X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FQspawn.pm;h=5e4fd5cb8541a2a8202d188ae1fd73c88fe64188;hb=HEAD;hp=6cb28b9ac061622b5223749474269a26e7cf4a90;hpb=dd57a7d007bf756d856fe3d2b414657ebf009941;p=public-inbox.git

diff --git a/lib/PublicInbox/Qspawn.pm b/lib/PublicInbox/Qspawn.pm
index 6cb28b9a..5e4fd5cb 100644
--- a/lib/PublicInbox/Qspawn.pm
+++ b/lib/PublicInbox/Qspawn.pm
@@ -1,4 +1,4 @@
-# Copyright (C) 2016-2019 all contributors <meta@public-inbox.org>
+# Copyright (C) all contributors <meta@public-inbox.org>
 # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt>
 
 # Like most Perl modules in public-inbox, this is internal and
@@ -12,22 +12,23 @@
 # operate in.  This can be useful to ensure smaller inboxes can
 # be cloned while cloning of large inboxes is maxed out.
 #
-# This does not depend on PublicInbox::DS or any other external
-# scheduling mechanism, you just need to call start() and finish()
-# appropriately. However, public-inbox-httpd (which uses PublicInbox::DS)
-# will be able to schedule this based on readability of stdout from
-# the spawned process.  See GitHTTPBackend.pm and SolverGit.pm for
-# usage examples.  It does not depend on any form of threading.
+# This does not depend on the PublicInbox::DS::event_loop or any
+# other external scheduling mechanism, you just need to call
+# start() and finish() appropriately. However, public-inbox-httpd
+# (which uses PublicInbox::DS)  will be able to schedule this
+# based on readability of stdout from the spawned process.
+# See GitHTTPBackend.pm and SolverGit.pm for usage examples.
+# It does not depend on any form of threading.
 #
 # This is useful for scheduling CGI execution of both long-lived
 # git-http-backend(1) process (for "git clone") as well as short-lived
 # processes such as git-apply(1).
 
 package PublicInbox::Qspawn;
-use strict;
-use warnings;
+use v5.12;
 use PublicInbox::Spawn qw(popen_rd);
-require Plack::Util;
+use PublicInbox::GzipFilter;
+use Scalar::Util qw(blessed);
 
 # n.b.: we get EAGAIN with public-inbox-httpd, and EINTR on other PSGI servers
 use Errno qw(EAGAIN EINTR);
@@ -38,7 +39,8 @@ my $def_limiter;
 # $cmd is the command to spawn
 # $cmd_env is the environ for the child process (not PSGI env)
 # $opt can include redirects and perhaps other process spawning options
-sub new ($$$;) {
+# {qsp_err} is an optional error buffer callers may access themselves
+sub new {
 	my ($class, $cmd, $cmd_env, $opt) = @_;
 	bless { args => [ $cmd, $cmd_env, $opt ] }, $class;
 }
@@ -46,59 +48,31 @@ sub new ($$$;) {
 sub _do_spawn {
 	my ($self, $start_cb, $limiter) = @_;
 	my $err;
-	my ($cmd, $cmd_env, $opts) = @{$self->{args}};
-	my %opts = %{$opts || {}};
+	my ($cmd, $cmd_env, $opt) = @{delete $self->{args}};
+	my %o = %{$opt || {}};
 	$self->{limiter} = $limiter;
-	foreach my $k (PublicInbox::Spawn::RLIMITS()) {
+	foreach my $k (@PublicInbox::Spawn::RLIMITS) {
 		if (defined(my $rlimit = $limiter->{$k})) {
-			$opts{$k} = $rlimit;
+			$o{$k} = $rlimit;
 		}
 	}
-
-	($self->{rpipe}, $self->{pid}) = popen_rd($cmd, $cmd_env, \%opts);
-
-	# drop any IO handles opt was holding open via $opt->{hold}
-	# No need to hold onto the descriptor once the child process has it.
-	$self->{args} = $cmd; # keep this around for logging
-
-	if (defined $self->{pid}) {
+	$self->{cmd} = $cmd;
+	$self->{-quiet} = 1 if $o{quiet};
+	$o{cb_arg} = [ \&waitpid_err, $self ];
+	eval {
+		# popen_rd may die on EMFILE, ENFILE
+		$self->{rpipe} = popen_rd($cmd, $cmd_env, \%o) // die "E: $!";
 		$limiter->{running}++;
-	} else {
-		$self->{err} = $!;
-	}
-	$start_cb->($self->{rpipe});
-}
-
-sub child_err ($) {
-	my ($child_error) = @_; # typically $?
-	my $exitstatus = ($child_error >> 8) or return;
-	my $sig = $child_error & 127;
-	my $msg = "exit status=$exitstatus";
-	$msg .= " signal=$sig" if $sig;
-	$msg;
-}
-
-sub log_err ($$) {
-	my ($env, $msg) = @_;
-	$env->{'psgi.errors'}->print($msg, "\n");
+		$start_cb->($self); # EPOLL_CTL_ADD may ENOSPC/ENOMEM
+	};
+	finish($self, $@) if $@;
 }
 
