diff options
Diffstat (limited to 'lib/PublicInbox/ViewVCS.pm')
| -rw-r--r-- | lib/PublicInbox/ViewVCS.pm | 655 |
1 files changed, 559 insertions, 96 deletions
diff --git a/lib/PublicInbox/ViewVCS.pm b/lib/PublicInbox/ViewVCS.pm index 3cbc363b..f47c2703 100644 --- a/lib/PublicInbox/ViewVCS.pm +++ b/lib/PublicInbox/ViewVCS.pm @@ -1,8 +1,7 @@ -# Copyright (C) 2019-2021 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> # show any VCS object, similar to "git show" -# FIXME: we only show blobs for now # # This can use a "solver" to reconstruct blobs based on git # patches (with abbreviated OIDs in the header). However, the @@ -16,11 +15,22 @@ package PublicInbox::ViewVCS; use strict; use v5.10.1; +use File::Temp 0.19 (); # newdir use PublicInbox::SolverGit; +use PublicInbox::Git; +use PublicInbox::GitAsyncCat; use PublicInbox::WwwStream qw(html_oneshot); use PublicInbox::Linkify; use PublicInbox::Tmpfile; -use PublicInbox::Hval qw(ascii_html to_filename); +use PublicInbox::ViewDiff qw(flush_diff uri_escape_path); +use PublicInbox::View; +use PublicInbox::Eml; +use PublicInbox::OnDestroy; +use Text::Wrap qw(wrap); +use PublicInbox::Hval qw(ascii_html to_filename prurl utf8_maybe); +use POSIX qw(strftime); +use autodie qw(open seek truncate); +use Fcntl qw(SEEK_SET); my $hl = eval { require PublicInbox::HlMod; PublicInbox::HlMod->new; @@ -29,22 +39,52 @@ my $hl = eval { my %QP_MAP = ( A => 'oid_a', a => 'path_a', b => 'path_b' ); our $MAX_SIZE = 1024 * 1024; # TODO: configurable my $BIN_DETECT = 8000; # same as git +my $SHOW_FMT = '--pretty=format:'.join('%n', '%P', '%p', '%H', '%T', '%s', '%f', + '%an <%ae> %ai', '%cn <%ce> %ci', '%b%x00'); -sub html_page ($$$) { - my ($ctx, $code, $strref) = @_; +my %GIT_MODE = ( + '100644' => ' ', # blob + '100755' => 'x', # executable blob + '040000' => 'd', # tree + '120000' => 'l', # symlink + '160000' => 'g', # commit (gitlink) +); + +# TODO: not fork safe, but we don't fork w/o exec in PublicInbox::WWW +my (@solver_q, $solver_lim); +my $solver_nr = 0; + +sub html_page ($$;@) { + my ($ctx, $code) = @_[0, 1]; my $wcb = delete $ctx->{-wcb}; - $ctx->{-upfx} = '../../'; # from "/$INBOX/$OID/s/" - my $res = html_oneshot($ctx, $code, $strref); + $ctx->{-upfx} //= '../../'; # from "/$INBOX/$OID/s/" + my $res = html_oneshot($ctx, $code, @_[2..$#_]); $wcb ? $wcb->($res) : $res; } +sub dbg_log ($) { + my ($ctx) = @_; + my $log = delete $ctx->{lh} // die 'BUG: already captured debug log'; + if (!CORE::seek($log, 0, SEEK_SET)) { + warn "seek(log): $!"; + return '<pre>debug log seek error</pre>'; + } + $log = eval { PublicInbox::IO::read_all $log } // do { + warn "read(log): $@"; + return '<pre>debug log read error</pre>'; + }; + return '' if $log eq ''; + $ctx->{-linkify} //= PublicInbox::Linkify->new; + "<hr><pre>debug log:\n\n". + $ctx->{-linkify}->to_html($log).'