X-Git-Url: http://www.git.stargrave.org/?a=blobdiff_plain;f=lib%2FPublicInbox%2FGitHTTPBackend.pm;h=744324294917df02bd07166a8d4e9423714b871d;hb=refs%2Fheads%2Fmaster;hp=5750d254233a7a6f12c9539473e2db5800579628;hpb=2537d9dd897a87f4149fe5dccfd11f6d1f1e2d8e;p=public-inbox.git diff --git a/lib/PublicInbox/GitHTTPBackend.pm b/lib/PublicInbox/GitHTTPBackend.pm index 5750d254..74432429 100644 --- a/lib/PublicInbox/GitHTTPBackend.pm +++ b/lib/PublicInbox/GitHTTPBackend.pm @@ -1,17 +1,18 @@ -# Copyright (C) 2016-2018 all contributors +# Copyright (C) all contributors # License: AGPL-3.0+ # when no endpoints match, fallback to this and serve a static file # or smart HTTP. This is our wrapper for git-http-backend(1) package PublicInbox::GitHTTPBackend; use strict; -use warnings; +use v5.10.1; use Fcntl qw(:seek); -use IO::Handle; +use IO::Handle; # ->flush use HTTP::Date qw(time2str); -use HTTP::Status qw(status_message); -use Plack::Util; use PublicInbox::Qspawn; +use PublicInbox::Tmpfile; +use PublicInbox::WwwStatic qw(r @NO_CACHE); +use Carp (); # 32 is same as the git-daemon connection limit my $default_limiter = PublicInbox::Qspawn::Limiter->new(32); @@ -22,37 +23,19 @@ my @text = qw[HEAD info/refs info/attributes objects/info/(?:http-alternates|alternates|packs) cloneurl description]; -my @binary = qw! - objects/[a-f0-9]{2}/[a-f0-9]{38} - objects/pack/pack-[a-f0-9]{40}\.(?:pack|idx) - !; +my @binary = ('objects/[a-f0-9]{2}/[a-f0-9]{38,62}', + 'objects/pack/pack-[a-f0-9]{40,64}\.(?:pack|idx)'); our $ANY = join('|', @binary, @text, 'git-upload-pack'); -my $BIN = join('|', @binary); my $TEXT = join('|', @text); -my @no_cache = ('Expires', 'Fri, 01 Jan 1980 00:00:00 GMT', - 'Pragma', 'no-cache', - 'Cache-Control', 'no-cache, max-age=0, must-revalidate'); - -sub r ($;$) { - my ($code, $msg) = @_; - $msg ||= status_message($code); - my $len = length($msg); - [ $code, [qw(Content-Type text/plain Content-Length), $len, @no_cache], - [$msg] ] -} - sub serve { my ($env, $git, $path) = @_; - # XXX compatibility... ugh, can we stop supporting this? - $git = PublicInbox::Git->new($git) unless ref($git); - # Documentation/technical/http-protocol.txt in git.git # requires one and exactly one query parameter: - if ($env->{QUERY_STRING} =~ /\Aservice=git-\w+-pack\z/ || - $path =~ /\Agit-\w+-pack\z/) { + if ($env->{QUERY_STRING} =~ /\Aservice=git-[A-Za-z0-9_]+-pack\z/ || + $path =~ /\Agit-[A-Za-z0-9_]+-pack\z/) { my $ok = serve_smart($env, $git, $path); return $ok if $ok; } @@ -60,16 +43,7 @@ sub serve { serve_dumb($env, $git, $path); } -sub err ($@) { - my ($env, @msg) = @_; - $env->{'psgi.errors'}->print(@msg, "\n"); -} - -sub drop_client ($) { - if (my $io = $_[0]->{'psgix.io'}) { - $io->close; # this is Danga::Socket::close - } -} +sub ucarp { Carp::carp(@_); undef } my $prev = 0; my $exp; @@ -83,116 +57,43 @@ sub cache_one_year { sub serve_dumb { my ($env, $git, $path) = @_; - my @h; + my $h = []; my $type; - if ($path =~ m!\Aobjects/[a-f0-9]{2}/[a-f0-9]{38}\z!) { + if ($path =~ m!\Aobjects/[a-f0-9]{2}/[a-f0-9]{38,62}\z!) { $type = 'application/x-git-loose-object'; - cache_one_year(\@h); - } elsif ($path =~ m!