-# callback for dwaitpid
-sub waitpid_err ($$) {
-	my ($self, $pid) = @_;
-	my $xpid = delete $self->{pid};
-	my $err;
-	if ($pid > 0) { # success!
-		$err = child_err($?);
-	} elsif ($pid < 0) { # ??? does this happen in our case?
-		$err = "W: waitpid($xpid, 0) => $pid: $!";
-	} # else should not be called with pid == 0
-
-	my ($env, $qx_cb, $qx_arg, $qx_buf) =
-		delete @$self{qw(psgi_env qx_cb qx_arg qx_buf)};
+sub finalize ($) {
+	my ($self) = @_;
 
-	# done, spawn whatever's in the queue
-	my $limiter = $self->{limiter};
+	# process is done, spawn whatever's in the queue
+	my $limiter = delete $self->{limiter} or return;
 	my $running = --$limiter->{running};
 
 	if ($running < $limiter->{max}) {
@@ -106,41 +80,57 @@ sub waitpid_err ($$) {
 			_do_spawn(@$next, $limiter);
 		}
 	}
-
-	if ($err) {
-		$self->{err} = $err;
-		if ($env && !$env->{'qspawn.quiet'}) {
-			log_err($env, join(' ', @{$self->{args}}) . ": $err");
+	if (my $err = $self->{_err}) { # set by finish or waitpid_err
+		utf8::decode($err);
+		if (my $dst = $self->{qsp_err}) {
+			$$dst .= $$dst ? " $err" : "; $err";
 		}
+		warn "@{$self->{cmd}}: $err\n" if !$self->{-quiet};
 	}
-	eval { $qx_cb->($qx_buf, $qx_arg) } if $qx_cb;
-}
 
-sub do_waitpid ($) {
-	my ($self) = @_;
-	my $pid = $self->{pid};
-	# PublicInbox::DS may not be loaded
-	eval { PublicInbox::DS::dwaitpid($pid, \&waitpid_err, $self) };
-	# done if we're running in PublicInbox::DS::EventLoop
-	if ($@) {
-		# non public-inbox-{httpd,nntpd} callers may block:
-		my $ret = waitpid($pid, 0);
-		waitpid_err($self, $ret);
+	my ($env, $qx_cb, $qx_arg, $qx_buf) =
+		delete @$self{qw(psgi_env qx_cb qx_arg qx_buf)};
+	if ($qx_cb) {
+		eval { $qx_cb->($qx_buf, $qx_arg) };
+		return unless $@;
+		warn "E: $@"; # hope qspawn.wcb can handle it
+	}
+	return if $self->{passed}; # another command chained it
+	if (my $wcb = delete $env->{'qspawn.wcb'}) {
+		# have we started writing, yet?
+		my $code = delete $env->{'qspawn.fallback'} // 500;
+		require PublicInbox::WwwStatic;
+		$wcb->(PublicInbox::WwwStatic::r($code));
 	}
 }
 
