X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FGit.pm;h=b2ae75c8ae312042562c35dc568047bc8af49a0e;hb=23af251dd607c4e75ab1e68063f2c885c48cc035;hp=77783000573f20692be001042d256ed20597cfce;hpb=770604494e1054ada8d9e3960e4a6c7e82a09ad5;p=public-inbox.git diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm index 77783000..b2ae75c8 100644 --- a/lib/PublicInbox/Git.pm +++ b/lib/PublicInbox/Git.pm @@ -1,4 +1,4 @@ -# Copyright (C) 2014-2021 all contributors +# Copyright (C) all contributors # License: GPLv2 or later # # Used to read files from a git repository without excessive forking. @@ -19,13 +19,14 @@ use Time::HiRes qw(stat); use PublicInbox::Spawn qw(popen_rd spawn); use PublicInbox::Tmpfile; use IO::Poll qw(POLLIN); -use Carp qw(croak); +use Carp qw(croak carp); use Digest::SHA (); use PublicInbox::DS qw(dwaitpid); our @EXPORT_OK = qw(git_unquote git_quote); our $PIPE_BUFSIZ = 65536; # Linux default our $in_cleanup; our $RDTIMEO = 60_000; # milliseconds +our $async_warn; # true in read-only daemons use constant MAX_INFLIGHT => (POSIX::PIPE_BUF * 3) / 65; # SHA-256 hex size + "\n" in preparation for git using non-SHA1 @@ -70,7 +71,7 @@ sub new { sub git_path ($$) { my ($self, $path) = @_; - $self->{-git_path}->{$path} ||= do { + $self->{-git_path}->{$path} //= do { local $/ = "\n"; chomp(my $str = $self->qx(qw(rev-parse --git-path), $path)); @@ -93,6 +94,13 @@ sub alternates_changed { $self->{alt_st} = $st; # always a true value } +sub object_format { + $_[0]->{object_format} //= do { + my $fmt = $_[0]->qx(qw(config extensions.objectformat)); + $fmt eq "sha256\n" ? \'sha256' : \undef; + } +} + sub last_check_err { my ($self) = @_; my $fh = $self->{err_c} or return; @@ -113,7 +121,7 @@ sub _bidi_pipe { return; } pipe(my ($out_r, $out_w)) or $self->fail("pipe failed: $!"); - my $rdr = { 0 => $out_r }; + my $rdr = { 0 => $out_r, pgid => 0 }; my $gd = $self->{git_dir}; if ($gd =~ s!/([^/]+/[^/]+)\z!/!) { $rdr->{-C} = $gd; @@ -157,7 +165,8 @@ sub my_read ($$$) { return; # unrecoverable error } } - \substr($$rbuf, 0, $len, ''); + my $no_pad = substr($$rbuf, 0, $len, ''); + \$no_pad; } sub my_readline ($$) { @@ -178,8 +187,8 @@ sub my_readline ($$) { } } -sub cat_async_retry ($$$$$) { - my ($self, $inflight, $req, $cb, $arg) = @_; +sub cat_async_retry ($$) { + my ($self, $inflight) = @_; # {inflight} may be non-existent, but if it isn't we delete it # here to prevent cleanup() from waiting: @@ -188,12 +197,13 @@ sub cat_async_retry ($$$$$) { $self->{inflight} = $inflight; batch_prepare($self); - my $buf = "$req\n"; + my $buf = ''; for (my $i = 0; $i < @$inflight; $i += 3) { $buf .= "$inflight->[$i]\n"; } print { $self->{out} } $buf or $self->fail("write error: $!"); - unshift(@$inflight, \$req, $cb, $arg); # \$ref to indicate retried + my $req = shift @$inflight; + unshift(@$inflight, \$req); # \$ref to indicate retried cat_async_step($self, $inflight); # take one step } @@ -201,8 +211,8 @@ sub cat_async_retry ($$$$$) { sub cat_async_step ($$) { my ($self, $inflight) = @_; die 'BUG: inflight empty or odd' if scalar(@$inflight) < 3; - my ($req, $cb, $arg) = splice(@$inflight, 0, 3); - my $rbuf = delete($self->{cat_rbuf}) // \(my $new = ''); + my ($req, $cb, $arg) = @$inflight[0, 1, 2]; + my $rbuf = delete($self->{rbuf}) // \(my $new = ''); my ($bref, $oid, $type, $size); my $head = my_readline($self->{in}, $rbuf); # ->fail may be called via Gcf2Client.pm @@ -216,8 +226,7 @@ sub cat_async_step ($$) { # ref($req) indicates it's already been retried # -gcf2 retries internally, so it never hits this path: if (!ref($req) && !