convert: add "status=delayed" to filter process protocol

Some `clean` / `smudge` filters may require a significant amount of
time to process a single blob (e.g. the Git LFS smudge filter might
perform network requests). During this process the Git checkout
operation is blocked and Git needs to wait until the filter is done to
continue with the checkout.

Teach the filter process protocol, introduced in edcc8581 ("convert: add
filter.<driver>.process option", 2016-10-16), to accept the status
"delayed" as response to a filter request. Upon this response Git
continues with the checkout operation. After the checkout operation Git
calls "finish_delayed_checkout" which queries the filter for remaining
blobs. If the filter is still working on the completion, then the filter
is expected to block. If the filter has completed all remaining blobs
then an empty response is expected.

Git has a multiple code paths that checkout a blob. Support delayed
checkouts only in `clone` (in unpack-trees.c) and `checkout` operations
for now. The optimization is most effective in these code paths as all
files of the tree are processed.

Signed-off-by: Lars Schneider <larsxschneider@gmail.com>
Signed-off-by: Junio C Hamano <gitster@pobox.com>

1 files changed, 139 insertions, 65 deletions

diff --git a/t/t0021/rot13-filter.pl b/t/t0021/rot13-filter.pl
index 5e43faeec1..ad685d92f8 100644
--- a/t/t0021/rot13-filter.pl
+++ b/t/t0021/rot13-filter.pl
@@ -18,6 +18,16 @@
 #     operation then the filter signals that it cannot or does not want
 #     to process the file and any file after that is processed with the
 #     same command.
+# (5) If data with a pathname that is a key in the DELAY hash is
+#     requested (e.g. "test-delay10.a") then the filter responds with
+#     a "delay" status and sets the "requested" field in the DELAY hash.
+#     The filter will signal the availability of this object after
+#     "count" (field in DELAY hash) "list_available_blobs" commands.
+# (6) If data with the pathname "missing-delay.a" is processed that the
+#     filter will drop the path from the "list_available_blobs" response.
+# (7) If data with the pathname "invalid-delay.a" is processed that the
+#     filter will add the path "unfiltered" which was not delayed before
+#     to the "list_available_blobs" response.
 #
 
 use strict;
@@ -30,6 +40,15 @@ my @capabilities            = @ARGV;
 
 open my $debug, ">>", $log_file or die "cannot open log file: $!";
 
+my %DELAY = (
+	'test-delay10.a' => { "requested" => 0, "count" => 1 },
+	'test-delay11.a' => { "requested" => 0, "count" => 1 },
+	'test-delay20.a' => { "requested" => 0, "count" => 2 },
+	'test-delay10.b' => { "requested" => 0, "count" => 1 },
+	'missing-delay.a' => { "requested" => 0, "count" => 1 },
+	'invalid-delay.a' => { "requested" => 0, "count" => 1 },
+);
+
 sub rot13 {
 	my $str = shift;
 	$str =~ y/A-Za-z/N-ZA-Mn-za-m/;
@@ -66,7 +85,7 @@ sub packet_bin_read {
 
 sub packet_txt_read {
 	my ( $res, $buf ) = packet_bin_read();
-	unless ( $buf =~ s/\n$// ) {
+	unless ( $buf eq '' or $buf =~ s/\n$// ) {
 		die "A non-binary line MUST be terminated by an LF.";
 	}
 	return ( $res, $buf );
@@ -101,6 +120,7 @@ packet_flush();
 
 ( packet_txt_read() eq ( 0, "capability=clean" ) )  || die "bad capability";
 ( packet_txt_read() eq ( 0, "capability=smudge" ) ) || die "bad capability";
+( packet_txt_read() eq ( 0, "capability=delay" ) )  || die "bad capability";
 ( packet_bin_read() eq ( 1, "" ) )                  || die "bad capability end";
 
 foreach (@capabilities) {
@@ -111,88 +131,142 @@ print $debug "init handshake complete\n";
 $debug->flush();
 
 while (1) {
-	my ($command) = packet_txt_read() =~ /^command=(.+)$/;
+	my ( $command ) = packet_txt_read() =~ /^command=(.+)$/;
 	print $debug "IN: $command";
 	$debug->flush();
 
-	my ($pathname) = packet_txt_read() =~ /^pathname=(.+)$/;
-	print $debug " $pathname";
-	$debug->flush();
-
-	if ( $pathname eq "" ) {
-		die "bad pathname '$pathname'";
-	}
+	if ( $command eq "list_available_blobs" ) {
+		# Flush
+		packet_bin_read();
 
-	# Flush
-	packet_bin_read();
-
-	my $input = "";
-	{
-		binmode(STDIN);
-		my $buffer;
-		my $done = 0;
-		while ( !$done ) {
-			( $done, $buffer ) = packet_bin_read();
-			$input .= $buffer;
+		foreach my $pathname ( sort keys %DELAY ) {
+			if ( $DELAY{$pathname}{"requested"} >= 1 ) {
+				$DELAY{$pathname}{"count"} = $DELAY{$pathname}{"count"} - 1;
+				if ( $pathname eq "invalid-delay.a" ) {
+					# Send Git a pathname that was not delayed earlier
+					packet_txt_write("pathname=unfiltered");
+				}
+				if ( $pathname eq "missing-delay.a" ) {
+					# Do not signal Git that this file is available
+				} elsif ( $DELAY{$pathname}{"count"} == 0 ) {
+					print $debug " $pathname";
+					packet_txt_write("pathname=$pathname");
+				}
+			}
 		}
-		print $debug " " . length($input) . " [OK] -- ";
-		$debug->flush();
-	}
-
-	my $output;
-	if ( $pathname eq "error.r" or $pathname eq "abort.r" ) {
-		$output = "";
-	}
-	elsif ( $command eq "clean" and grep( /^clean$/, @capabilities ) ) {
-		$output = rot13($input);
-	}
-	elsif ( $command eq "smudge" and grep( /^smudge$/, @capabilities ) ) {
-		$output = rot13($input);
-	}
-	else {
-		die "bad command '$command'";
-	}
 