-sub finish ($) {
-	my ($self) = @_;
-	if (delete $self->{rpipe}) {
-		do_waitpid($self);
-	} else {
-		my ($env, $qx_cb, $qx_arg, $qx_buf) =
-			delete @$self{qw(psgi_env qx_cb qx_arg qx_buf)};
-		eval { $qx_cb->($qx_buf, $qx_arg) } if $qx_cb;
+sub DESTROY { finalize($_[0]) } # ->finalize is idempotent
+
+sub waitpid_err { # callback for awaitpid
+	my (undef, $self) = @_; # $_[0]: pid
+	$self->{_err} = ''; # for defined check in ->finish
+	if ($?) {
+		my $status = $? >> 8;
+		my $sig = $? & 127;
+		$self->{_err} .= "exit status=$status";
+		$self->{_err} .= " signal=$sig" if $sig;
 	}
+	finalize($self) if !$self->{rpipe};
 }
 
-sub start {
+sub finish ($;$) {
+	my ($self, $err) = @_;
+	$self->{_err} //= $err; # only for $@
+
+	# we can safely finalize if pipe was closed before, or if
+	# {_err} is defined by waitpid_err.  Deleting {rpipe} will
+	# trigger PublicInbox::ProcessPipe::DESTROY -> waitpid_err,
+	# but it may not fire right away if inside the event loop.
+	my $closed_before = !delete($self->{rpipe});
+	finalize($self) if $closed_before || defined($self->{_err});
+}
+
+sub start ($$$) {
 	my ($self, $limiter, $start_cb) = @_;
 	if ($limiter->{running} < $limiter->{max}) {
 		_do_spawn($self, $start_cb, $limiter);
@@ -149,6 +139,37 @@ sub start {
 	}
 }
 
+sub psgi_qx_init_cb { # this may be PublicInbox::HTTPD::Async {cb}
+	my ($self) = @_;
+	my ($r, $buf);
+reread:
+	$r = sysread($self->{rpipe}, $buf, 65536);
+	if (!defined($r)) {
+		return if $! == EAGAIN; # try again when notified
+		goto reread if $! == EINTR;
+		event_step($self, $!);
+	} elsif (my $as = delete $self->{async}) { # PublicInbox::HTTPD::Async
+		$as->async_pass($self->{psgi_env}->{'psgix.io'},
+				$self->{qx_fh}, \$buf);
+	} elsif ($r) { # generic PSGI:
+		print { $self->{qx_fh} } $buf;
+	} else { # EOF
+		event_step($self, undef);
+	}
+}
+
+sub psgi_qx_start {
+	my ($self) = @_;
+	if (my $async = $self->{psgi_env}->{'pi-httpd.async'}) {
+		# PublicInbox::HTTPD::Async->new(rpipe, $cb, cb_arg, $end_obj)
+		$self->{async} = $async->($self->{rpipe},
+					\&psgi_qx_init_cb, $self, $self);
+		# init_cb will call ->async_pass or ->close
+	} else { # generic PSGI
+		psgi_qx_init_cb($self) while $self->{qx_fh};
+	}
+}
+
 # Similar to `backtick` or "qx" ("perldoc -f qx"), it calls $qx_cb with
 # the stdout of the given command when done; but respects the given limiter
 # $env is the PSGI env.  As with ``/qx; only use this when output is small
@@ -162,46 +183,8 @@ sub psgi_qx {
 	$self->{qx_arg} = $qx_arg;
 	$self->{qx_fh} = $qx_fh;
 	$self->{qx_buf} = \$qx_buf;
-	my $rpipe; # comes from popen_rd
-	my $async = $env->{'pi-httpd.async'};
-	my $cb = sub {
-		my ($r, $buf);
-reread:
-		$r = sysread($rpipe, $buf, 65536);
-		if ($async) {
-			$async->async_pass($env->{'psgix.io'}, $qx_fh, \$buf);
-		} elsif (defined $r) {
-			$r ? $qx_fh->write($buf) : event_step($self, undef);
-		} else {
-			return if $! == EAGAIN; # try again when notified
-			goto reread if $! == EINTR;
-			event_step($self, $!);
-		}
-	};
 	$limiter ||= $def_limiter ||= PublicInbox::Qspawn::Limiter->new(32);
-	$self->start($limiter, sub { # start_cb, may run later, much later...
-		($rpipe) = @_; # popen_rd result
-		if ($async) {
-		# PublicInbox::HTTPD::Async->new($rpipe, $cb, $end_obj)
-			$async = $async->($rpipe, $cb, undef, $self);
-			# $cb will call ->async_pass or ->close
-		} else { # generic PSGI
-			$cb->() while $self->{qx_fh};
-		}
-	});
-}
-
-# create a filter for "push"-based streaming PSGI writes used by HTTPD::Async
-sub filter_fh ($$) {
-	my ($fh, $filter) = @_;
-	Plack::Util::inline_object(
-		close => sub {
-			$fh->write($filter->(undef));
-			$fh->close;
-		},
-		write => sub {
-			$fh->write($filter->($_[0]));
-		});
+	start($self, $limiter, \&psgi_qx_start);
 }
 
