use PublicInbox::Qspawn;
use URI::Escape qw(uri_escape_utf8);
+# POSIX requires _POSIX_ARG_MAX >= 4096, and xargs is required to
+# subtract 2048 bytes. We also don't factor in environment variable
+# headroom into this.
+use POSIX qw(sysconf _SC_ARG_MAX);
+my $ARG_SIZE_MAX = (sysconf(_SC_ARG_MAX) || 4096) - 2048;
+
+# By default, "git format-patch" generates filenames with a four-digit
+# prefix, so that means 9999 patch series are OK, right? :>
+# Maybe we can make this configurable, main concern is disk space overhead
+# for uncompressed patch fragments. Aside from space, public-inbox-httpd
+# is otherwise unaffected by having many patches, here, as it can share
+# work fairly. Other PSGI servers may have trouble, though.
+my $MAX_PATCH = 9999;
+
# di = diff info / a hashref with information about a diff ($di):
# {
# oid_a => abbreviated pre-image oid,
scalar(@ambiguous) ? \@ambiguous : undef;
}
-sub extract_diff ($$$$) {
- my ($p, $re, $ibx, $smsg) = @_;
+sub extract_diff ($$$$$) {
+ my ($self, $p, $re, $ibx, $smsg) = @_;
my ($part) = @$p; # ignore $depth and @idx;
my $hdr_lines; # diff --git a/... b/...
my $tmp;
}
}
+
# start writing the diff out to a tempfile
- open($tmp, '+>', undef) or die "open(tmp): $!";
- $di->{tmp} = $tmp;
+ my $pn = ++$self->{tot};
+ open($tmp, '>', $self->{tmp}->dirname . "/$pn") or
+ die "open(tmp): $!";
push @$hdr_lines, $l;
$di->{hdr_lines} = $hdr_lines;
+ utf8::encode($_) for @$hdr_lines;
print $tmp @$hdr_lines or die "print(tmp): $!";
# for debugging/diagnostics:
$di->{ibx} = $ibx;
$di->{smsg} = $smsg;
} elsif ($l =~ m!\Adiff --git ("?a/.+) ("?b/.+)$!) {
- return $di if $tmp; # got our blob, done!
+ last if $tmp; # got our blob, done!
my ($path_a, $path_b) = ($1, $2);
$di->{path_b} = join('/', @b);
$hdr_lines = [ $l ];
} elsif ($tmp) {
+ utf8::encode($l);
print $tmp $l or die "print(tmp): $!";
} elsif ($hdr_lines) {
push @$hdr_lines, $l;
}
}
}
- $tmp ? $di : undef;
+ return undef unless $tmp;
+ close $tmp or die "close(tmp): $!";
+ $di;
}
sub path_searchable ($) { defined($_[0]) && $_[0] =~ m!\A[\w/\. \-]+\z! }
foreach my $smsg (@$msgs) {
$ibx->smsg_mime($smsg) or next;
msg_iter(delete($smsg->{mime}), sub {
- $di ||= extract_diff($_[0], $re, $ibx, $smsg);
+ $di ||= extract_diff($self, $_[0], $re, $ibx, $smsg);
});
return $di if $di;
}
my ($self) = @_;
my $patches = $self->{patches};
$self->{nr} = 0;
- $self->{tot} = scalar @$patches;
my $di = $patches->[0] or die 'no patches';
my $oid_a = $di->{oid_a} or die '{oid_a} unset';
sub do_git_apply ($) {
my ($self) = @_;
-
- my $di = shift @{$self->{patches}} or die 'empty {patches}';
- my $tmp = delete $di->{tmp} or die 'no tmp ', di_url($self, $di);
- $tmp->flush or die "tmp->flush failed: $!";
- sysseek($tmp, 0, SEEK_SET) or die "sysseek(tmp) failed: $!";
-
- my $i = ++$self->{nr};
- dbg($self, "\napplying [$i/$self->{tot}] " . di_url($self, $di) .
- "\n" . join('', @{$di->{hdr_lines}}));
+ my $dn = $self->{tmp}->dirname;
+ my $patches = $self->{patches};
# we need --ignore-whitespace because some patches are CRLF
- my $cmd = [ qw(git apply --cached --ignore-whitespace
- --whitespace=warn --verbose) ];
- my $rdr = { 0 => fileno($tmp), 2 => 1 };
- my $qsp = PublicInbox::Qspawn->new($cmd, $self->{git_env}, $rdr);
+ my @cmd = (qw(git -C), $dn, qw(apply --cached --ignore-whitespace
+ --whitespace=warn --verbose));
+ my $len = length(join(' ', @cmd));
+ my $total = $self->{tot};
+ my $di; # keep track of the last one for "git ls-files"
+
+ do {
+ my $i = ++$self->{nr};
+ $di = shift @$patches;
+ dbg($self, "\napplying [$i/$total] " . di_url($self, $di) .
+ "\n" . join('', @{$di->{hdr_lines}}));
+ my $path = $total + 1 - $i;
+ $len += length($path) + 1;
+ push @cmd, $path;
+ } while (@$patches && $len < $ARG_SIZE_MAX);
+
+ my $rdr = { 2 => 1 };
+ my $qsp = PublicInbox::Qspawn->new(\@cmd, $self->{git_env}, $rdr);
$qsp->psgi_qx($self->{psgi_env}, undef, sub {
my ($bref) = @_;
- close $tmp;
dbg($self, $$bref);
if (my $err = $qsp->{err}) {
ERR($self, "git apply error: $err");
sub resolve_patch ($$) {
my ($self, $want) = @_;
- if (scalar(@{$self->{patches}}) > $self->{max_patch}) {
+ if (scalar(@{$self->{patches}}) > $MAX_PATCH) {
die "Aborting, too many steps to $self->{oid_want}";
}
# see if we can find the blob in an existing git repo:
my $cur_want = $want->{oid_b};
+ if ($self->{seen_oid}->{$cur_want}++) {
+ die "Loop detected solving $cur_want\n";
+ }
if (my $existing = solve_existing($self, $want)) {
dbg($self, "found $cur_want in " .
join("\n", $existing->[0]->pub_urls));
bless {
gits => $ibx->{-repo_objs},
user_cb => $user_cb,
- max_patch => 100,
# TODO: config option for searching related inboxes
inboxes => [ $ibx ],
$self->{oid_want} = $oid_want;
$self->{out} = $out;
+ $self->{seen_oid} = {};
+ $self->{tot} = 0;
$self->{psgi_env} = $env;
$self->{todo} = [ { %$hints, oid_b => $oid_want } ];
$self->{patches} = []; # [ $di, $di, ... ]