\n/g));
-	}
-}
-
 sub is_bad_blob ($$$$) {
 	my ($oid, $type, $size, $expect_oid) = @_;
 	if ($type ne 'blob') {
@@ -640,8 +629,8 @@ 	return if is_bad_blob($oid, $type, $size, $sync->{oid});
 	my ($nr, $max) = @$sync{qw(nr max)};
 	++$$nr;
 	$$max -= $size;
-	$size += crlf_adjust($$bref);
-	my $smsg = bless { bytes => $size, blob => $oid }, 'PublicInbox::Smsg';
+	my $smsg = bless { blob => $oid }, 'PublicInbox::Smsg';
+	$smsg->set_bytes($$bref, $size);
 	my $self = $sync->{sidx};
 	local $self->{current_info} = "$self->{current_info}: $oid";
 	my $eml = PublicInbox::Eml->new($bref);
diff --git a/lib/PublicInbox/Smsg.pm b/lib/PublicInbox/Smsg.pm
index 571cbb6faca9417d804b1771b0bb94fdac5e6987..c6ff7f52b0ac90a2aad2326f80c4b862b4cc982b 100644
--- a/lib/PublicInbox/Smsg.pm
+++ b/lib/PublicInbox/Smsg.pm
@@ -135,4 +135,17 @@ 	$subj =~ s/$REPLY_RE//igo; # remove reply prefix
 	$subj;
 }
 
+# returns the number of bytes to add if given a non-CRLF arg
+sub crlf_adjust ($) {
+	if (index($_[0], "\r\n") < 0) {
+		# common case is LF-only, every \n needs an \r;
+		# so favor a cheap tr// over an expensive m//g
+		$_[0] =~ tr/\n/\n/;
+	} else { # count number of '\n' w/o '\r', expensive:
+		scalar(my @n = ($_[0] =~ m/(?<!\r)\n/g));
+	}
+}
+
+sub set_bytes { $_[0]->{bytes} = $_[2] + crlf_adjust($_[1]) }
+
 1;
diff --git a/lib/PublicInbox/V2Writable.pm b/lib/PublicInbox/V2Writable.pm
index 7b6b93a0d124baf8635cf13fc8d6ecc08221980d..c4efbdd203ab71497af35f99fdcc4fd3d9b90ffa 100644
--- a/lib/PublicInbox/V2Writable.pm
+++ b/lib/PublicInbox/V2Writable.pm
@@ -17,8 +17,7 @@ use PublicInbox::InboxWritable;
 use PublicInbox::OverIdx;
 use PublicInbox::Msgmap;
 use PublicInbox::Spawn qw(spawn popen_rd run_die);
-use PublicInbox::SearchIdx qw(log2stack crlf_adjust is_ancestor check_size
-	is_bad_blob);
+use PublicInbox::SearchIdx qw(log2stack is_ancestor check_size is_bad_blob);
 use IO::Handle; # ->autoflush
 use File::Temp ();
 
@@ -139,13 +138,12 @@ 	$self->{idx_shards}->[$num % scalar(@{$self->{idx_shards}})];
 }
 
 # indexes a message, returns true if checkpointing is needed
-sub do_idx ($$$$) {
-	my ($self, $msgref, $eml, $smsg) = @_;
-	$smsg->{bytes} = $smsg->{raw_bytes} + crlf_adjust($$msgref);
+sub do_idx ($$$) {
+	my ($self, $eml, $smsg) = @_;
 	$self->{oidx}->add_overview($eml, $smsg);
 	my $idx = idx_shard($self, $smsg->{num});
 	$idx->index_eml($eml, $smsg);
-	my $n = $self->{transact_bytes} += $smsg->{raw_bytes};
+	my $n = $self->{transact_bytes} += $smsg->{bytes};
 	$n >= $self->{batch_bytes};
 }
 
@@ -173,7 +171,7 @@ 	my $cmt = $im->add($mime, undef, $smsg); # sets $smsg->{ds|ts|blob}
 	$cmt = $im->get_mark($cmt);
 	$self->{last_commit}->[$self->{epoch_max}] = $cmt;
 
-	if (do_idx($self, delete $smsg->{-raw_email}, $mime, $smsg)) {
+	if (do_idx($self, $mime, $smsg)) {
 		$self->checkpoint;
 	}
 