</pre>'; +} + sub stream_blob_parse_hdr { # {parse_hdr} for Qspawn my ($r, $bref, $ctx) = @_; - my ($res, $logref) = delete @$ctx{qw(-res -logref)}; - my ($git, $oid, $type, $size, $di) = @$res; + my ($git, $oid, $type, $size, $di) = @{$ctx->{-res}}; my @cl = ('Content-Length', $size); - if (!defined $r) { # error - html_page($ctx, 500, $logref); + if (!defined $r) { # sysread error + html_page($ctx, 500, dbg_log($ctx)); } elsif (index($$bref, "\0") >= 0) { [200, [qw(Content-Type application/octet-stream), @cl] ]; } else { @@ -54,114 +94,512 @@ sub stream_blob_parse_hdr { # {parse_hdr} for Qspawn 'text/plain; charset=UTF-8', @cl ] ]; } if ($r == 0) { - warn "premature EOF on $oid $$logref"; - return html_page($ctx, 500, $logref); + my $log = dbg_log($ctx); + warn "premature EOF on $oid $log"; + return html_page($ctx, 500, $log); } - @$ctx{qw(-res -logref)} = ($res, $logref); undef; # bref keeps growing } } -sub stream_large_blob ($$$$) { - my ($ctx, $res, $logref, $fn) = @_; - $ctx->{-logref} = $logref; +sub stream_large_blob ($$) { + my ($ctx, $res) = @_; $ctx->{-res} = $res; my ($git, $oid, $type, $size, $di) = @$res; my $cmd = ['git', "--git-dir=$git->{git_dir}", 'cat-file', $type, $oid]; my $qsp = PublicInbox::Qspawn->new($cmd); - my $env = $ctx->{env}; - $env->{'qspawn.wcb'} = delete $ctx->{-wcb}; - $qsp->psgi_return($env, undef, \&stream_blob_parse_hdr, $ctx); + $ctx->{env}->{'qspawn.wcb'} = $ctx->{-wcb}; + $qsp->psgi_yield($ctx->{env}, undef, \&stream_blob_parse_hdr, $ctx); } -sub show_other_result ($$) { +sub show_other_result ($$) { # future-proofing my ($bref, $ctx) = @_; - my ($qsp, $logref) = delete @$ctx{qw(-qsp -logref)}; - if (my $err = $qsp->{err}) { - utf8::decode($$err); - $$logref .= "git show error: $err"; - return html_page($ctx, 500, $logref); + if (my $qsp_err = delete $ctx->{-qsp_err}) { + return html_page($ctx, 500, dbg_log($ctx) . + "git show error:$qsp_err"); } my $l = PublicInbox::Linkify->new; - utf8::decode($$bref); - $$bref = '<pre>'. $l->to_html($$bref); - $$bref .= '</pre><hr>' . $$logref; - html_page($ctx, 200, $bref); + utf8_maybe($$bref); + html_page($ctx, 200, '<pre>', $l->to_html($$bref), '</pre><hr>', + dbg_log($ctx)); } -sub show_other ($$$$) { - my ($ctx, $res, $logref, $fn) = @_; - my ($git, $oid, $type, $size) = @$res; - if ($size > $MAX_SIZE) { - $$logref = "$oid is too big to show\n" . $$logref; - return html_page($ctx, 200, $logref); +sub cmt_title { # git->cat_async callback + my ($bref, $oid, $type, $size, $ctx_cb) = @_; + utf8_maybe($$bref); + my $title = $$bref =~ /\r?\n\r?\n([^\r\n]+)\r?\n?