diff options
Diffstat (limited to 'lib/PublicInbox/WwwCoderepo.pm')
-rw-r--r-- | lib/PublicInbox/WwwCoderepo.pm | 396 |
1 files changed, 288 insertions, 108 deletions
diff --git a/lib/PublicInbox/WwwCoderepo.pm b/lib/PublicInbox/WwwCoderepo.pm index fb510b28..a5e2dc4a 100644 --- a/lib/PublicInbox/WwwCoderepo.pm +++ b/lib/PublicInbox/WwwCoderepo.pm @@ -1,21 +1,45 @@ # Copyright (C) all contributors <meta@public-inbox.org> # License: AGPL-3.0+ <https://www.gnu.org/licenses/agpl-3.0.txt> # -# Standalone code repository viewer for users w/o cgit +# Standalone code repository viewer for users w/o cgit. +# This isn't intended to replicate all of cgit, but merely to be a +# "good enough" viewer with search support and some UI hints to encourage +# cloning + command-line usage. package PublicInbox::WwwCoderepo; use v5.12; +use parent qw(PublicInbox::WwwStream); use File::Temp 0.19 (); # newdir +use POSIX qw(O_RDWR F_GETFL); use PublicInbox::ViewVCS; use PublicInbox::WwwStatic qw(r); use PublicInbox::GitHTTPBackend; -use PublicInbox::Git; -use PublicInbox::GitAsyncCat; use PublicInbox::WwwStream; -use PublicInbox::Hval qw(ascii_html); +use PublicInbox::Hval qw(prurl ascii_html utf8_maybe); +use PublicInbox::ViewDiff qw(uri_escape_path); +use PublicInbox::RepoSnapshot; +use PublicInbox::RepoAtom; +use PublicInbox::RepoTree; +use PublicInbox::RepoList; +use PublicInbox::OnDestroy; +use URI::Escape qw(uri_escape_utf8); +use File::Spec; +use autodie qw(fcntl open); -my $EACH_REF = "git for-each-ref --sort=-creatordate --format='%(HEAD)%00". - join('%00', map { "%($_)" } - qw(objectname refname:short subject creatordate:short))."'"; +my @EACH_REF = (qw(git for-each-ref --sort=-creatordate), + "--format=%(HEAD)%00".join('%00', map { "%($_)" } + qw(objectname refname:short subject creatordate:short))); +my $HEADS_CMD = <<''; +# heads (aka `branches'): +$ git for-each-ref --sort=-creatordate refs/heads \ + --format='%(HEAD) %(refname:short) %(subject) (%(creatordate:short))' + +my $TAGS_CMD = <<''; +# tags: +$ git for-each-ref --sort=-creatordate refs/tags \ + --format='%(refname:short) %(subject) (%(creatordate:short))' + +my $NO_HEADS = "# no heads (branches), yet...\n"; +my $NO_TAGS = "# no tags, yet...\n"; # shared with PublicInbox::Cgit sub prepare_coderepos { @@ -23,40 +47,132 @@ sub prepare_coderepos { my $pi_cfg = $self->{pi_cfg}; # TODO: support gitweb and other repository viewers? - if (defined(my $cgitrc = $pi_cfg->{-cgitrc_unparsed})) { - $pi_cfg->parse_cgitrc($cgitrc, 0); - } - my $code_repos = $pi_cfg->{-code_repos}; + $pi_cfg->parse_cgitrc(undef, 0); + + my $coderepos = $pi_cfg->{-coderepos}; for my $k (grep(/\Acoderepo\.(?:.+)\.dir\z/, keys %$pi_cfg)) { $k = substr($k, length('coderepo.'), -length('.dir')); - $code_repos->{$k} //= $pi_cfg->fill_code_repo($k); + $coderepos->{$k} //= $pi_cfg->fill_coderepo($k); + } + + # associate inboxes and extindices with coderepos for search: + for my $k (grep(/\Apublicinbox\.(?:.