-	if ( $pathname eq "error.r" ) {
-		print $debug "[ERROR]\n";
-		$debug->flush();
-		packet_txt_write("status=error");
 		packet_flush();
-	}
-	elsif ( $pathname eq "abort.r" ) {
-		print $debug "[ABORT]\n";
+
+		print $debug " [OK]\n";
 		$debug->flush();
-		packet_txt_write("status=abort");
+		packet_txt_write("status=success");
 		packet_flush();
 	}
 	else {
-		packet_txt_write("status=success");
-		packet_flush();
+		my ( $pathname ) = packet_txt_read() =~ /^pathname=(.+)$/;
+		print $debug " $pathname";
+		$debug->flush();
+
+		if ( $pathname eq "" ) {
+			die "bad pathname '$pathname'";
+		}
+
+		# Read until flush
+		my ( $done, $buffer ) = packet_txt_read();
+		while ( $buffer ne '' ) {
+			if ( $buffer eq "can-delay=1" ) {
+				if ( exists $DELAY{$pathname} and $DELAY{$pathname}{"requested"} == 0 ) {
+					$DELAY{$pathname}{"requested"} = 1;
+				}
+			} else {
+				die "Unknown message '$buffer'";
+			}
 
-		if ( $pathname eq "${command}-write-fail.r" ) {
-			print $debug "[WRITE FAIL]\n";
+			( $done, $buffer ) = packet_txt_read();
+		}
+
+		my $input = "";
+		{
+			binmode(STDIN);
+			my $buffer;
+			my $done = 0;
+			while ( !$done ) {
+				( $done, $buffer ) = packet_bin_read();
+				$input .= $buffer;
+			}
+			print $debug " " . length($input) . " [OK] -- ";
 			$debug->flush();
-			die "${command} write error";
 		}
 
-		print $debug "OUT: " . length($output) . " ";
-		$debug->flush();
+		my $output;
+		if ( exists $DELAY{$pathname} and exists $DELAY{$pathname}{"output"} ) {
+			$output = $DELAY{$pathname}{"output"}
+		}
+		elsif ( $pathname eq "error.r" or $pathname eq "abort.r" ) {
+			$output = "";
+		}
+		elsif ( $command eq "clean" and grep( /^clean$/, @capabilities ) ) {
+			$output = rot13($input);
+		}
+		elsif ( $command eq "smudge" and grep( /^smudge$/, @capabilities ) ) {
+			$output = rot13($input);
+		}
+		else {
+			die "bad command '$command'";
+		}
+
+		if ( $pathname eq "error.r" ) {
+			print $debug "[ERROR]\n";
+			$debug->flush();
+			packet_txt_write("status=error");
+			packet_flush();
+		}
+		elsif ( $pathname eq "abort.r" ) {
+			print $debug "[ABORT]\n";
+			$debug->flush();
+			packet_txt_write("status=abort");
+			packet_flush();
+		}
+		elsif ( $command eq "smudge" and
+			exists $DELAY{$pathname} and
+			$DELAY{$pathname}{"requested"} == 1
+		) {
+			print $debug "[DELAYED]\n";
+			$debug->flush();
+			packet_txt_write("status=delayed");
+			packet_flush();
+			$DELAY{$pathname}{"requested"} = 2;
+			$DELAY{$pathname}{"output"} = $output;
+		}
+		else {
+			packet_txt_write("status=success");
+			packet_flush();
 
-		while ( length($output) > 0 ) {
-			my $packet = substr( $output, 0, $MAX_PACKET_CONTENT_SIZE );
-			packet_bin_write($packet);
-			# dots represent the number of packets
-			print $debug ".";
-			if ( length($output) > $MAX_PACKET_CONTENT_SIZE ) {
-				$output = substr( $output, $MAX_PACKET_CONTENT_SIZE );
+			if ( $pathname eq "${command}-write-fail.r" ) {
+				print $debug "[WRITE FAIL]\n";
+				$debug->flush();
+				die "${command} write error";
 			}
-			else {
-				$output = "";
+
+			print $debug "OUT: " . length($output) . " ";
+			$debug->flush();
+
+			while ( length($output) > 0 ) {
+				my $packet = substr( $output, 0, $MAX_PACKET_CONTENT_SIZE );
+				packet_bin_write($packet);
+				# dots represent the number of packets
+				print $debug ".";
+				if ( length($output) > $MAX_PACKET_CONTENT_SIZE ) {
+					$output = substr( $output, $MAX_PACKET_CONTENT_SIZE );
+				}
+				else {
+					$output = "";
+				}
 			}
+			packet_flush();
+			print $debug " [OK]\n";
+			$debug->flush();
+			packet_flush();
 		}
-		packet_flush();
-		print $debug " [OK]\n";
-		$debug->flush();
-		packet_flush();
 	}
 }