/ ? $1 : ''; + # $ctx_cb is [ $ctx, $cmt_fin ] + push @{$ctx_cb->[0]->{-cmt_pt}}, ascii_html($title); +} + +sub do_cat_async { + my ($arg, $cb, @req) = @_; + # favor git(1) over Gcf2 (libgit2) for SHA-256 support + my $ctx = ref $arg eq 'ARRAY' ? $arg->[0] : $arg; + $ctx->{git}->cat_async($_, $cb, $arg) for @req; + if ($ctx->{env}->{'pi-httpd.async'}) { + $ctx->{git}->watch_async; + } else { # synchronous, generic PSGI + $ctx->{git}->cat_async_wait; + } +} + +sub do_check_async { + my ($ctx, $cb, @req) = @_; + if ($ctx->{env}->{'pi-httpd.async'}) { + async_check($ctx, $_, $cb, $ctx) for @req; + } else { # synchronous, generic PSGI + $ctx->{git}->check_async($_, $cb, $ctx) for @req; + $ctx->{git}->check_async_wait; + } +} + +sub cmt_hdr_prep { # psgi_qx cb + my ($fh, $ctx, $cmt_fin) = @_; + return if $ctx->{-qsp_err_h}; # let cmt_fin handle it + seek $fh, 0, SEEK_SET; + my $buf = do { local $/ = "\0"; <$fh> } // die "readline: $!"; + chop($buf) eq "\0" or die 'no NUL in git show -z output'; + utf8_maybe($buf); # non-UTF-8 commits exist + chomp $buf; + (my $P, my $p, @{$ctx->{cmt_info}}) = split(/\n/, $buf, 9); + truncate $fh, 0; + return unless $P; + seek $fh, 0, SEEK_SET; + my $qsp_p = PublicInbox::Qspawn->new($ctx->{git}->cmd(qw(show + --encoding=UTF-8 --pretty=format:%n -M --stat -p), $ctx->{oid}), + undef, { 1 => $fh }); + $qsp_p->{qsp_err} = \($ctx->{-qsp_err_p} = ''); + $qsp_p->psgi_qx($ctx->{env}, undef, \&cmt_patch_prep, $ctx, $cmt_fin); + @{$ctx->{-cmt_P}} = split / /, $P; + @{$ctx->{-cmt_p}} = split / /, $p; # abbreviated + do_cat_async([$ctx, $cmt_fin], \&cmt_title, @{$ctx->{-cmt_P}}); +} + +sub read_patchid { # psgi_qx cb + my ($bref, $ctx, $cmt_fin) = @_; + my ($patchid) = split(/ /, $$bref); # ignore commit + $ctx->{-q_value_html} = "patchid:$patchid" if defined $patchid; +} + +sub cmt_patch_prep { # psgi_qx cb + my ($fh, $ctx, $cmt_fin) = @_; + return if $ctx->{-qsp_err_p}; # let cmt_fin handle error + return if -s $fh > $MAX_SIZE; # too big to show, too big to patch-id + seek $fh, 0, SEEK_SET; + my $qsp = PublicInbox::Qspawn->new( + $ctx->{git}->cmd(qw(patch-id --stable)), + undef, { 0 => $fh }); + $qsp->{qsp_err} = \$ctx->{-qsp_err_p}; + $qsp->psgi_qx($ctx->{env}, undef, \&read_patchid, $ctx, $cmt_fin); +} + +sub ibx_url_for { + my ($ctx) = @_; + $ctx->{ibx} and return; # fall back to $upfx + $ctx->{git} or die 'BUG: no {git}'; + if (my $ALL = $ctx->{www}->{pi_cfg}->ALL) { + if (defined(my $u = $ALL->base_url($ctx->{env}))) { + return wantarray ? ($u) : $u; + } + } + my @ret; + if (my $ibx_names = $ctx->{git}->{ibx_names}) { + my $by_name = $ctx->{www}->{pi_cfg}->{-by_name}; + for my $name (@$ibx_names) { + my $ibx = $by_name->{$name} // do { + warn "inbox `$name' no longer exists\n"; + next; + }; + $ibx->isrch // next; + my $u = defined($ibx->{url}) ? + prurl($ctx->{env}, $ibx->{url}) : $name; + $u .