+)\.coderepo\z/, keys %$pi_cfg)) { + $k = substr($k, length('publicinbox.'), -length('.coderepo')); + my $ibx = $pi_cfg->lookup_name($k) // next; + $pi_cfg->repo_objs($ibx); } - while (my ($nick, $repo) = each %$code_repos) { - $self->{"\0$nick"} = $repo; + for my $k (grep(/\Aextindex\.(?:.+)\.coderepo\z/, keys %$pi_cfg)) { + $k = substr($k, length('extindex.'), -length('.coderepo')); + my $eidx = $pi_cfg->lookup_ei($k) // next; + $pi_cfg->repo_objs($eidx); } + $pi_cfg->each_cindex('load_coderepos', $pi_cfg); } sub new { my ($cls, $pi_cfg) = @_; my $self = bless { pi_cfg => $pi_cfg }, $cls; prepare_coderepos($self); + $self->{snapshots} = do { + my $s = $pi_cfg->{'coderepo.snapshots'} // ''; + $s eq 'all' ? \%PublicInbox::RepoSnapshot::FMT_TYPES : + +{ map { $_ => 1 } split(/\s+/, $s) }; + }; $self->{$_} = 10 for qw(summary_branches summary_tags); $self->{$_} = 10 for qw(summary_log); + + # try reuse STDIN if it's already /dev/null + open $self->{log_fh}, '+>', '/dev/null'; + my @l = stat($self->{log_fh}) or die "stat: $!"; + my @s = stat(STDIN) or die "stat(STDIN): $!"; + if ("@l[0, 1]" eq "@s[0, 1]") { + my $f = fcntl(STDIN, F_GETFL, 0); + $self->{log_fh} = *STDIN{IO} if $f & O_RDWR; + } $self; } -sub summary_finish { +sub _snapshot_link_prep { + my ($ctx) = @_; + my @s = sort keys %{$ctx->{wcr}->{snapshots}} or return (); + my $n = $ctx->{git}->local_nick // die "BUG: $ctx->{git_dir} nick"; + $n =~ s!\.git/*\z!!; + ($n) = ($n =~ m!([^/]+)/*\z!); + (ascii_html($n).'-', @s); +} + +sub _refs_heads_link { + my ($line, $upfx) = @_; + my ($pfx, $oid, $ref, $s, $cd) = split(/\0/, $line); + my $align = length($ref) < 12 ? ' ' x (12 - length($ref)) : ''; + ("$pfx <a\nhref=$upfx$oid/s/>", ascii_html($ref), + "</a>$align ", ascii_html($s), " ($cd)\n") +} + +sub _refs_tags_link { + my ($line, $upfx, $snap_pfx, @snap_fmt) = @_; + my (undef, $oid, $ref, $s, $cd) = split(/\0/, $line); + my $align = length($ref) < 12 ? ' ' x (12 - length($ref)) : ''; + if (@snap_fmt) { + my $v = $ref; + $v =~ s/\A[vV]//; + @snap_fmt = map { + qq{ <a href="${upfx}snapshot/$snap_pfx$v.$_">$_</a>} + } @snap_fmt; + } + ("<a\nhref=$upfx$oid/s/>", ascii_html($ref), + "</a>$align ", ascii_html($s), " ($cd)", @snap_fmt, "\n"); +} + +sub emit_joined_inboxes ($) { + my ($ctx) = @_; + my $names = $ctx->{git}->{ibx_names}; # coderepo directives in config + my $score = $ctx->{git}->{ibx_score}; # generated w/ cindex --join + ($names || $score) or return; + my $pi_cfg = $ctx->{wcr}->{pi_cfg}; + my ($u, $h); + my $zfh = $ctx->zfh; + print $zfh "\n# associated public inboxes:", + "\n# (number on the left is used for dev purposes)"; + my @ns = map { [ 0, $_ ] } @$names; + my $env = $ctx->{env}; + for (@ns, @$score) { + my ($nr, $name) = @$_; + my $ibx = $pi_cfg->lookup_name($name) // do { + warn "W: inbox `$name' gone for $ctx->{git}->{git_dir}"; + say $zfh '# ', ascii_html($name), ' (missing inbox?)'; + next; + }; + if (scalar(@{$ibx->{url} // []})) { + $u = $h = ascii_html(prurl($env, $ibx->{url})); + } else { + $h = ascii_html(prurl($env, uri_escape_utf8($name))); + $h .