 # this is called on pipe EOF to reap the process, may be called
@@ -209,12 +192,103 @@ sub filter_fh ($$) {
 # PSGI servers.
 sub event_step {
 	my ($self, $err) = @_; # $err: $!
-	log_err($self->{psgi_env}, "psgi_{return,qx} $err") if defined($err);
+	warn "psgi_{return,qx} $err" if defined($err);
 	finish($self);
-	my ($fh, $qx_fh) = delete(@$self{qw(fh qx_fh)});
+	my ($fh, $qx_fh) = delete(@$self{qw(qfh qx_fh)});
 	$fh->close if $fh; # async-only (psgi_return)
 }
 
+sub rd_hdr ($) {
+	my ($self) = @_;
+	# typically used for reading CGI headers
+	# We also need to check EINTR for generic PSGI servers.
+	my $ret;
+	my $total_rd = 0;
+	my $hdr_buf = $self->{hdr_buf};
+	my ($ph_cb, $ph_arg) = @{$self->{parse_hdr}};
+	until (defined($ret)) {
+		my $r = sysread($self->{rpipe}, $$hdr_buf, 4096,
+				length($$hdr_buf));
+		if (defined($r)) {
+			$total_rd += $r;
+			eval { $ret = $ph_cb->($total_rd, $hdr_buf, $ph_arg) };
+			if ($@) {
+				warn "parse_hdr: $@";
+				$ret = [ 500, [], [ "Internal error\n" ] ];
+			} elsif (!defined($ret) && !$r) {
+				warn <<EOM;
+EOF parsing headers from @{$self->{cmd}} ($self->{psgi_env}->{REQUEST_URI})
+EOM
+				$ret = [ 500, [], [ "Internal error\n" ] ];
+			}
+		} else {
+			# caller should notify us when it's ready:
+			return if $! == EAGAIN;
+			next if $! == EINTR; # immediate retry
+			warn "error reading header: $!";
+			$ret = [ 500, [], [ "Internal error\n" ] ];
+		}
+	}
+	delete $self->{parse_hdr}; # done parsing headers
+	$ret;
+}
+
+sub psgi_return_init_cb { # this may be PublicInbox::HTTPD::Async {cb}
+	my ($self) = @_;
+	my $r = rd_hdr($self) or return;
+	my $env = $self->{psgi_env};
+	my $filter;
+
+	# this is for RepoAtom since that can fire after parse_cgi_headers
+	if (ref($r) eq 'ARRAY' && blessed($r->[2]) && $r->[2]->can('attach')) {
+		$filter = pop @$r;
+	}
+	$filter //= delete($env->{'qspawn.filter'}) // (ref($r) eq 'ARRAY' ?
+		PublicInbox::GzipFilter::qsp_maybe($r->[1], $env) : undef);
+
+	my $wcb = delete $env->{'qspawn.wcb'};
+	my $async = delete $self->{async}; # PublicInbox::HTTPD::Async
+	if (ref($r) ne 'ARRAY' || scalar(@$r) == 3) { # error
+		if ($async) { # calls rpipe->close && ->event_step
+			$async->close; # PublicInbox::HTTPD::Async::close
+		} else { # generic PSGI, use PublicInbox::ProcessPipe::CLOSE
+			delete($self->{rpipe})->close;
+			event_step($self);
+		}
+		if (ref($r) eq 'ARRAY') { # error
+			$wcb->($r)
+		} elsif (ref($r) eq 'CODE') { # chain another command
+			$r->($wcb);
+			$self->{passed} = 1;
+		}
+		# else do nothing
+	} elsif ($async) {
+		# done reading headers, handoff to read body
+		my $fh = $wcb->($r); # scalar @$r == 2
+		$fh = $filter->attach($fh) if $filter;
+		$self->{qfh} = $fh;
+		$async->async_pass($env->{'psgix.io'}, $fh,
+					delete($self->{hdr_buf}));
+	} else { # for synchronous PSGI servers
+		require PublicInbox::GetlineBody;
+		my $buf = delete $self->{hdr_buf};
+		$r->[2] = PublicInbox::GetlineBody->new($self->{rpipe},
+					\&event_step, $self, $$buf, $filter);
+		$wcb->($r);
+	}
+}
+
+sub psgi_return_start { # may run later, much later...
+	my ($self) = @_;
+	if (my $cb = $self->{psgi_env}->{'pi-httpd.async'}) {
+		# PublicInbox::HTTPD::Async->new(rpipe, $cb, $cb_arg, $end_obj)
+		$self->{async} = $cb->($self->{rpipe},