= '/' if substr($u, -1) ne '/'; + push @ret, $u; + } } + wantarray ? (@ret) : $ret[0]; +} + +sub cmt_fin { # OnDestroy cb + my ($ctx) = @_; + my ($eh, $ep) = delete @$ctx{qw(-qsp_err_h -qsp_err_p)}; + if ($eh || $ep) { + my $e = join(' - ', grep defined, $eh, $ep); + return html_page($ctx, 500, dbg_log($ctx) . + "git show/patch-id error:$e"); + } + $ctx->{-linkify} //= PublicInbox::Linkify->new; + my $upfx = $ctx->{-upfx} = '../../'; # from "/$INBOX/$OID/s/" + my ($H, $T, $s, $f, $au, $co, $bdy) = @{delete $ctx->{cmt_info}}; + # try to keep author and committer dates lined up + my $x = length($au) - length($co); + if ($x > 0) { + $x = ' ' x $x; + $co =~ s/>/>$x/; + } elsif ($x < 0) { + $x = ' ' x (-$x); + $au =~ s/>/>$x/; + } + $_ = ascii_html($_) for ($au, $co); + my $ibx_url = ibx_url_for($ctx) // $upfx; + $au =~ s!(> +)([0-9]{4,}-\S+ \S+)! + my ($gt, $t) = ($1, $2); + $t =~ tr/ :-//d; + qq($gt<a +href="$ibx_url?t=$t" +title="list contemporary emails">$2</a>) + !e; + + $ctx->{-title_html} = $s = $ctx->{-linkify}->to_html($s); + my ($P, $p, $pt) = delete @$ctx{qw(-cmt_P -cmt_p -cmt_pt)}; + $_ = qq(<a href="$upfx$_/s/">).shift(@$p).'</a> '.shift(@$pt) for @$P; + if (@$P == 1) { + $x = qq{ (<a +href="$f.patch">patch</a>)\n <a href=#parent>parent</a> $P->[0]}; + } elsif (@$P > 1) { + $x = qq(\n <a href=#parents>parents</a> $P->[0]\n); + shift @$P; + $x .= qq( $_\n) for @$P; + chop $x; + } else { + $x = ' (<a href=#root_commit>root commit</a>)'; + } + PublicInbox::WwwStream::html_init($ctx); + my $zfh = $ctx->zfh; + print $zfh <<EOM; +<pre> <a href=#commit>commit</a> $H$x + <a href=#tree>tree</a> <a href="$upfx$T/s/?b=">$T</a> + author $au +committer $co + +<b>$s</b> +EOM + print $zfh "\n", $ctx->{-linkify}->to_html($bdy) if length($bdy); + undef $bdy; # free memory + my $fh = delete $ctx->{patch_fh}; + if (-s $fh > $MAX_SIZE) { + print $zfh '</pre><hr><pre>patch is too large to show</pre>'; + } else { # prepare flush_diff: + seek $fh, 0, SEEK_SET; + PublicInbox::IO::read_all $fh, -s _, \$x; + utf8_maybe($x); + $ctx->{-apfx} = $ctx->{-spfx} = $upfx; + $x =~ s/\r?\n/\n/gs; + $ctx->{-anchors} = {} if $x =~ /^diff --git /sm; + flush_diff($ctx, \$x); # undefs $x + # TODO: should there be another textarea which attempts to + # search for the exact email which was applied to make this + # commit? + if (my $qry = delete $ctx->{-qry}) { + my $q = ''; + for (@{$qry->{dfpost}}, @{$qry->{dfpre}}) { + # keep blobs as short as reasonable, emails + # are going to be older than what's in git + substr($_, 7, 64, ''); + $q .= "dfblob:$_ "; + } + chop $q; # no trailing SP + local $Text::Wrap::columns = PublicInbox::View::COLS; + local $Text::Wrap::huge = 'overflow'; + $q = wrap('', '', $q); + my $rows = ($q =~ tr/\n/\n/) + 1; + $q = ascii_html($q); + my $ibx_url = ibx_url_for($ctx); + my $alt; + if (defined $ibx_url) { + $alt = " `$ibx_url'"; + $ibx_url =~ m!://! or + substr($ibx_url, 0, 0, '../../../'); + $ibx_url = ascii_html($ibx_url); + } else { + $ibx_url = $upfx; + $alt = ''; + } + print $zfh <<EOM; +</pre><hr><form action="$ibx_url" +id=related><pre>find related emails, including ancestors/descendants/conflicts +<textarea name=q cols=${\PublicInbox::View::COLS} rows=$rows>$q</textarea> +<input type=submit value="search$alt" +/>\t(<a href="${ibx_url}_/text/help/">help</a>)</pre></form> +EOM + } + } + chop($x = <<EOM); +<hr><pre>glossary +-------- +<dfn +id=commit>Commit</dfn> objects reference one tree, and zero or more parents. + +Single <dfn +id=parent>parent</dfn> commits can typically generate a patch in +unified diff format via `git format-patch'. + +Multiple <dfn id=parents>parents</dfn> means the commit is a merge. + +<dfn id=root_commit>Root commits</dfn> have no ancestor. Note that it is +possible to have multiple root commits when merging independent histories. + +Every commit references one top-level <dfn id=tree>tree</dfn> object.</pre> +EOM + delete($ctx->{-wcb})->($ctx->html_done($x)); +} + +sub stream_patch_parse_hdr { # {parse_hdr} for Qspawn + my ($r, $bref, $ctx) = @_; + if (!defined $r) { # sysread error + html_page($ctx, 500, dbg_log($ctx)); + } elsif (index($$bref, "\n\n") >= 0) { + my $eml = bless { hdr => $bref }, 'PublicInbox::Eml'; + my $fn = to_filename($eml->header('Subject') // ''); + $fn = substr($fn // 'PATCH-no-subject', 6); # drop "PATCH-" + return [ 200, [ 'Content-Type', 'text/plain; charset=UTF-8', + 'Content-Disposition', + qq(inline; filename=$fn.patch) ] ]; + } elsif ($r == 0) { + my $log = dbg_log($ctx); + warn "premature EOF on $ctx->{patch_oid} $log"; + return html_page($ctx, 500, $log); + } else { + undef; # bref keeps growing until "\n\n" + } +} + +sub show_patch ($$) { + my ($ctx, $res) = @_; + my ($git, $oid) = @$res; + my @cmd = ('git', "--git-dir=$git->{git_dir}", + qw(format-patch -1 --stdout -C), + "--signature=git format-patch -1 --stdout -C $oid", $oid); + my $qsp = PublicInbox::Qspawn->new(\@cmd); + $ctx->{env}->{'qspawn.wcb'} = $ctx->{-wcb}; + $ctx->{patch_oid} = $oid; + $qsp->psgi_yield($ctx->{env}, undef, \&stream_patch_parse_hdr, $ctx); +} + +sub show_commit ($$) { + my ($ctx, $res) = @_; + return show_patch($ctx, $res) if ($ctx->{fn} // '') =~ /\.patch\z/; + my ($git, $oid) = @$res; + # patch-id needs two passes, and we use the initial show to ensure + # a patch embedded inside the commit message body doesn't get fed + # to patch-id: + open $ctx->{patch_fh}, '+>', "$ctx->{-tmp}/show"; + my $qsp_h = PublicInbox::Qspawn->new($git->cmd('show', $SHOW_FMT, + qw(--encoding=UTF-8 -z --no-notes --no-patch), $oid), + undef, { 1 => $ctx->{patch_fh} }); + $qsp_h->{qsp_err} = \($ctx->{-qsp_err_h} = ''); + my $cmt_fin = on_destroy \&cmt_fin, $ctx; + $ctx->{git} = $git; + $ctx->{oid} = $oid; + $qsp_h->psgi_qx($ctx->{env}, undef, \&cmt_hdr_prep, $ctx, $cmt_fin); +} + +sub show_other ($$) { # just in case... + my ($ctx, $res) = @_; + my ($git, $oid, $type, $size) = @$res; + $size > $MAX_SIZE and return html_page($ctx, 200, + ascii_html($type)." $oid is too big to