[PATCH 05/12] v2writable: initial cut for repo-rotation

["Eric Wong (Contractor, The Linux Foundation)" <e@80x24.org>]

Wrap the old Import package to enable creating new repos based
on size thresholds.  This is better than relying on time-based
rotation as LKML traffic seems to be increasing.
---
 MANIFEST                      |   1 +
 lib/PublicInbox/Git.pm        |  12 +++
 lib/PublicInbox/Import.pm     |   9 ++-
 lib/PublicInbox/V2Writable.pm | 180 ++++++++++++++++++++++++++++++++++++++++++
 scripts/import_vger_from_mbox |  19 +++--
 t/git.t                       |   3 +
 6 files changed, 211 insertions(+), 13 deletions(-)
 create mode 100644 lib/PublicInbox/V2Writable.pm

diff --git a/MANIFEST b/MANIFEST
index 1df27f2..4b51b54 100644
--- a/MANIFEST
+++ b/MANIFEST
@@ -91,6 +91,7 @@ lib/PublicInbox/Spamcheck/Spamc.pm
 lib/PublicInbox/Spawn.pm
 lib/PublicInbox/SpawnPP.pm
 lib/PublicInbox/Unsubscribe.pm
+lib/PublicInbox/V2Writable.pm
 lib/PublicInbox/View.pm
 lib/PublicInbox/WWW.pm
 lib/PublicInbox/WWW.pod
diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm
index ea2b814..6437643 100644
--- a/lib/PublicInbox/Git.pm
+++ b/lib/PublicInbox/Git.pm
@@ -139,6 +139,18 @@ sub cleanup {
 	_destroy($self, qw(in_c out_c pid_c));
 }
 
+# assuming a well-maintained repo, this should be a somewhat
+# accurate estimation of its size
+# TODO: show this in the WWW UI as a hint to potential cloners
+sub packed_bytes {
+	my ($self) = @_;
+	my $n = 0;
+	foreach my $p (glob("$self->{git_dir}/objects/pack/*.pack")) {
+		$n += -s $p;
+	}
+	$n
+}
+
 sub DESTROY { cleanup(@_) }
 
 1;
diff --git a/lib/PublicInbox/Import.pm b/lib/PublicInbox/Import.pm
index 1f831a7..364ab60 100644
--- a/lib/PublicInbox/Import.pm
+++ b/lib/PublicInbox/Import.pm
@@ -31,6 +31,7 @@ sub new {
 		inbox => $ibx,
 		path_type => '2/38', # or 'v2'
 		ssoma_lock => 1, # disable for v2
+		bytes_added => 0,
 	}, $class
 }
 
@@ -275,7 +276,9 @@ sub add {
 
 	my $blob = $self->{mark}++;
 	my $str = $mime->as_string;
-	print $w "blob\nmark :$blob\ndata ", length($str), "\n" or wfail;
+	my $n = length($str);
+	$self->{bytes_added} += $n;
+	print $w "blob\nmark :$blob\ndata ", $n, "\n" or wfail;
 	print $w $str, "\n" or wfail;
 	$str = undef;
 