+					\&psgi_return_init_cb, $self, $self);
+	} else { # generic PSGI
+		psgi_return_init_cb($self) while $self->{parse_hdr};
+	}
+}
+
 # Used for streaming the stdout of one process as a PSGI response.
 #
 # $env is the PSGI env.
@@ -225,8 +299,9 @@ sub event_step {
 #                          psgi_return will return an anonymous
 #                          sub for the PSGI server to call
 #
-#   $env->{'qspawn.filter'} - filter callback, receives a string as input,
-#                             undef on EOF
+#   $env->{'qspawn.filter'} - filter object, responds to ->attach for
+#                             pi-httpd.async and ->translate for generic
+#                             PSGI servers
 #
 # $limiter - the Limiter object to use (uses the def_limiter if not given)
 #
@@ -238,96 +313,29 @@ sub event_step {
 #              psgix.io.  3-element arrays means the body is available
 #              immediately (or streamed via ->getline (pull-based)).
 sub psgi_return {
-	my ($self, $env, $limiter, $parse_hdr) = @_;
+	my ($self, $env, $limiter, $parse_hdr, $hdr_arg) = @_;
 	$self->{psgi_env} = $env;
-	my $rpipe;
-	my $buf = '';
-	my $rd_hdr = sub {
-		# typically used for reading CGI headers
-		# we must loop until EAGAIN for EPOLLET in HTTPD/Async.pm
-		# We also need to check EINTR for generic PSGI servers.
-		my $ret;
-		my $total_rd = 0;
-		do {
-			my $r = sysread($rpipe, $buf, 4096, length($buf));
-			if (defined($r)) {
-				$total_rd += $r;
-				$ret = $parse_hdr->($r ? $total_rd : 0, \$buf);
-			} else {
-				# caller should notify us when it's ready:
-				return if $! == EAGAIN;
-				next if $! == EINTR; # immediate retry
-				log_err($env, "error reading header: $!");
-				$ret = [ 500, [], [ "Internal error\n" ] ];
-			}
-		} until (defined $ret);
-		$ret;
-	};
-
-	my $wcb = delete $env->{'qspawn.wcb'}; # or PSGI server supplies it
-	my $async = $env->{'pi-httpd.async'};
-
-	my $cb = sub {
-		my $r = $rd_hdr->() or return;
-		$rd_hdr = undef; # done reading headers
-		my $filter = delete $env->{'qspawn.filter'};
-		if (scalar(@$r) == 3) { # error
-			if ($async) {
-				# calls rpipe->close && ->event_step
-				$async->close;
-			} else {
-				$rpipe->close;
-				event_step($self);
-			}
-			$wcb->($r);
-		} elsif ($async) {
-			# done reading headers, handoff to read body
-			my $fh = $wcb->($r); # scalar @$r == 2
-			$fh = filter_fh($fh, $filter) if $filter;
-			$self->{fh} = $fh;
-			$async->async_pass($env->{'psgix.io'}, $fh, \$buf);
-		} else { # for synchronous PSGI servers
-			require PublicInbox::GetlineBody;
-			$r->[2] = PublicInbox::GetlineBody->new($rpipe,
-						\&event_step, $self,
-						$buf, $filter);
-			$wcb->($r);
-		}
-
-		# Workaround a leak under Perl 5.16.3 when combined with
-		# Plack::Middleware::Deflater:
-		$wcb = undef;
-	};
+	$self->{hdr_buf} = \(my $hdr_buf = '');
+	$self->{parse_hdr} = [ $parse_hdr, $hdr_arg ];
 	$limiter ||= $def_limiter ||= PublicInbox::Qspawn::Limiter->new(32);
-	my $start_cb = sub { # may run later, much later...
-		($rpipe) = @_;
-		if ($async) {
-			# PublicInbox::HTTPD::Async->new($rpipe, $cb, $cb_arg,
-			#				 $end_obj)
-			$async = $async->($rpipe, $cb, undef, $self);
-			# $cb will call ->async_pass or ->close
-		} else { # generic PSGI
-			$cb->() while $rd_hdr;
-		}
-	};
 