show\n". dbg_log($ctx)); my $cmd = ['git', "--git-dir=$git->{git_dir}", qw(show --encoding=UTF-8 --no-color --no-abbrev), $oid ]; my $qsp = PublicInbox::Qspawn->new($cmd); - my $env = $ctx->{env}; - $ctx->{-qsp} = $qsp; - $ctx->{-logref} = $logref; - $qsp->psgi_qx($env, undef, \&show_other_result, $ctx); + $qsp->{qsp_err} = \($ctx->{-qsp_err} = ''); + $qsp->psgi_qx($ctx->{env}, undef, \&show_other_result, $ctx); } -# user_cb for SolverGit, called as: user_cb->($result_or_error, $uarg) -sub solve_result { - my ($res, $ctx) = @_; - my ($log, $hints, $fn) = delete @$ctx{qw(log hints fn)}; - - unless (seek($log, 0, 0)) { - warn "seek(log): $!"; - return html_page($ctx, 500, \'seek error'); +sub show_tree_result ($$) { + my ($bref, $ctx) = @_; + if (my $qsp_err = delete $ctx->{-qsp_err}) { + return html_page($ctx, 500, dbg_log($ctx) . + "git ls-tree -z error:$qsp_err"); } - $log = do { local $/; <$log> }; + my @ent = split(/\0/, $$bref); + my $qp = delete $ctx->{qp}; + my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new; + my $pfx = $ctx->{-path} // $qp->{b}; # {-path} is from RepoTree + $$bref = "<pre><a href=#tree>tree</a> $ctx->{tree_oid}"; + # $REPO/tree/$path already sets {-upfx} + my $upfx = $ctx->{-upfx} //= '../../'; + if (defined $pfx) { + $pfx =~ s!/+\z!!s; + if (my $t = $ctx->{-obj}) { + my $t = ascii_html($t); + $$bref .= <<EOM +\n\$ git ls-tree -l $t # shows similar output on the CLI +EOM + } elsif ($pfx eq '') { + $$bref .= " (root)\n"; + } else { + my $x = ascii_html($pfx); + $pfx .= '/'; + $$bref .= qq( <a href=#path>path</a>: $x</a>\n); + } + } else { + $pfx = ''; + $$bref .= qq[ (<a href=#path>path</a> unknown)\n]; + } + my ($x, $m, $t, $oid, $sz, $f, $n, $gitlink); + $$bref .= "\n size name"; + for (@ent) { + ($x, $f) = split(/\t/, $_, 2); + undef $_; + ($m, $t, $oid, $sz) = split(/ +/, $x, 4); + $m = $GIT_MODE{$m} // '?'; + utf8_maybe($f); + $n = ascii_html($f); + if ($m eq 'g') { # gitlink submodule commit + $$bref .= "\ng\t\t$n @ <a\nhref=#g>commit</a>$oid"; + $gitlink = 1; + next; + } + my $q = 'b='.ascii_html(uri_escape_path($pfx.$f)); + if ($m eq 'd') { $n .= '/' } + elsif ($m eq 'x') { $n = "<b>$n</b>" } + elsif ($m eq 'l') { $n = "<i>$n</i>" } + $$bref .= qq(\n$m\t$sz\t<a\nhref="$upfx$oid/s/?$q">$n</a>); + } + $$bref .= dbg_log($ctx); + $$bref .= <<EOM; +<hr><pre>glossary +-------- +<dfn +id=tree>Tree</dfn> objects belong to commits or other tree objects. Trees may +reference blobs, sub-trees, or (rarely) commits of submodules. + +<dfn +id=path>Path</dfn> names are stored in tree objects, but trees do not know +their own path name. A tree's path name comes from their parent tree, +or it is the root tree referenced by a commit object. Thus, this web UI +relies on the `b=' URI parameter as a hint to display the path name. +EOM + + $$bref .