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;
+my $OID_MIN = 7;
+
+# 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 ($$$$) {
- my ($p, $re, $ibx, $smsg) = @_;
+sub extract_diff ($$$$$) {
+ my ($self, $p, $re, $ibx, $smsg) = @_;
my ($part) = @$p; # ignore $depth and @idx;
my $hdr_lines; # diff --git a/... b/...
my $tmp;
}
}
+
# start writing the diff out to a tempfile
- open($tmp, '+>', undef) or die "open(tmp): $!";
- $di->{tmp} = $tmp;
+ my $pn = ++$self->{tot};
+ open($tmp, '>', $self->{tmp}->dirname . "/$pn") or
+ die "open(tmp): $!";
push @$hdr_lines, $l;
$di->{hdr_lines} = $hdr_lines;
+ utf8::encode($_) for @$hdr_lines;
print $tmp @$hdr_lines or die "print(tmp): $!";
# for debugging/diagnostics:
$di->{ibx} = $ibx;
$di->{smsg} = $smsg;
- } elsif ($l =~ m!\Adiff --git ("?a/.+) ("?b/.+)$!) {
- return $di if $tmp; # got our blob, done!
+ } elsif ($l =~ m!\Adiff --git ("?[^/]+/.+) ("?[^/]+/.+)$!) {
+ last if $tmp; # got our blob, done!
my ($path_a, $path_b) = ($1, $2);
$di->{path_b} = join('/', @b);
$hdr_lines = [ $l ];
} elsif ($tmp) {
+ utf8::encode($l);
print $tmp $l or die "print(tmp): $!";
} elsif ($hdr_lines) {
push @$hdr_lines, $l;
}
}
}
- $tmp ? $di : undef;
+ return undef unless $tmp;
+ close $tmp or die "close(tmp): $!";
+ $di;
}
sub path_searchable ($) { defined($_[0]) && $_[0] =~ m!\A[\w/\. \-]+\z! }
+# ".." appears in path names, which confuses Xapian into treating
+# it as a range query. So we split on ".." since Xapian breaks
+# on punctuation anyways:
+sub filename_query ($) {
+ join('', map { qq( dfn:"$_") } split(/\.\./, $_[0]));
+}
+
sub find_extract_diff ($$$) {
my ($self, $ibx, $want) = @_;
my $srch = $ibx->search or return;
my $path_b = $want->{path_b};
if (path_searchable($path_b)) {
- $q .= qq{ dfn:"$path_b"};
+ $q .= filename_query($path_b);
my $path_a = $want->{path_a};
if (path_searchable($path_a) && $path_a ne $path_b) {
- $q .= qq{ dfn:"$path_a"};
+ $q .= filename_query($path_a);
}
}
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 = $di->{oid_a} or die '{oid_a} unset';
dbg($self, 'preparing index');
my $rdr = { 0 => fileno($in) };
- my $cmd = [ qw(git -C), $self->{wt_dir},
- qw(update-index -z --index-info) ];
- my $qsp = PublicInbox::Qspawn->new($cmd, undef, $rdr);
+ my $cmd = [ qw(git update-index -z --index-info) ];
+ my $qsp = PublicInbox::Qspawn->new($cmd, $self->{git_env}, $rdr);
$qsp->psgi_qx($self->{psgi_env}, undef, sub {
my ($bref) = @_;
if (my $err = $qsp->{err}) {
}
# pure Perl "git init"
-sub do_git_init_wt ($) {
+sub do_git_init ($) {
my ($self) = @_;
- my $wt = File::Temp->newdir('solver.wt-XXXXXXXX', TMPDIR => 1);
- my $dir = $self->{wt_dir} = $wt->dirname;
+ my $dir = $self->{tmp}->dirname;
+ my $git_dir = "$dir/git";
foreach ('', qw(objects refs objects/info refs/heads)) {
- mkdir("$dir/.git/$_") or die "mkdir $_: $!";
+ mkdir("$git_dir/$_") or die "mkdir $_: $!";
}
- open my $fh, '>', "$dir/.git/config" or die "open .git/config: $!";
- print $fh <<'EOF' or die "print .git/config $!";
+ open my $fh, '>', "$git_dir/config" or die "open git/config: $!";
+ print $fh <<'EOF' or die "print git/config $!";
[core]
repositoryFormatVersion = 0
filemode = true
fsyncObjectfiles = false
logAllRefUpdates = false
EOF
- close $fh or die "close .git/config: $!";
+ close $fh or die "close git/config: $!";
- open $fh, '>', "$dir/.git/HEAD" or die "open .git/HEAD: $!";
- print $fh "ref: refs/heads/master\n" or die "print .git/HEAD: $!";
- close $fh or die "close .git/HEAD: $!";
+ open $fh, '>', "$git_dir/HEAD" or die "open git/HEAD: $!";
+ print $fh "ref: refs/heads/master\n" or die "print git/HEAD: $!";
+ close $fh or die "close git/HEAD: $!";
- my $f = '.git/objects/info/alternates';
- open $fh, '>', "$dir/$f" or die "open: $f: $!";
- print($fh (map { "$_->{git_dir}/objects\n" } @{$self->{gits}})) or
- die "print $f: $!";
+ my $f = 'objects/info/alternates';
+ open $fh, '>', "$git_dir/$f" or die "open: $f: $!";
+ foreach my $git (@{$self->{gits}}) {
+ print $fh $git->git_path('objects'),"\n" or die "print $f: $!";
+ }
close $fh or die "close: $f: $!";