@@ -325,7 +328,7 @@ sub add {
 	$self->{tip} = ":$commit";
 }
 
-sub run_die ($$) {
+sub run_die ($;$) {
 	my ($cmd, $env) = @_;
 	my $pid = spawn($cmd, $env, undef);
 	defined $pid or die "spawning ".join(' ', @$cmd)." failed: $!";
@@ -354,7 +357,7 @@ sub done {
 	}
 	if ($nchg) {
 		run_die([@cmd, 'update-server-info'], undef);
-		eval {
+		($self->{path_type} eq '2/38') and eval {
 			require PublicInbox::SearchIdx;
 			my $inbox = $self->{inbox} || $git_dir;
 			my $s = PublicInbox::SearchIdx->new($inbox);
diff --git a/lib/PublicInbox/V2Writable.pm b/lib/PublicInbox/V2Writable.pm
new file mode 100644
index 0000000..9b68e9b
--- /dev/null
+++ b/lib/PublicInbox/V2Writable.pm
@@ -0,0 +1,180 @@
+# Copyright (C) 2018 all contributors <meta@public-inbox.org>
+# License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
+
+# This interface wraps and mimics PublicInbox::Import
+package PublicInbox::V2Writable;
+use strict;
+use warnings;
+use Fcntl qw(:flock :DEFAULT);
+use PublicInbox::SearchIdx;
+use PublicInbox::MIME;
+use PublicInbox::Git;
+use PublicInbox::Import;
+use Email::MIME::ContentType;
+$Email::MIME::ContentType::STRICT_PARAMS = 0;
+
+# an estimate of the post-packed size to the raw uncompressed size
+my $PACKING_FACTOR = 0.4;
+
+sub new {
+	my ($class, $v2ibx, $creat) = @_;
+	my $dir = $v2ibx->{mainrepo} or die "no mainrepo in inbox\n";
+	unless (-d $dir) {
+		if ($creat) {
+			require File::Path;
+			File::Path::mkpath($dir);
+		} else {
+			die "$dir does not exist\n";
+		}
+	}
+	my $self = {
+		-inbox => $v2ibx,
+		im => undef, #  PublicInbox::Import
+		xap_rw => undef, # PublicInbox::V2SearchIdx
+		xap_ro => undef,
+
+		# limit each repo to 1GB or so
+		rotate_bytes => int((100 * 1024 * 1024) / $PACKING_FACTOR),
+	};
+	bless $self, $class
+}
+
+# returns undef on duplicate or spam
+# mimics Import::add and wraps it for v2
+sub add {
+	my ($self, $mime, $check_cb) = @_;
+	my $existing = $self->lookup_content($mime);
+
+	if ($existing) {
+		return undef if $existing->type eq 'mail'; # duplicate
+	}
+
+	my $im = $self->importer;
+
+	# im->add returns undef if check_cb fails
+	my $cmt = $im->add($mime, $check_cb) or return;
+	$cmt = $im->get_mark($cmt);
+	my $oid = $im->{last_object_id};
+	$self->index_msg($mime, $existing, $cmt, $oid);
+	$mime;
+}
+
+sub index_msg {  # TODO
+}
+
+sub remove {
+	my ($self, $mime, $msg) = @_;
+	my $existing = $self->lookup_content($mime) or return;
+
+	# don't touch ghosts or already junked messages
+	return unless $existing->type eq 'mail';
+
+	# always write removals to the current (latest) git repo since
+	# we process chronologically
+	my $im = $self->importer;
+	my ($cmt, undef) = $im->remove($mime, $msg);
+	$cmt = $im->get_mark($cmt);
+	$self->unindex_msg($existing, $cmt);
+}
+
+sub done {
+	my ($self) = @_;
+	$self->{im}->done; # PublicInbox::Import::done
+}
+
+sub checkpoint {
+	my ($self) = @_;
+	$self->{im}->checkpoint; # PublicInbox::Import::checkpoint
+}
+
+sub git_init {
+	my ($self, $new) = @_;
+	my $pfx = "$self->{-inbox}->{mainrepo}/git";
+	my $git_dir = "$pfx/$new.git";
+	die "$git_dir exists\n" if -e $git_dir;
+	my @cmd = (qw(git init --bare -q), $git_dir);
+	PublicInbox::Import::run_die(\@cmd);
+	@cmd = (qw/git config/, "--file=$git_dir/config",
+			'repack.writeBitmaps', 'true');
+	PublicInbox::Import::run_die(\@cmd);
+
+	my $all = "$self->{-inbox}->{mainrepo}/all.git";
+	unless (-d $all) {
+		@cmd = (qw(git init --bare -q), $all);
+		PublicInbox::Import::run_die(\@cmd);
+	}
+
+	my $alt = "$all/objects/info/alternates";
