X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FImport.pm;h=31dc769583bd32777e2d25e16de743fd507a65c0;hb=3ef89f98a662a135ae18dddf3ff3e61e8e3ca996;hp=c2beb19c42e8fcfec42283df04f0d02ddbdef44c;hpb=5cbb8df04448baf6c07c3de652d287bd2a0a3299;p=public-inbox.git

diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index c2beb19c..31dc7695 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -54,6 +54,8 @@ sub gfi_start {
 	$self->{lockfh} = $lockfh;
 	$self->{pid} = $pid;
 	$self->{nchg} = 0;
+	binmode $out_w, ':raw' or die "binmode :raw failed: $!";
+	binmode $in_r, ':raw' or die "binmode :raw failed: $!";
 	($in_r, $out_w);
 }
 
@@ -75,6 +77,15 @@ sub norm_body ($) {
 	$b
 }
 
+sub _check_path ($$$$) {
+	my ($r, $w, $tip, $path) = @_;
+	return if $tip eq '';
+	print $w "ls $tip $path\n" or wfail;
+	local $/ = "\n";
+	defined(my $info = <$r>) or die "EOF from fast-import: $!";
+	$info =~ /\Amissing / ? undef : $info;
+}
+
 # returns undef on non-existent
 # ('MISMATCH', msg) on mismatch
 # (:MARK, msg) on success
@@ -86,20 +97,16 @@ sub remove {
 
 	my ($r, $w) = $self->gfi_start;
 	my $tip = $self->{tip};
-	return ('MISSING', undef) if $tip eq '';
-
-	print $w "ls $tip $path\n" or wfail;
-	local $/ = "\n";
-	my $check = <$r>;
-	defined $check or die "EOF from fast-import / ls: $!";
-	return ('MISSING', undef) if $check =~ /\Amissing /;
-	$check =~ m!\A100644 blob ([a-f0-9]{40})\t!s or die "not blob: $check";
+	my $info = _check_path($r, $w, $tip, $path) or return ('MISSING',undef);
+	$info =~ m!\A100644 blob ([a-f0-9]{40})\t!s or die "not blob: $info";
 	my $blob = $1;
+
 	print $w "cat-blob $blob\n" or wfail;
-	$check = <$r>;
-	defined $check or die "EOF from fast-import / cat-blob: $!";
-	$check =~ /\A[a-f0-9]{40} blob (\d+)\n\z/ or
-				die "unexpected cat-blob response: $check";
+	local $/ = "\n";
+	$info = <$r>;
+	defined $info or die "EOF from fast-import / cat-blob: $!";
+	$info =~ /\A[a-f0-9]{40} blob (\d+)\n\z/ or
+				die "unexpected cat-blob response: $info";
 	my $left = $1;
 	my $offset = 0;
 	my $buf = '';
@@ -114,7 +121,7 @@ sub remove {
 	$n = read($r, my $lf, 1);
 	defined($n) or die "read final byte of cat-blob failed: $!";
 	die "bad read on final byte: <$lf>" if $lf ne "\n";
-	my $cur = Email::MIME->new($buf);
+	my $cur = PublicInbox::MIME->new($buf);
 	my $cur_s = $cur->header('Subject');
 	$cur_s = '' unless defined $cur_s;
 	my $cur_m = $mime->header('Subject');
@@ -162,13 +169,7 @@ sub add {
 
 	my ($r, $w) = $self->gfi_start;
 	my $tip = $self->{tip};
-	if ($tip ne '') {
-		print $w "ls $tip $path\n" or wfail;
-		local $/ = "\n";
-		my $check = <$r>;
-		defined $check or die "EOF from fast-import: $!";
-		return unless $check =~ /\Amissing /;
-	}
+	_check_path($r, $w, $tip, $path) and return;
 
 	# kill potentially confusing/misleading headers
 	$mime->header_set($_) for qw(bytes lines content-length status);
@@ -188,15 +189,15 @@ sub add {
 		print $w "reset $ref\n" or wfail;
 	}
 
+	utf8::encode($email);
+	utf8::encode($name);
+	utf8::encode($subject);
 	# quiet down wide character warnings:
-	binmode $w, ':utf8' or die "binmode :utf8 failed: $!";
 	print $w "commit $ref\nmark :$commit\n",
 		"author $name <$email> $date\n",
-		"committer $self->{ident} ", now2822(), "\n",
-		"data ", (bytes::length($subject) + 1), "\n",
+		"committer $self->{ident} ", now2822(), "\n" or wfail;
+	print $w "data ", (length($subject) + 1), "\n",
 		$subject, "\n\n" or wfail;
-	binmode $w, ':raw' or die "binmode :raw failed: $!";
-
 	if ($tip ne '') {
 		print $w 'from ', ($parent ? $parent : $tip), "\n" or wfail;
 	}
@@ -205,6 +206,14 @@ sub add {
 	$self->{tip} = ":$commit";
 }
 
+sub run_die ($$) {
+	my ($cmd, $env) = @_;
+	my $pid = spawn($cmd, $env, undef);
+	defined $pid or die "spawning ".join(' ', @$cmd)." failed: $!";
+	waitpid($pid, 0) == $pid or die join(' ', @$cmd) .' did not finish';
+	$? == 0 or die join(' ', @$cmd) . " failed: $?\n";
+}
+
 sub done {
 	my ($self) = @_;
 	my $w = delete $self->{out} or return;
@@ -223,25 +232,18 @@ sub done {
 	if ($nchg && !$ENV{FAST}) {
 		my $index = "$git_dir/ssoma.index";
 		my $env = { GIT_INDEX_FILE => $index };
-		my @rt = (@cmd, qw(read-tree -m -v -i), $self->{ref});
-		$pid = spawn(\@rt, $env, undef);
-		defined $pid or die "spawn read-tree failed: $!";
-		waitpid($pid, 0) == $pid or die 'read-tree did not finish';
-		$? == 0 or die "failed to update $git_dir/ssoma.index: $?\n";
+		run_die([@cmd, qw(read-tree -m -v -i), $self->{ref}], $env);
 	}
 	if ($nchg) {
-		$pid = spawn([@cmd, 'update-server-info'], undef, undef);
-		defined $pid or die "spawn update-server-info failed: $!\n";
-		waitpid($pid, 0) == $pid or
-			die 'update-server-info did not finish';
-		$? == 0 or die "failed to update-server-info: $?\n";
-
+		run_die([@cmd, 'update-server-info'], undef);
 		eval {
 			require PublicInbox::SearchIdx;
 			my $inbox = $self->{inbox} || $git_dir;
 			my $s = PublicInbox::SearchIdx->new($inbox);
 			$s->index_sync({ ref => $self->{ref} });
 		};
+
+		eval { run_die([@cmd, qw(gc --auto)], undef) };
 	}
 
 	my $lockfh = delete $self->{lockfh} or die "BUG: not locked: $!";