- my $wt_git = $self->{wt_git} = PublicInbox::Git->new("$dir/.git");
- $wt_git->{-wt} = $wt;
+ my $tmp_git = $self->{tmp_git} = PublicInbox::Git->new($git_dir);
+ $tmp_git->{-tmp} = $self->{tmp};
+ $self->{git_env} = {
+ GIT_DIR => $git_dir,
+ GIT_INDEX_FILE => "$git_dir/index",
+ };
prepare_index($self);
}
sub extract_old_mode ($) {
my ($di) = @_;
- if (grep(/\Aold mode (100644|100755|120000)$/, @{$di->{hdr_lines}})) {
+ if (join('', @{$di->{hdr_lines}}) =~
+ /^old mode (100644|100755|120000)\b/) {
return $1;
}
'100644';
}
+sub do_finish ($$) {
+ my ($self, $user_cb) = @_;
+ my $found = $self->{found};
+ my $oid_want = $self->{oid_want};
+ if (my $exists = $found->{$oid_want}) {
+ return $user_cb->($exists);
+ }
+
+ # let git disambiguate if oid_want was too short,
+ # but long enough to be unambiguous:
+ my $tmp_git = $self->{tmp_git};
+ if (my @res = $tmp_git->check($oid_want)) {
+ return $user_cb->($found->{$res[0]});
+ }
+ if (my $err = $tmp_git->last_check_err) {
+ dbg($self, $err);
+ }
+ $user_cb->(undef);
+}
+
sub do_step ($) {
my ($self) = @_;
eval {
# step 2: then we instantiate a working tree once
# the todo queue is finally empty:
- } elsif (!defined($self->{wt_git})) {
- do_git_init_wt($self);
+ } elsif (!defined($self->{tmp_git})) {
+ do_git_init($self);
# step 3: apply each patch in the stack
} elsif (scalar @{$self->{patches}}) {
# our result: (which may be undef)
# Other steps may call user_cb to terminate prematurely
# on error
- } elsif (my $ucb = delete($self->{user_cb})) {
- $ucb->($self->{found}->{$self->{oid_want}});
+ } elsif (my $user_cb = delete($self->{user_cb})) {
+ do_finish($self, $user_cb);
} else {
die 'about to call user_cb twice'; # Oops :x
}
sub mark_found ($$$) {
my ($self, $oid, $found_info) = @_;
- $self->{found}->{$oid} = $found_info;
+ my $found = $self->{found};
+ $found->{$oid} = $found_info;
+ my $oid_cur = $found_info->[1];
+ while ($oid_cur ne $oid && length($oid_cur) > $OID_MIN) {
+ $found->{$oid_cur} = $found_info;
+ chop($oid_cur);
+ }
}
sub parse_ls_files ($$$$) {
"BUG: index mismatch: file=$file != path_b=$di->{path_b}";
}
- my $wt_git = $self->{wt_git} or die 'no git working tree';
- my (undef, undef, $size) = $wt_git->check($oid_b_full);
+ my $tmp_git = $self->{tmp_git} or die 'no git working tree';
+ my (undef, undef, $size) = $tmp_git->check($oid_b_full);
defined($size) or die "check $oid_b_full failed";
dbg($self, "index at:\n$mode_b $oid_b_full\t$file");
- my $created = [ $wt_git, $oid_b_full, 'blob', $size, $di ];
+ my $created = [ $tmp_git, $oid_b_full, 'blob', $size, $di ];
mark_found($self, $di->{oid_b}, $created);
next_step($self); # onto the next patch
}
sub start_ls_files ($$) {
my ($self, $di) = @_;
- my $cmd = [qw(git -C), $self->{wt_dir}, qw(ls-files -s -z)];
- my $qsp = PublicInbox::Qspawn->new($cmd);
+ my $cmd = [qw(git ls-files -s -z)];
+ my $qsp = PublicInbox::Qspawn->new($cmd, $self->{git_env});
$qsp->psgi_qx($self->{psgi_env}, undef, sub {
my ($bref) = @_;
eval { parse_ls_files($self, $qsp, $bref, $di) };
sub do_git_apply ($) {
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 -C), $self->{wt_dir},
- qw(apply --cached --ignore-whitespace
- --whitespace=warn --verbose) ];
- my $rdr = { 0 => fileno($tmp), 2 => 1 };
- my $qsp = PublicInbox::Qspawn->new($cmd, undef, $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", $existing->[0]->pub_urls));
}
return next_step($self); # onto the next todo item
}
+ if (length($cur_want) > $OID_MIN) {
+ chop($cur_want);
+ dbg($self, "retrying $want->{oid_b} as $cur_want");
+ $want->{oid_b} = $cur_want;
+ push @{$self->{todo}}, $want;
+ return next_step($self); # retry with shorter abbrev
+ }
+
dbg($self, "could not find $cur_want");
eval { delete($self->{user_cb})->(undef) }; # not found! :<
die "E: $@" if $@;
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, ... ]
$self->{found} = {}; # { abbr => [ ::Git, oid, type, size, $di ] }
+ $self->{tmp} = File::Temp->newdir('solver.tmp-XXXXXXXX', TMPDIR => 1);
dbg($self, "solving $oid_want ...");
my $step_cb = step_cb($self);