 	# the caller already captured the PSGI write callback from
 	# the PSGI server, so we can call ->start, here:
-	return $self->start($limiter, $start_cb) if $wcb;
+	$env->{'qspawn.wcb'} and
+		return start($self, $limiter, \&psgi_return_start);
 
 	# the caller will return this sub to the PSGI server, so
-	# it can set the response callback (that is, for PublicInbox::HTTP,
-	# the chunked_wcb or identity_wcb callback), but other HTTP servers
-	# are supported:
+	# it can set the response callback (that is, for
+	# PublicInbox::HTTP, the chunked_wcb or identity_wcb callback),
+	# but other HTTP servers are supported:
 	sub {
-		($wcb) = @_;
-		$self->start($limiter, $start_cb);
-	};
+		$env->{'qspawn.wcb'} = $_[0];
+		start($self, $limiter, \&psgi_return_start);
+	}
 }
 
 package PublicInbox::Qspawn::Limiter;
-use strict;
-use warnings;
+use v5.12;
 
 sub new {
 	my ($class, $max) = @_;
@@ -343,12 +351,12 @@ sub new {
 }
 
 sub setup_rlimit {
-	my ($self, $name, $config) = @_;
-	foreach my $rlim (PublicInbox::Spawn::RLIMITS()) {
+	my ($self, $name, $cfg) = @_;
+	foreach my $rlim (@PublicInbox::Spawn::RLIMITS) {
 		my $k = lc($rlim);
 		$k =~ tr/_//d;
 		$k = "publicinboxlimiter.$name.$k";
-		defined(my $v = $config->{$k}) or next;
+		defined(my $v = $cfg->{$k}) or next;
 		my @rlimit = split(/\s*,\s*/, $v);
 		if (scalar(@rlimit) == 1) {
 			push @rlimit, $rlimit[0];