= '/'; + $u = ascii_html($name); + } + if ($nr) { + printf $zfh "\n% 11u", $nr; + } else { + print $zfh "\n", ' 'x11; + } + print $zfh qq{ <a\nhref="$h">$u</a>}; + } +} + +sub summary_END { # called via OnDestroy my ($ctx) = @_; - my $wcb = delete($ctx->{env}->{'qspawn.wcb'}) or return; # already done - my @x = split(/\n\n/sm, delete($ctx->{-each_refs})); + my $wcb = delete($ctx->{-wcb}) or return; # already done PublicInbox::WwwStream::html_init($ctx); my $zfh = $ctx->zfh; - # git log - my @r = split(/\n/s, pop(@x) // ''); - my $last = pop(@r) if scalar(@r) > $ctx->{wcr}->{summary_log}; - print $zfh '<pre><a id=log>$</a> '. - "git log --pretty=format:'%h %s (%cs)%d'\n"; + my @r = split(/\n/s, delete($ctx->{qx_res}->{'log'}) // ''); + my $last = scalar(@r) > $ctx->{wcr}->{summary_log} ? pop(@r) : undef; + my $tip_html = ''; + my $tip = $ctx->{qp}->{h}; + $tip_html .= ' '.ascii_html($tip).' --' if defined $tip; + print $zfh <<EOM; +<pre><a id=log>\$</a> git log --pretty=format:'%h %s (%cs)%d'$tip_html +EOM for (@r) { my $d; # decorations s/^ \(([^\)]+)\)// and $d = $1; @@ -66,98 +182,143 @@ sub summary_finish { " (", $cs, ")\n"; print $zfh "\t(", ascii_html($d), ")\n" if $d; } - print $zfh "# no commits, yet\n" if !@r; + print $zfh '# no commits in `', ($tip//'HEAD'),"', yet\n\n" if !@r; print $zfh "...\n" if $last; # README - my ($bref, $oid, $ref_path) = @{delete $ctx->{-readme}}; + my ($bref, $oid, $ref_path) = @{delete $ctx->{qx_res}->{readme}}; if ($bref) { my $l = PublicInbox::Linkify->new; $$bref =~ s/\s*\z//sm; + my (undef, $path) = split(/:/, $ref_path, 2); # HEAD:README print $zfh "\n<a id=readme>\$</a> " . - "git cat-file blob <a href=./$oid/s/>", + qq(git cat-file blob <a href="./$oid/s/?b=) . + ascii_html(uri_escape_path($path)) . q(">). ascii_html($ref_path), "</a>\n", $l->to_html($$bref), '</pre><hr><pre>'; } # refs/heads - print $zfh "<a id=heads># heads (aka `branches'):</a>\n\$ " . - "git for-each-ref --sort=-creatordate refs/heads" . - " \\\n\t--format='%(HEAD) ". # no space for %(align:) hint - "%(refname:short) %(subject) (%(creatordate:short))'\n"; - @r = split(/^/sm, shift(@x) // ''); - $last = pop(@r) if scalar(@r) > $ctx->{wcr}->{summary_branches}; - for (@r) { - my ($pfx, $oid, $ref, $s, $cd) = split(/\0/); - utf8::decode($_) for ($ref, $s); - chomp $cd; - my $align = length($ref) < 12 ? ' ' x (12 - length($ref)) : ''; - print $zfh "$pfx <a\nhref=./$oid/s/>", ascii_html($ref), - "</a>$align ", ascii_html($s), " ($cd)\n"; - } - print $zfh "# no heads (branches) yet...\n" if !@r; - print $zfh "...