treewide: "require" + "use" cleanup and docs

[public-inbox.git] / lib / PublicInbox / Git.pm

diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm
index 218846f3d1d25d42699a4a4be55351394c6a5341..6a527f82f9145ada72a1abbf053bf1987009283d 100644 (file)
--- a/lib/PublicInbox/Git.pm
+++ b/lib/PublicInbox/Git.pm
@@ -10,12 +10,17 @@ package PublicInbox::Git;
 use strict;
 use warnings;
 use POSIX qw(dup2);
-require IO::Handle;
+use IO::Handle; # ->autoflush
 use PublicInbox::Spawn qw(spawn popen_rd);
 use PublicInbox::Tmpfile;
 use base qw(Exporter);
 our @EXPORT_OK = qw(git_unquote git_quote);
 
+use constant MAX_INFLIGHT =>
+       ($^O eq 'linux' ? 4096 : POSIX::_POSIX_PIPE_BUF())
+       /
+       65; # SHA-256 hex size + "\n" in preparation for git using non-SHA1
+
 my %GIT_ESC = (
        a => "\a",
        b => "\b",
@@ -109,12 +114,12 @@ sub _bidi_pipe {
 
        my @cmd = (qw(git), "--git-dir=$self->{git_dir}",
                        qw(-c core.abbrev=40 cat-file), $batch);
-       my $redir = { 0 => fileno($out_r), 1 => fileno($in_w) };
+       my $redir = { 0 => $out_r, 1 => $in_w };
        if ($err) {
                my $id = "git.$self->{git_dir}$batch.err";
                my $fh = tmpfile($id) or fail($self, "tmpfile($id): $!");
                $self->{$err} = $fh;
-               $redir->{2} = fileno($fh);
+               $redir->{2} = $fh;
        }
        my $p = spawn(\@cmd, undef, $redir);
        defined $p or fail($self, "spawn failed: $!");
@@ -124,10 +129,52 @@ sub _bidi_pipe {
        $self->{$in} = $in_r;
 }
 
+sub read_cat_in_full ($$$) {
+       my ($self, $in, $left) = @_;
+       my $offset = 0;
+       my $buf = '';
+       while ($left > 0) {
+               my $r = read($in, $buf, $left, $offset);
+               defined($r) or fail($self, "read failed: $!");
+               $r == 0 and fail($self, 'exited unexpectedly');
+               $left -= $r;
+               $offset += $r;
+       }
+       my $r = read($in, my $lf, 1);
+       defined($r) or fail($self, "read failed: $!");
+       fail($self, 'newline missing after blob') if ($r != 1 || $lf ne "\n");
+       \$buf;
+}
+
+sub _cat_async_step ($$$) {
+       my ($self, $inflight, $in) = @_;
+       my $pair = shift @$inflight or die 'BUG: inflight empty';
+       my ($cb, $arg) = @$pair;
+       local $/ = "\n";
+       my $head = $in->getline;
+       $head =~ / missing$/ and return
+               eval { $cb->(undef, undef, undef, undef, $arg) };
+
+       $head =~ /^([0-9a-f]{40}) (\S+) ([0-9]+)$/ or
+               fail($self, "Unexpected result from async git cat-file: $head");
+       my ($oid_hex, $type, $size) = ($1, $2, $3 + 0);
+       my $bref = read_cat_in_full($self, $in, $size);
+       eval { $cb->($bref, $oid_hex, $type, $size, $arg) };
+}
+
+sub cat_async_wait ($) {
+       my ($self) = @_;
+       my $inflight = delete $self->{inflight} or return;
+       my $in = $self->{in};
+       while (scalar(@$inflight)) {
+               _cat_async_step($self, $inflight, $in);
+       }
+}
+
 sub cat_file {
        my ($self, $obj, $ref) = @_;
        my ($retried, $in, $head);
-
+       cat_async_wait($self);
 again:
        batch_prepare($self);
        $self->{out}->print($obj, "\n") or fail($self, "write error: $!");
@@ -147,26 +194,8 @@ again:
                fail($self, "Unexpected result from git cat-file: $head");
 
        my $size = $1;
-       my $rv;
-       my $left = $size;
        $$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 $lf, 1);
-       defined($r) or fail($self, "read failed: $!");
-       fail($self, 'newline missing after blob') if ($r != 1 || $lf ne "\n");
-
-       $rv;
+       read_cat_in_full($self, $in, $size);
 }
 
 sub batch_prepare ($) { _bidi_pipe($_[0], qw(--batch in out pid)) }
@@ -206,9 +235,15 @@ sub _destroy {
        waitpid($p, 0) if $@; # wait synchronously if not in event loop
 }
 
+sub cat_async_abort ($) {
+       my ($self) = @_;
+       my $inflight = delete $self->{inflight} or die 'BUG: not in async';
+       cleanup($self);
+}
+
 sub fail {
        my ($self, $msg) = @_;
-       cleanup($self);
+       $self->{inflight} ? cat_async_abort($self) : cleanup($self);
        die $msg;
 }
 
@@ -278,6 +313,25 @@ sub pub_urls {
        local_nick($self);
 }
 
+sub cat_async_begin {
+       my ($self) = @_;
+       cleanup($self) if alternates_changed($self);
+       batch_prepare($self);
+       die 'BUG: already in async' if $self->{inflight};
+       $self->{inflight} = [];
+}
+
+sub cat_async ($$$;$) {
+       my ($self, $oid, $cb, $arg) = @_;
+       my $inflight = $self->{inflight} or die 'BUG: not in async';
+       if (scalar(@$inflight) >= MAX_INFLIGHT) {
+               _cat_async_step($self, $inflight, $self->{in});
+       }
+
+       $self->{out}->print($oid, "\n") or fail($self, "write error: $!");
+       push(@$inflight, [ $cb, $arg ]);
+}
+
 sub commit_title ($$) {
        my ($self, $oid) = @_; # PublicInbox::Git, $sha1hex
        my $buf = cat_file($self, $oid) or return;