\Aobjects/pack/pack-[a-f0-9]{40}\.pack\z!) { + cache_one_year($h); + } elsif ($path =~ m!\Aobjects/pack/pack-[a-f0-9]{40,64}\.pack\z!) { $type = 'application/x-git-packed-objects'; - cache_one_year(\@h); - } elsif ($path =~ m!\Aobjects/pack/pack-[a-f0-9]{40}\.idx\z!) { + cache_one_year($h); + } elsif ($path =~ m!\Aobjects/pack/pack-[a-f0-9]{40,64}\.idx\z!) { $type = 'application/x-git-packed-objects-toc'; - cache_one_year(\@h); + cache_one_year($h); } elsif ($path =~ /\A(?:$TEXT)\z/o) { $type = 'text/plain'; - push @h, @no_cache; + push @$h, @NO_CACHE; } else { return r(404); } - - my $f = $git->{git_dir} . '/' . $path; - return r(404) unless -f $f && -r _; # just in case it's a FIFO :P - my $size = -s _; - - # TODO: If-Modified-Since and Last-Modified? - open my $in, '<', $f or return r(404); - my $len = $size; - my $code = 200; - push @h, 'Content-Type', $type; - if (($env->{HTTP_RANGE} || '') =~ /\bbytes=(\d*)-(\d*)\z/) { - ($code, $len) = prepare_range($env, $in, \@h, $1, $2, $size); - if ($code == 416) { - push @h, 'Content-Range', "bytes */$size"; - return [ 416, \@h, [] ]; - } - } - push @h, 'Content-Length', $len; - my $n = 65536; - [ $code, \@h, Plack::Util::inline_object(close => sub { close $in }, - getline => sub { - return if $len == 0; - $n = $len if $len < $n; - my $r = sysread($in, my $buf, $n); - if (!defined $r) { - err($env, "$f read error: $!"); - } elsif ($r <= 0) { - err($env, "$f EOF with $len bytes left"); - } else { - $len -= $r; - $n = 8192; - return $buf; - } - drop_client($env); - return; - })] + $path = "$git->{git_dir}/$path"; + PublicInbox::WwwStatic::response($env, $h, $path, $type); } -sub prepare_range { - my ($env, $in, $h, $beg, $end, $size) = @_; - my $code = 200; - my $len = $size; - if ($beg eq '') { - if ($end ne '') { # "bytes=-$end" => last N bytes - $beg = $size - $end; - $beg = 0 if $beg < 0; - $end = $size - 1; - $code = 206; - } else { - $code = 416; - } - } else { - if ($beg > $size) { - $code = 416; - } elsif ($end eq '' || $end >= $size) { - $end = $size - 1; - $code = 206; - } elsif ($end < $size) { - $code = 206; - } else { - $code = 416; - } - } - if ($code == 206) { - $len = $end - $beg + 1; - if ($len <= 0) { - $code = 416; - } else { - sysseek($in, $beg, SEEK_SET) or return [ 500, [], [] ]; - push @$h, qw(Accept-Ranges bytes Content-Range); - push @$h, "bytes $beg-$end/$size"; - - # FIXME: Plack::Middleware::Deflater bug? - $env->{'psgix.no-compress'} = 1; - } - } - ($code, $len); +sub git_parse_hdr { # {parse_hdr} for Qspawn + my ($r, $bref, $dumb_args) = @_; + my $res = parse_cgi_headers($r, $bref) or return; # incomplete + $res->[0] == 403 ? serve_dumb(@$dumb_args) : $res; } # returns undef if 403 so it falls back to dumb HTTP sub serve_smart { my ($env, $git, $path) = @_; - my $in = $env->{'psgi.input'}; - my $fd = eval { fileno($in) }; - unless (defined $fd && $fd >= 0) { - $in = input_to_file($env) or return r(500); - } my %env = %ENV; # GIT_COMMITTER_NAME, GIT_COMMITTER_EMAIL # may be set in the server-process and are passed as-is foreach my $name (qw(QUERY_STRING REMOTE_USER