\n" if $last; - print $zfh "\n<a id=tags># tags:</a>\n\$ " . - "git for-each-ref --sort=-creatordate refs/tags" . - " \\\n\t--format='". # no space for %(align:) hint - "%(refname:short) %(subject) (%(creatordate:short))'\n"; - @r = split(/^/sm, shift(@x) // ''); - $last = pop(@r) if scalar(@r) > $ctx->{wcr}->{summary_tags}; - for (@r) { - my (undef, $oid, $ref, $s, $cd) = split(/\0/); - utf8::decode($_) for ($ref, $s); - chomp $cd; - my $align = length($ref) < 12 ? ' ' x (12 - length($ref)) : ''; - print $zfh "<a\nhref=./$oid/s/>", ascii_html($ref), - "</a>$align ", ascii_html($s), " ($cd)\n"; - } - print $zfh "# no tags yet...\n" if !@r; - print $zfh "...\n" if $last; + print $zfh '<a id=heads>', $HEADS_CMD , '</a>'; + @r = split(/^/sm, delete($ctx->{qx_res}->{heads}) // ''); + $last = scalar(@r) > $ctx->{wcr}->{summary_branches} ? pop(@r) : undef; + chomp(@r); + for (@r) { print $zfh _refs_heads_link($_, './') } + print $zfh $NO_HEADS if !@r; + print $zfh qq(<a href="refs/heads/">...</a>\n) if $last; + print $zfh "\n<a id=tags>", $TAGS_CMD, '</a>'; + @r = split(/^/sm, delete($ctx->{qx_res}->{tags}) // ''); + $last = scalar(@r) > $ctx->{wcr}->{summary_tags} ? pop(@r) : undef; + my ($snap_pfx, @snap_fmt) = _snapshot_link_prep($ctx); + chomp @r; + for (@r) { print $zfh _refs_tags_link($_, './', $snap_pfx, @snap_fmt) } + print $zfh $NO_TAGS if !@r; + print $zfh qq(<a href="refs/tags/">...</a>\n) if $last; + emit_joined_inboxes $ctx; $wcb->($ctx->html_done('</pre>')); } -sub capture_refs ($$) { # psgi_qx callback to capture git-for-each-ref + git-log - my ($bref, $ctx) = @_; - my $qsp_err = delete $ctx->{-qsp_err}; - $ctx->{-each_refs} = $$bref; - summary_finish($ctx) if $ctx->{-readme}; +sub capture { # psgi_qx callback to capture git-for-each-ref + my ($bref, $ctx, $key) = @_; # $_[3] = OnDestroy(summary_END) + $ctx->{qx_res}->{$key} = $$bref; + # summary_END may be called via OnDestroy $arg->[2] } sub set_readme { # git->cat_async callback my ($bref, $oid, $type, $size, $ctx) = @_; - my $ref_path = shift @{$ctx->{-nr_readme_tries}}; # e.g. HEAD:README - if ($type eq 'blob' && !$ctx->{-readme}) { - $ctx->{-readme} = [ $bref, $oid, $ref_path ]; - } elsif (scalar @{$ctx->{-nr_readme_tries}} == 0) { - $ctx->{-readme} //= []; # nothing left to try + my $ref_path = shift @{$ctx->{-readme_tries}}; # e.g. HEAD:README + if ($type eq 'blob' && !$ctx->{qx_res}->{readme}) { + $ctx->{qx_res}->{readme} = [ $bref, $oid, $ref_path ]; + } elsif (scalar @{$ctx->{-readme_tries}} == 0) { + $ctx->{qx_res}->{readme} //= []; # nothing left to try } # or try another README... - summary_finish($ctx) if $ctx->{-each_refs} && $ctx->{-readme}; + # summary_END may be called via OnDestroy ($ctx->{-END}) } -sub summary { - my ($self, $ctx) = @_; - $ctx->{wcr} = $self; - my $nb = $self->{summary_branches} + 1; - my $nt = $self->{summary_tags} + 1; - my $nl = $self->{summary_log} + 1; - my $qsp = PublicInbox::Qspawn->new([qw(/bin/sh -c), - "$EACH_REF --count=$nb refs/heads; echo && " . - "$EACH_REF --count=$nt refs/tags; echo && " . - "git log -$nl --pretty=format:'%d %H %h %cs %s' --" ], - { GIT_DIR => $ctx->{git}->{git_dir} }); - $qsp->{qsp_err} = \($ctx->{-qsp_err} = ''); - my @try = qw(HEAD:README HEAD:README.md); # TODO: configurable - $ctx->{-nr_readme_tries} = [ @try ]; - $ctx->{git}->cat_async($_, \&set_readme, $ctx) for @try; - if ($ctx->{env}->{'pi-httpd.async'}) { - PublicInbox::GitAsyncCat::watch_cat($ctx->{git}); - } else { # synchronous - $ctx->{git}->cat_async_wait; +sub summary ($$) { + my ($ctx, $wcb) = @_; + $ctx->{-wcb} = $wcb; # PublicInbox::HTTP::{Identity,Chunked} + my $tip = $ctx->{qp}->{h}; # same as cgit + if (defined $tip && $tip eq '') { + delete $ctx->{qp}->{h}; + undef($tip); + } + my ($nb, $nt, $nl) = map { $_ + 1 } @{$ctx->{wcr}}{qw( + summary_branches summary_tags summary_log)}; + $ctx->{qx_res} = {}; + my $qsp_err = \($ctx->{-qsp_err} = ''); + my %opt = (quiet => 1, 2 => $ctx->{wcr}->{log_fh}); + my %env = (GIT_DIR => $ctx->{git}->{git_dir}); + my @log = (qw(git log), "-$nl", '--pretty=format:%d %H %h %cs %s'); + push(@log, $tip) if defined $tip; + + # limit scope for MockHTTP test (t/solver_git.t) + my $END = on_destroy \&summary_END, $ctx; + for (['log', \@log], + [ 'heads', [@EACH_REF, "--count=$nb", 'refs/heads'] ], + [ 'tags', [@EACH_REF, "--count=$nt", 'refs/tags'] ]) { + my ($k, $cmd) = @$_; + my $qsp = PublicInbox::Qspawn->new($cmd, \%env, \%opt); + $qsp->{qsp_err} = $qsp_err; + $qsp->psgi_qx($ctx->{env}, undef, \&capture, $ctx, $k, $END); } - sub { # $_[0] => PublicInbox::HTTP::{Identity,Chunked} - $ctx->{env}->{'qspawn.wcb'} = $_[0]; - $qsp->psgi_qx($ctx->{env}, undef, \&capture_refs, $ctx); + $tip //= 'HEAD'; + my @try = ("$tip:README", "$tip:README.md"); # TODO: configurable + my %ctx = (%$ctx, -END => $END, -readme_tries => [ @try ]); + PublicInbox::ViewVCS::do_cat_async(\%ctx, \&set_readme, @try); +} + +# called by GzipFilter->close after translate +sub zflush { $_[0]->SUPER::zflush('</pre>', $_[0]->_html_end) } + +# called by GzipFilter->write or GetlineResponse->getline +sub translate { + my $ctx = shift; + $_[0] // return zflush($ctx); # getline caller + my @out; + my $fbuf = delete($ctx->{fbuf}) // shift; + $fbuf .= shift while @_; + if ($ctx->{-heads}) { + while ($fbuf =~ s/\A([^\n]+)\n//s) { + utf8_maybe(my $x = $1); + push @out, _refs_heads_link($x, '../../'); + } + } else { + my ($snap_pfx, @snap_fmt) = _snapshot_link_prep($ctx); + while ($fbuf =~ s/\A([^\n]+)\n//s) { + utf8_maybe(my $x = $1); + push @out, _refs_tags_link($x, '../../', + $snap_pfx, @snap_fmt); + } } + $ctx->{fbuf} = $fbuf; # may be incomplete + @out ? $ctx->SUPER::translate(@out) : ''; # not EOF, yet +} + +sub _refs_parse_hdr { # {parse_hdr} for Qspawn + my ($r, $bref, $ctx) = @_; + my ($code, $top); + if ($r == 0) { + $code = 404; + $top = $ctx->{-heads} ? $NO_HEADS : $NO_TAGS; + } else { + $code = 200; + $top = $ctx->{-heads} ? $HEADS_CMD : $TAGS_CMD; + } + PublicInbox::WwwStream::html_init($ctx); + bless $ctx, __PACKAGE__; # re-bless for ->translate + print { $ctx->{zfh} } '<pre>', $top; + [ $code, delete($ctx->{-res_hdr}), $ctx ]; # [2] is qspawn.filter +} + +sub refs_foo { # /$REPO/refs/{heads,tags} endpoints + my ($self, $ctx, $pfx) = @_; + $ctx->{wcr} = $self; + $ctx->{-upfx} = '../../'; + $ctx->{-heads} = 1 if $pfx eq 'refs/heads'; + my $qsp = PublicInbox::Qspawn->new([@EACH_REF, $pfx ], + { GIT_DIR => $ctx->{git}->{git_dir} }); + $qsp->psgi_yield($ctx->{env}, undef, \&_refs_parse_hdr, $ctx); } sub srv { # endpoint called by PublicInbox::WWW @@ -165,25 +326,44 @@ sub srv { # endpoint called by PublicInbox::WWW my $path_info = $ctx->{env}->{PATH_INFO}; my $git; # handle clone requests - if ($path_info =~ m!\A/(.+?)/($PublicInbox::GitHTTPBackend::ANY)\z!x) { - $git = $self->{"\0$1"} and return + my $pi_cfg = $self->{pi_cfg}; + if ($path_info =~ m!\A/(.+?)/($PublicInbox::GitHTTPBackend::ANY)\z!x and + ($git = $pi_cfg->get_coderepo($1))) { PublicInbox::GitHTTPBackend::serve($ctx->{env},$git,$2); - } - $path_info =~ m!\A/(.+?)/\z! and - ($ctx->{git} = $self->{"\0$1"}) and return summary($self, $ctx); - $path_info =~ m!\A/(.+?)/([a-f0-9]+)/s/\z! and - ($ctx->{git} = $self->{"\0$1"}) and - return PublicInbox::ViewVCS::show($ctx, $2); - - # snapshots: - if ($path_info =~ m!\A/(.+?)/snapshot/([^/]+)\z! and - ($ctx->{git} = $self->{"\0$1"})) { - require PublicInbox::RepoSnapshot; - return PublicInbox::RepoSnapshot::srv($ctx, $2) // r(404); - } - - # enforce trailing slash: - if ($path_info =~ m!\A/(.+?)\z! and ($git = $self->{"\0$1"})) { + } elsif ($path_info =~ m!\A/(.+?)/\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + $ctx->{wcr} = $self; + sub { summary($ctx, $_[0]) }; # $_[0] = wcb + } elsif ($path_info =~ m!\A/(.+?)/([a-f0-9]+)/s/([^/]+)?\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + $ctx->{lh} = $self->{log_fh}; + PublicInbox::ViewVCS::show($ctx, $2, $3); + } elsif ($path_info =~ m!\A/(.+?)/tree/(.*)\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + $ctx->{lh} = $self->{log_fh}; + PublicInbox::RepoTree::srv_tree($ctx, $2) // r(404); + } elsif ($path_info =~ m!\A/(.+?)/snapshot/([^/]+)\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + $ctx->{wcr} = $self; + PublicInbox::RepoSnapshot::srv($ctx, $2) // r(404); + } elsif ($path_info =~ m!\A/(.+?)/atom/(.*)\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + $ctx->{lh} = $self->{log_fh}; + PublicInbox::RepoAtom::srv_atom($ctx, $2) // r(404); + } elsif ($path_info =~ m!\A/(.+?)/tags\.atom\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + PublicInbox::RepoAtom::srv_tags_atom($ctx); + } elsif ($path_info =~ m!\A/(.+?)/(refs/(?:heads|tags))/\z! and + ($ctx->{git} = $pi_cfg->get_coderepo($1))) { + refs_foo($self, $ctx, $2); + } elsif ($path_info =~ m!\A/(.*?\*.*?)/*\z!) { + my $re = PublicInbox::Config::glob2re($1); + PublicInbox::RepoList::html($self, $ctx, qr!$re\z!) // r(404); + } elsif ($path_info =~ m!\A/(.+?)/\z!) { + my $re = qr!\A\Q$1\E/!; + PublicInbox::RepoList::html($self, $ctx, $re) // r(404); + } elsif ($path_info =~ m!\A/(.+?)\z! and + ($git = $pi_cfg->get_coderepo($1))) { my $qs = $ctx->{env}->{QUERY_STRING}; my $url = $git->base_url($ctx->{env}); $url .= "?$qs" if $qs ne ''; |