= <<EOM if $gitlink; - my $ref = ref($res); +<dfn title="submodule commit" +id=g>Commit</dfn> objects may be stored in trees to reference submodules.</pre> +EOM + chop $$bref; + html_page($ctx, 200, $$bref); +} + +sub show_tree ($$) { # also used by RepoTree + my ($ctx, $res) = @_; + my ($git, $oid, undef, $size) = @$res; + $size > $MAX_SIZE and return html_page($ctx, 200, + "tree $oid is too big to show\n". dbg_log($ctx)); + my $cmd = [ 'git', "--git-dir=$git->{git_dir}", + qw(ls-tree -z -l --no-abbrev), $oid ]; + my $qsp = PublicInbox::Qspawn->new($cmd); + $ctx->{tree_oid} = $oid; + $qsp->{qsp_err} = \($ctx->{-qsp_err} = ''); + $qsp->psgi_qx($ctx->{env}, undef, \&show_tree_result, $ctx); +} + +# returns seconds offset from git TZ offset +sub tz_adj ($) { + my ($tz) = @_; # e.g "-0700" + $tz = int($tz); + my $mm = $tz < 0 ? -$tz : $tz; + $mm = int($mm / 100) * 60 + ($mm % 100); + $mm = $tz < 0 ? -$mm : $mm; + ($mm * 60); +} + +sub show_tag_result { # git->cat_async callback + my ($bref, $oid, $type, $size, $ctx) = @_; + utf8_maybe($$bref); my $l = PublicInbox::Linkify->new; - $log = '<pre>debug log:</pre><hr /><pre>' . - $l->to_html($log) . '</pre>'; + $$bref = $l->to_html($$bref); + $$bref =~ s!^object ([a-f0-9]+)!object <a +href=../../$1/s/>$1</a>!; - $res or return html_page($ctx, 404, \$log); - $ref eq 'ARRAY' or return html_page($ctx, 500, \$log); + $$bref =~ s/^(tagger .*> )([0-9]+) ([\-+]?[0-9]+)/$1.strftime( + '%Y-%m-%d %H:%M:%S', gmtime($2 + tz_adj($3)))." $3"/sme; + # TODO: download link + html_page($ctx, 200, '<pre>', $$bref, '</pre>', dbg_log($ctx)); +} + +sub show_tag ($$) { + my ($ctx, $res) = @_; + my ($git, $oid) = @$res; + $ctx->{git} = $git; + do_cat_async($ctx, \&show_tag_result, $oid); +} + +# user_cb for SolverGit, called as: user_cb->($result_or_error, $uarg) +sub solve_result { + my ($res, $ctx) = @_; + my $hints = delete $ctx->{hints}; + $res or return html_page($ctx, 404, 'Not found', dbg_log($ctx)); + ref($res) eq 'ARRAY' or + return html_page($ctx, 500, 'Internal error', dbg_log($ctx)); my ($git, $oid, $type, $size, $di) = @$res; - return show_other($ctx, $res, \$log, $fn) if $type ne 'blob'; - my $path = to_filename($di->{path_b} // $hints->{path_b} // 'blob'); - my $raw_link = "(<a\nhref=$path>raw</a>)"; + return show_commit($ctx, $res) if $type eq 'commit'; + return show_tree($ctx, $res) if $type eq 'tree'; + return show_tag($ctx, $res) if $type eq 'tag'; + return show_other($ctx, $res) if $type ne 'blob'; + my $fn = $di->{path_b} // $hints->{path_b}; + my $paths = $ctx->{-paths} //= do { + my $path = to_filename($fn // 'blob') // 'blob'; + my $raw_more = qq[(<a\nhref="$path">raw</a>)]; + my @def; + + # XXX not sure if this is the correct wording + if (defined($fn)) { + $raw_more .= qq( +name: ${\ascii_html($fn)} \t # note: path name is non-authoritative<a +href="#pathdef" id=top>(*)</a>); + $def[0] = "<hr><pre\nid=pathdef>" . +'(*) Git path names are given by the tree(s) the blob belongs to. + Blobs themselves have no identifier aside from the hash of its contents.'