From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <e@80x24.org>
X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net
X-Spam-Level: 
X-Spam-ASN:  
X-Spam-Status: No, score=-4.0 required=3.0 tests=ALL_TRUSTED,BAYES_00
	shortcircuit=no autolearn=ham autolearn_force=no version=3.4.2
Received: from localhost (dcvr.yhbt.net [127.0.0.1])
	by dcvr.yhbt.net (Postfix) with ESMTP id 4DB6D21420
	for <meta@public-inbox.org>; Mon, 21 Jan 2019 20:52:55 +0000 (UTC)
From: Eric Wong <e@80x24.org>
To: meta@public-inbox.org
Subject: [PATCH 07/37] git: check saves error on disambiguation
Date: Mon, 21 Jan 2019 20:52:23 +0000
Message-Id: <20190121205253.10455-8-e@80x24.org>
In-Reply-To: <20190121205253.10455-1-e@80x24.org>
References: <20190121205253.10455-1-e@80x24.org>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
List-Id: <meta.public-inbox.org>

This will be useful for disambiguating short OIDs in older
emails when abbreviations were shorter.

Tested against the following script with /path/to/git.git
==> t.perl <==
use strict;
use PublicInbox::Git;
use Data::Dumper;
my $dir = shift or die "Usage: $0 GIT_DIR # (of git.git)";
my $git = PublicInbox::Git->new($dir);
my @res = $git->check('dead');
print Dumper({res => \@res, err=> $git->last_check_err});
@res = $git->check('5335669531d83d7d6c905bcfca9b5f8e182dc4d4');
print Dumper({res => \@res, err=> $git->last_check_err});
---
 lib/PublicInbox/Git.pm | 38 +++++++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm
index d0ac6b6..a61cb31 100644
--- a/lib/PublicInbox/Git.pm
+++ b/lib/PublicInbox/Git.pm
@@ -64,9 +64,25 @@ sub alternates_changed {
 	$self->{st} = \@st;
 }
 
+sub last_check_err {
+	my ($self) = @_;
+	my $fh = $self->{err_c} or return;
+	sysseek($fh, 0, 0) or fail($self, "sysseek failed: $!");
+	defined(sysread($fh, my $buf, -s $fh)) or
+			fail($self, "sysread failed: $!");
+	$buf;
+}
+
 sub _bidi_pipe {
-	my ($self, $batch, $in, $out, $pid) = @_;
-	return if $self->{$pid};
+	my ($self, $batch, $in, $out, $pid, $err) = @_;
+	if ($self->{$pid}) {
+		if (defined $err) { # "err_c"
+			my $fh = $self->{$err};
+			sysseek($fh, 0, 0) or fail($self, "sysseek failed: $!");
+			truncate($fh, 0) or fail($self, "truncate failed: $!");
+		}
+		return;
+	}
 	my ($in_r, $in_w, $out_r, $out_w);
 
 	pipe($in_r, $in_w) or fail($self, "pipe failed: $!");
@@ -78,6 +94,11 @@ sub _bidi_pipe {
 
 	my @cmd = ('git', "--git-dir=$self->{git_dir}", qw(cat-file), $batch);
 	my $redir = { 0 => fileno($out_r), 1 => fileno($in_w) };
+	if ($err) {
+		open(my $fh, '+>', undef) or fail($self, "open.err failed: $!");
+		$self->{$err} = $fh;
+		$redir->{2} = fileno($fh);
+	}
 	my $p = spawn(\@cmd, undef, $redir);
 	defined $p or fail($self, "spawn failed: $!");
 	$self->{$pid} = $p;
@@ -152,12 +173,23 @@ sub batch_prepare ($) { _bidi_pipe($_[0], qw(--batch in out pid)) }
 
 sub check {
 	my ($self, $obj) = @_;
-	$self->_bidi_pipe(qw(--batch-check in_c out_c pid_c));
+	_bidi_pipe($self, qw(--batch-check in_c out_c pid_c err_c));
 	$self->{out_c}->print($obj, "\n") or fail($self, "write error: $!");
 	local $/ = "\n";
 	chomp(my $line = $self->{in_c}->getline);
 	my ($hex, $type, $size) = split(' ', $line);
 	return if $type eq 'missing';
+
+	# "dead" in git.git shows "dangling 4\ndead\n", not sure why
+	# https://public-inbox.org/git/20190118033845.s2vlrb3wd3m2jfzu@dcvr/
+	# so handle the oddball stuff just in case
+	if ($hex eq 'dangling' || $hex eq 'notdir' || $hex eq 'loop') {
+		$size = $type + length("\n");
+		my $r = read($self->{in_c}, my $buf, $size);
+		defined($r) or fail($self, "read failed: $!");
+		return;
+	}
+
 	($hex, $type, $size);
 }
 
-- 
EW