$in_cleanup && $self->alternates_changed) { - return cat_async_retry($self, $inflight, - $req, $cb, $arg); + return cat_async_retry($self, $inflight); } $type = 'missing'; $oid = ref($req) ? $$req : $req if $oid eq ''; @@ -225,9 +234,10 @@ sub cat_async_step ($$) { my $err = $! ? " ($!)" : ''; $self->fail("bad result from async cat-file: $head$err"); } - $self->{cat_rbuf} = $rbuf if $$rbuf ne ''; + $self->{rbuf} = $rbuf if $$rbuf ne ''; + splice(@$inflight, 0, 3); # don't retry $cb on ->fail eval { $cb->($bref, $oid, $type, $size, $arg) }; - warn "E: $oid: $@\n" if $@; + async_err($self, $req, $oid, $@, 'cat') if $@; } sub cat_async_wait ($) { @@ -258,8 +268,8 @@ sub cat_file { sub check_async_step ($$) { my ($self, $inflight_c) = @_; die 'BUG: inflight empty or odd' if scalar(@$inflight_c) < 3; - my ($req, $cb, $arg) = splice(@$inflight_c, 0, 3); - my $rbuf = delete($self->{chk_rbuf}) // \(my $new = ''); + my ($req, $cb, $arg) = @$inflight_c[0, 1, 2]; + my $rbuf = delete($self->{rbuf_c}) // \(my $new = ''); chomp(my $line = my_readline($self->{in_c}, $rbuf)); my ($hex, $type, $size) = split(/ /, $line); @@ -271,9 +281,10 @@ sub check_async_step ($$) { my $ret = my_read($self->{in_c}, $rbuf, $type + 1); $self->fail(defined($ret) ? 'read EOF' : "read: $!") if !$ret; } - $self->{chk_rbuf} = $rbuf if $$rbuf ne ''; + $self->{rbuf_c} = $rbuf if $$rbuf ne ''; + splice(@$inflight_c, 0, 3); # don't retry $cb on ->fail eval { $cb->($hex, $type, $size, $arg, $self) }; - warn "E: check($req) $@\n" if $@; + async_err($self, $req, $hex, $@, 'check') if $@; } sub check_async_wait ($) { @@ -333,27 +344,39 @@ sub _destroy { dwaitpid($p) if $$ == $self->{"$pid.owner"}; } -sub cat_async_abort ($) { +sub async_abort ($) { my ($self) = @_; - if (my $inflight = $self->{inflight}) { - while (@$inflight) { - my ($req, $cb, $arg) = splice(@$inflight, 0, 3); - $req =~ s/ .*//; # drop git_dir for Gcf2Client - eval { $cb->(undef, $req, undef, undef, $arg) }; - warn "E: $req: $@ (in abort)\n" if $@; + while (scalar(@{$self->{inflight_c} // []}) || + scalar(@{$self->{inflight} // []})) { + for my $c ('', '_c') { + my $q = $self->{"inflight$c"} or next; + while (@$q) { + my ($req, $cb, $arg) = splice(@$q, 0, 3); + $req = $$req if ref($req); + $req =~ s/ .*//; # drop git_dir for Gcf2Client + eval { $cb->(undef, $req, undef, undef, $arg) }; + warn "E: (in abort) $req: $@" if $@; + } + delete $self->{"inflight$c"}; + delete $self->{"rbuf$c"}; } - delete $self->{cat_rbuf}; - delete $self->{inflight}; } cleanup($self); } sub fail { # may be augmented in subclasses my ($self, $msg) = @_; - cat_async_abort($self); + async_abort($self); croak(ref($self) . ' ' . ($self->{git_dir} // '') . ": $msg"); } +sub async_err ($$$$$) { + my ($self, $req, $oid, $err, $action) = @_; + $req = $$req if ref($req); # retried + my $msg = "E: $action $req ($oid): $err"; + $async_warn ? carp($msg) : $self->fail($msg); +} + # $git->popen(qw(show f00)); # or # $git->popen(qw(show f00), { GIT_CONFIG => ... }, { 2 => ... }); sub popen { @@ -391,8 +414,8 @@ sub async_wait_all ($) { my ($self) = @_; while (scalar(@{$self->{inflight_c} // []}) || scalar(@{$self->{inflight} // []})) { - $self->check_async_wait; - $self->cat_async_wait; + check_async_wait($self); + cat_async_wait($self); } } @@ -406,8 +429,8 @@ sub cleanup { async_wait_all($self); delete $self->{inflight}; delete $self->{inflight_c}; - _destroy($self, qw(cat_rbuf in out pid)); - _destroy($self, qw(chk_rbuf in_c out_c pid_c err_c)); + _destroy($self, qw(rbuf in out pid)); + _destroy($self, qw(rbuf_c in_c out_c pid_c err_c)); undef; } @@ -427,13 +450,8 @@ sub packed_bytes { sub DESTROY { cleanup(@_) } sub local_nick ($) { - my ($self) = @_; - my $ret = '???'; # don't show full FS path, basename should be OK: - if ($self->{git_dir} =~ m!/([^/]+)(?:/*\.git/*)?\z!) { - $ret = "$1.git"; - } - wantarray ? ($ret) : $ret; + $_[0]->{git_dir} =~ m!/([^/]+?)(?:/*\.git/*)?\z! ? "$1.git" : '???'; } sub host_prefix_url ($$) { @@ -450,7 +468,7 @@ sub pub_urls { if (my $urls = $self->{cgit_url}) { return map { host_prefix_url($env, $_) } @$urls; } - local_nick($self); + (local_nick($self)); } sub cat_async_begin {