+	my $new_obj_dir = "../../git/$new.git/objects";
+	my %alts;
+	if (-e $alt) {
+		open(my $fh, '<', $alt) or die "open < $alt: $!\n";
+		%alts = map { chomp; $_ => 1 } (<$fh>);
+	}
+	return $git_dir if $alts{$new_obj_dir};
+	open my $fh, '>>', $alt or die "open >> $alt: $!\n";
+	print $fh "$new_obj_dir\n" or die "print >> $alt: $!\n";
+	close $fh or die "close $alt: $!\n";
+	$git_dir
+}
+
+sub importer {
+	my ($self) = @_;
+	my $im = $self->{im};
+	if ($im) {
+		if ($im->{bytes_added} < $self->{rotate_bytes}) {
+			return $im;
+		} else {
+			$self->{im} = undef;
+			$im->done;
+			$im = undef;
+			my $git_dir = $self->git_init(++$self->{max_git});
+			my $git = PublicInbox::Git->new($git_dir);
+			return $self->import_init($git, 0);
+		}
+	}
+	my $latest;
+	my $max = -1;
+	my $new = 0;
+	my $pfx = "$self->{-inbox}->{mainrepo}/git";
+	if (-d $pfx) {
+		foreach my $git_dir (glob("$pfx/*.git")) {
+			$git_dir =~ m!/(\d+)\.git\z! or next;
+			my $n = $1;
+			if ($n > $max) {
+				$max = $n;
+				$latest = $git_dir;
+			}
+		}
+	}
+	if (defined $latest) {
+		my $git = PublicInbox::Git->new($latest);
+		my $packed_bytes = $git->packed_bytes;
+		if ($packed_bytes >= $self->{rotate_bytes}) {
+			$new = $max + 1;
+		} else {
+			$self->{max_git} = $max;
+			return $self->import_init($git, $packed_bytes);
+		}
+	} else {
+		warn "latest not found in $pfx\n";
+	}
+	$self->{max_git} = $new;
+	$latest = $self->git_init($new);
+	$self->import_init(PublicInbox::Git->new($latest), 0);
+}
+
+sub import_init {
+	my ($self, $git, $packed_bytes) = @_;
+	my $im = PublicInbox::Import->new($git, undef, undef, $self->{-inbox});
+	$im->{bytes_added} = int($packed_bytes / $PACKING_FACTOR);
+	$im->{ssoma_lock} = 0;
+	$im->{path_type} = 'v2';
+	$self->{im} = $im;
+}
+
+sub lookup_content {
+	undef # TODO
+}
+
+1;
diff --git a/scripts/import_vger_from_mbox b/scripts/import_vger_from_mbox
index 6ea2ca5..c45dc4e 100644
--- a/scripts/import_vger_from_mbox
+++ b/scripts/import_vger_from_mbox
@@ -7,25 +7,24 @@ use Getopt::Long qw/:config gnu_getopt no_ignore_case auto_abbrev/;
 use Date::Parse qw/str2time/;
 use Email::MIME;
 $Email::MIME::ContentType::STRICT_PARAMS = 0; # user input is imperfect
-use PublicInbox::Git;
-use PublicInbox::Import;
-my $usage = "usage: $0 NAME EMAIL <MBOX\n";
+use PublicInbox::V2Writable;
+my $usage = "usage: $0 NAME EMAIL DIR <MBOX\n";
 my $dry_run;
 my %opts = ( 'n|dry-run' => \$dry_run );
 GetOptions(%opts) or die $usage;
-chomp(my $git_dir = `git rev-parse --git-dir`);
-my $git = PublicInbox::Git->new($git_dir);
 my $name = shift or die $usage; # git
 my $email = shift or die $usage; # git@vger.kernel.org
-my $im = $dry_run ? undef : PublicInbox::Import->new($git, $name, $email);
+my $mainrepo = shift or die $usage; # /path/to/v2/repo
+my $v2ibx = {
+	mainrepo => $mainrepo,
+	name => $name,
+	-primary_address => $email,
+};
+my $im = $dry_run ? undef : PublicInbox::V2Writable->new($v2ibx, 1);
 binmode STDIN;
 my $msg = '';
 use PublicInbox::Filter::Vger;
 my $vger = PublicInbox::Filter::Vger->new;
-if ($im) {
-	$im->{ssoma_lock} = 0;
-	$im->{path_type} = 'v2';
-}
 
 sub do_add ($$) {
 	my ($im, $msg) = @_;
diff --git a/t/git.t b/t/git.t
index 5efc18a..ab588a1 100644
--- a/t/git.t
+++ b/t/git.t
@@ -137,6 +137,9 @@ if (1) {
 	is($all, join('', @ref), 'qx returned array when wanted');
 	my $nl = scalar @ref;
 	ok($nl > 1, "qx returned array length of $nl");
+
+	$gcf->qx(qw(repack -adbq));
+	ok($gcf->packed_bytes > 0, 'packed size is positive');
 }
 
 done_testing();
-- 
EW