@@ -536,13 +534,13 @@ 	# reindex modified messages:
 	for my $smsg (@$need_reindex) {
 		my $new_smsg = bless {
 			blob => $blob,
-			raw_bytes => $bytes,
 			num => $smsg->{num},
 			mid => $smsg->{mid},
 		}, 'PublicInbox::Smsg';
 		my $sync = { autime => $smsg->{ds}, cotime => $smsg->{ts} };
 		$new_smsg->populate($new_mime, $sync);
-		do_idx($self, \$raw, $new_mime, $new_smsg);
+		$new_smsg->set_bytes($raw, $bytes);
+		do_idx($self, $new_mime, $new_smsg);
 	}
 	$rewritten->{rewrites};
 }
@@ -937,13 +935,13 @@ 		return;
 	}
 	++${$arg->{nr}};
 	my $smsg = bless {
-		raw_bytes => $size,
 		num => $num,
 		blob => $oid,
 		mid => $mid0,
 	}, 'PublicInbox::Smsg';
 	$smsg->populate($eml, $arg);
-	if (do_idx($self, $bref, $eml, $smsg)) {
+	$smsg->set_bytes($$bref, $size);
+	if (do_idx($self, $eml, $smsg)) {
 		${$arg->{need_checkpoint}} = 1;
 	}
 	index_finalize($arg, 1);
@@ -1217,9 +1215,8 @@ sub index_xap_only { # git->cat_async callback
 	my ($bref, $oid, $type, $size, $smsg) = @_;
 	my $self = $smsg->{self};
 	my $idx = idx_shard($self, $smsg->{num});
-	$smsg->{raw_bytes} = $size;
 	$idx->index_eml(PublicInbox::Eml->new($bref), $smsg);
-	$self->{transact_bytes} += $size;
+	$self->{transact_bytes} += $smsg->{bytes};
 }
 
 sub index_xap_step ($$$;$) {
diff --git a/t/import.t b/t/import.t
index 855b5d7cff13207ce0e5847c9f76d7b2222be588..ae76858ba2ab7827bc7aaa1d45db377a0954ee64 100644
--- a/t/import.t
+++ b/t/import.t
@@ -7,7 +7,6 @@ use PublicInbox::Eml;
 use PublicInbox::Smsg;
 use PublicInbox::Git;
 use PublicInbox::Import;
-use PublicInbox::Spawn qw(spawn);
 use Fcntl qw(:DEFAULT SEEK_SET);
 use PublicInbox::TestCommon;
 use MIME::Base64 3.05; # Perl 5.10.0 / 5.9.2
@@ -32,20 +31,13 @@ like($im->add($mime, undef, $smsg), qr/\A:[0-9]+\z/, 'added one message');
 
 if ($v2) {
 	like($smsg->{blob}, qr/\A[a-f0-9]{40}\z/, 'got last object_id');
-	my $raw_email = $smsg->{-raw_email};
-	is($mime->as_string, $$raw_email, 'string matches');
-	is($smsg->{raw_bytes}, length($$raw_email), 'length matches');
 	my @cmd = ('git', "--git-dir=$git->{git_dir}", qw(hash-object --stdin));
 	open my $in, '+<', undef or BAIL_OUT "open(+<): $!";
 	print $in $mime->as_string or die "write failed: $!";
 	$in->flush or die "flush failed: $!";
-	seek($in, 0, SEEK_SET);
-	open my $out, '+<', undef or BAIL_OUT "open(+<): $!";
-	my $pid = spawn(\@cmd, {}, { 0 => $in, 1 => $out });
-	is(waitpid($pid, 0), $pid, 'waitpid succeeds on hash-object');
+	seek($in, 0, SEEK_SET) or die "seek: $!";
+	chomp(my $hashed_obj = xqx(\@cmd, undef, { 0 => $in }));
 	is($?, 0, 'hash-object');
-	seek($out, 0, SEEK_SET);
-	chomp(my $hashed_obj = <$out>);
 	is($hashed_obj, $smsg->{blob}, "blob object_id matches exp");
 }
 
diff --git a/t/search.t b/t/search.t
index 7495233ed20a881ceb8fa5f9deb077b516ef638a..b2958c002c2a85a1a52f26b9e64e7115a6aad3c0 100644
--- a/t/search.t
+++ b/t/search.t
@@ -60,7 +60,7 @@ 	}
 }
 
 {
-	my $crlf_adjust = \&PublicInbox::SearchIdx::crlf_adjust;
+	my $crlf_adjust = \&PublicInbox::Smsg::crlf_adjust;
 	is($crlf_adjust->("hi\r\nworld\r\n"), 0, 'no adjustment needed');
 	is($crlf_adjust->("hi\nworld\n"), 2, 'LF-only counts two CR');
 	is($crlf_adjust->("hi\r\nworld\n"), 1, 'CRLF/LF-mix 1 counts 1 CR');