REMOTE_ADDR HTTP_CONTENT_ENCODING + HTTP_GIT_PROTOCOL CONTENT_TYPE SERVER_PROTOCOL REQUEST_METHOD)) { @@ -202,65 +103,62 @@ sub serve_smart { my $limiter = $git->{-httpbackend_limiter} || $default_limiter; $env{GIT_HTTP_EXPORT_ALL} = '1'; $env{PATH_TRANSLATED} = "$git->{git_dir}/$path"; - my $rdr = { 0 => fileno($in) }; + my $rdr = input_prepare($env) or return r(500); my $qsp = PublicInbox::Qspawn->new([qw(git http-backend)], \%env, $rdr); - $qsp->psgi_return($env, $limiter, sub { - my ($r, $bref) = @_; - $r = parse_cgi_headers($bref) or return; # incomplete headers - $r->[0] == 403 ? serve_dumb($env, $git, $path) : $r; - }); + $qsp->psgi_return($env, $limiter, \&git_parse_hdr, [$env, $git, $path]); } -sub input_to_file { +sub input_prepare { my ($env) = @_; - open(my $in, '+>', undef); - unless (defined $in) { - err($env, "could not open temporary file: $!"); - return; - } + my $input = $env->{'psgi.input'}; + my $fd = eval { fileno($input) }; + return { 0 => $fd } if (defined $fd && $fd >= 0); + my $id = "git-http.input.$env->{REMOTE_ADDR}:$env->{REMOTE_PORT}"; + my $in = tmpfile($id) // return ucarp("tmpfile: $!"); my $buf; while (1) { - my $r = $input->read($buf, 8192); - unless (defined $r) { - err($env, "error reading input: $!"); - return; - } + my $r = $input->read($buf, 8192) // return ucarp("read $!"); last if $r == 0; - my $off = 0; - while ($r > 0) { - my $w = syswrite($in, $buf, $r, $off); - if (defined $w) { - $r -= $w; - $off += $w; - } else { - err($env, "error writing temporary file: $!"); - return; - } - } + print $in $buf // return ucarp("print: $!"); } - unless (defined(sysseek($in, 0, SEEK_SET))) { - err($env, "error seeking temporary file: $!"); - return; - } - return $in; + # ensure it's visible to git-http-backend(1): + $in->flush // return ucarp("flush: $!"); + sysseek($in, 0, SEEK_SET) // return ucarp($env, "seek: $!"); + { 0 => $in }; } -sub parse_cgi_headers { - my ($bref) = @_; - $$bref =~ s/\A(.*?)\r\n\r\n//s or return; +sub parse_cgi_headers { # {parse_hdr} for Qspawn + my ($r, $bref, $ctx) = @_; + return r(500) unless defined $r && $r >= 0; + $$bref =~ s/\A(.*?)\r?\n\r?\n//s or return $r == 0 ? r(500) : undef; my $h = $1; my $code = 200; my @h; - foreach my $l (split(/\r\n/, $h)) { + foreach my $l (split(/\r?\n/, $h)) { my ($k, $v) = split(/:\s*/, $l, 2); if ($k =~ /\AStatus\z/i) { - ($code) = ($v =~ /\b(\d+)\b/); + ($code) = ($v =~ /\b([0-9]+)\b/); } else { push @h, $k, $v; } } - [ $code, \@h ] + + # fallback to WwwCoderepo if cgit 404s. Duplicating $ctx prevents + # ->finalize from the current Qspawn from using qspawn.wcb. + # This makes qspawn skip ->async_pass and causes + # PublicInbox::HTTPD::Async::event_step to close shortly after + if ($code == 404 && $ctx->{www} && !$ctx->{_coderepo_tried}++) { + my $wcb = delete $ctx->{env}->{'qspawn.wcb'}; + $ctx->{env}->{'plack.skip-deflater'} = 1; # prevent 2x gzip + $ctx->{env}->{'qspawn.fallback'} = $code; + my $res = $ctx->{www}->coderepo->srv($ctx); + # for ->psgi_return_init_cb + $ctx->{env}->{'qspawn.wcb'} = $wcb; + $res; # CODE or ARRAY ref + } else { + [ $code, \@h ] + } } 1;