user/dev discussion of public-inbox itself
 help / color / Atom feed
* [PATCH] git: remove cat_file sub callback interface
@ 2019-06-13  8:10 Eric Wong
  0 siblings, 0 replies; only message in thread
From: Eric Wong @ 2019-06-13  8:10 UTC (permalink / raw)
  To: meta

We weren't using it, and in retrospect, it makes no sense to use
this API cat_file for giant responses which can't read quickly
with minimal context-switching (or sanely fit into memory for
Email::Simple/Email::MIME).

For giant blobs which we don't want slurped in memory, we'll
spawn a short-lived git-cat-file process like we do in ViewVCS.

Otherwise, monopolizing a git-cat-file process for a giant
blob is harmful to other PSGI/NNTP users.

A better interface is coming which will be more suitable for
for batch processing of "small" objects such as commits and
email blobs.
---
 lib/PublicInbox/Git.pm | 43 ++++++++++---------------------
 t/git.t                | 58 ------------------------------------------
 2 files changed, 13 insertions(+), 88 deletions(-)

diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm
index 68445b3..6a87661 100644
--- a/lib/PublicInbox/Git.pm
+++ b/lib/PublicInbox/Git.pm
@@ -145,41 +145,24 @@ again:
 		fail($self, "Unexpected result from git cat-file: $head");
 
 	my $size = $1;
-	my $ref_type = $ref ? ref($ref) : '';
-
 	my $rv;
 	my $left = $size;
-	$$ref = $size if ($ref_type eq 'SCALAR');
-	my $cb_err;
-
-	if ($ref_type eq 'CODE') {
-		$rv = eval { $ref->($in, \$left) };
-		$cb_err = $@;
-		# drain the rest
-		my $max = 8192;
-		while ($left > 0) {
-			my $r = read($in, my $x, $left > $max ? $max : $left);
-			defined($r) or fail($self, "read failed: $!");
-			$r == 0 and fail($self, 'exited unexpectedly');
-			$left -= $r;
-		}
-	} else {
-		my $offset = 0;
-		my $buf = '';
-		while ($left > 0) {
-			my $r = read($in, $buf, $left, $offset);
-			defined($r) or fail($self, "read failed: $!");
-			$r == 0 and fail($self, 'exited unexpectedly');
-			$left -= $r;
-			$offset += $r;
-		}
-		$rv = \$buf;
+	$$ref = $size if $ref;
+
+	my $offset = 0;
+	my $buf = '';
+	while ($left > 0) {
+		my $r = read($in, $buf, $left, $offset);
+		defined($r) or fail($self, "read failed: $!");
+		$r == 0 and fail($self, 'exited unexpectedly');
+		$left -= $r;
+		$offset += $r;
 	}
+	$rv = \$buf;
 
-	my $r = read($in, my $buf, 1);
+	my $r = read($in, my $lf, 1);
 	defined($r) or fail($self, "read failed: $!");
-	fail($self, 'newline missing after blob') if ($r != 1 || $buf ne "\n");
-	die $cb_err if $cb_err;
+	fail($self, 'newline missing after blob') if ($r != 1 || $lf ne "\n");
 
 	$rv;
 }
diff --git a/t/git.t b/t/git.t
index 913f6e5..9bc8900 100644
--- a/t/git.t
+++ b/t/git.t
@@ -33,33 +33,7 @@ use_ok 'PublicInbox::Git';
 	my $raw = $gcf->cat_file($f);
 	is($x[2], length($$raw), 'length matches');
 
-	{
-		my $size;
-		my $rv = $gcf->cat_file($f, sub {
-			my ($in, $left) = @_;
-			$size = $$left;
-			'nothing'
-		});
-		is($rv, 'nothing', 'returned from callback without reading');
-		is($size, $x[2], 'set size for callback correctly');
-	}
-
-	eval { $gcf->cat_file($f, sub { die 'OMG' }) };
-	like($@, qr/\bOMG\b/, 'died in callback propagated');
 	is(${$gcf->cat_file($f)}, $$raw, 'not broken after failures');
-
-	{
-		my ($buf, $r);
-		my $rv = $gcf->cat_file($f, sub {
-			my ($in, $left) = @_;
-			$r = read($in, $buf, 2);
-			$$left -= $r;
-			'blah'
-		});
-		is($r, 2, 'only read 2 bytes');
-		is($buf, '--', 'partial read succeeded');
-		is($rv, 'blah', 'return value propagated');
-	}
 	is(${$gcf->cat_file($f)}, $$raw, 'not broken after partial read');
 }
 
@@ -79,44 +53,12 @@ if (1) {
 
 	my $gcf = PublicInbox::Git->new($dir);
 	my $rsize;
-	is($gcf->cat_file($buf, sub {
-		$rsize = ${$_[1]};
-		'x';
-	}), 'x', 'checked input');
-	is($rsize, $size, 'got correct size on big file');
-
 	my $x = $gcf->cat_file($buf, \$rsize);
 	is($rsize, $size, 'got correct size ref on big file');
 	is(length($$x), $size, 'read correct number of bytes');
 
-	my $rline;
-	$gcf->cat_file($buf, sub {
-		my ($in, $left) = @_;
-		$rline = <$in>;
-		$$left -= length($rline);
-	});
-	{
-		open my $fh, '<', $big_data or die "open failed: $!\n";
-		is($rline, <$fh>, 'first line matches');
-	};
-
-	my $all;
-	$gcf->cat_file($buf, sub {
-		my ($in, $left) = @_;
-		my $x = read($in, $all, $$left);
-		$$left -= $x;
-	});
-	{
-		open my $fh, '<', $big_data or die "open failed: $!\n";
-		local $/;
-		is($all, <$fh>, 'entire read matches');
-	};
-
 	my $ref = $gcf->qx(qw(cat-file blob), $buf);
-	is($all, $ref, 'qx read giant single string');
-
 	my @ref = $gcf->qx(qw(cat-file blob), $buf);
-	is($all, join('', @ref), 'qx returned array when wanted');
 	my $nl = scalar @ref;
 	ok($nl > 1, "qx returned array length of $nl");
 
-- 
EW


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, back to index

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-06-13  8:10 [PATCH] git: remove cat_file sub callback interface Eric Wong

user/dev discussion of public-inbox itself

Archives are clonable:
	git clone --mirror http://public-inbox.org/meta
	git clone --mirror http://czquwvybam4bgbro.onion/meta
	git clone --mirror http://hjrcffqmbrq6wope.onion/meta
	git clone --mirror http://ou63pmih66umazou.onion/meta

Newsgroups are available over NNTP:
	nntp://news.public-inbox.org/inbox.comp.mail.public-inbox.meta
	nntp://ou63pmih66umazou.onion/inbox.comp.mail.public-inbox.meta
	nntp://czquwvybam4bgbro.onion/inbox.comp.mail.public-inbox.meta
	nntp://hjrcffqmbrq6wope.onion/inbox.comp.mail.public-inbox.meta
	nntp://news.gmane.org/gmane.mail.public-inbox.general

 note: .onion URLs require Tor: https://www.torproject.org/

AGPL code for this site: git clone https://public-inbox.org/ public-inbox