. +qq(<a\nhref="#top">^</a></pre>); + } + [ $path, $raw_more, @def ]; + }; + $ctx->{-q_value_html} //= do { + my $s = defined($fn) ? 'dfn:'.ascii_html($fn).' ' : ''; + $s.'dfpost:'.substr($oid, 0, 7); + }; + if ($size > $MAX_SIZE) { - return stream_large_blob($ctx, $res, \$log, $fn) if defined $fn; - $log = "<pre><b>Too big to show, download available</b>\n" . - "$oid $type $size bytes $raw_link</pre>" . $log; - return html_page($ctx, 200, \$log); + return stream_large_blob($ctx, $res) if defined $ctx->{fn}; + return html_page($ctx, 200, <<EOM . dbg_log($ctx)); +<pre><b>Too big to show, download available</b> +blob $oid $size bytes $paths->[1]</pre> +EOM } + bless $ctx, 'PublicInbox::WwwStream'; # for DESTROY + $ctx->{git} = $git; + do_cat_async($ctx, \&show_blob, $oid); +} - my $blob = $git->cat_file($oid); - if (!$blob) { # WTF? +sub show_blob { # git->cat_async callback + my ($blob, $oid, $type, $size, $ctx) = @_; + if (!$blob) { my $e = "Failed to retrieve generated blob ($oid)"; - warn "$e ($git->{git_dir})"; - $log = "<pre><b>$e</b></pre>" . $log; - return html_page($ctx, 500, \$log); + warn "$e ($ctx->{git}->{git_dir}) type=$type"; + return html_page($ctx, 500, "<pre><b>$e</b></pre>".dbg_log($ctx)) } my $bin = index(substr($$blob, 0, $BIN_DETECT), "\0") >= 0; - if (defined $fn) { + if (defined $ctx->{fn}) { my $h = [ 'Content-Length', $size, 'Content-Type' ]; push(@$h, ($bin ? 'application/octet-stream' : 'text/plain')); return delete($ctx->{-wcb})->([200, $h, [ $$blob ]]); } - if ($bin) { - $log = "<pre>$oid $type $size bytes (binary)" . - " $raw_link</pre>" . $log; - return html_page($ctx, 200, \$log); - } + my ($path, $raw_more, @def) = @{delete $ctx->{-paths}}; + $bin and return html_page($ctx, 200, + "<pre>blob $oid $size bytes (binary)" . + " $raw_more</pre>".dbg_log($ctx)); # TODO: detect + convert to ensure validity - utf8::decode($$blob); + utf8_maybe($$blob); my $nl = ($$blob =~ s/\r?\n/\n/sg); my $pad = length($nl); - $l->linkify_1($$blob); + ($ctx->{-linkify} //= PublicInbox::Linkify->new)->linkify_1($$blob); my $ok = $hl->do_hl($blob, $path) if $hl; if ($ok) { $blob = $ok; @@ -170,38 +608,63 @@ sub solve_result { } # using some of the same CSS class names and ids as cgit - $log = "<pre>$oid $type $size bytes $raw_link</pre>" . + my $x = "<pre>blob $oid $size bytes $raw_more</pre>" . "<hr /><table\nclass=blob>". - "<tr><td\nclass=linenumbers><pre>" . join('', map { - sprintf("<a id=n$_ href=#n$_>% ${pad}u</a>\n", $_) - } (1..$nl)) . '</pre></td>' . - '<td><pre> </pre></td>'. # pad for non-CSS users - "<td\nclass=lines><pre\nstyle='white-space:pre'><code>" . - $l->linkify_2($$blob) . - '</code></pre></td></tr></table>' . $log; - - html_page($ctx, 200, \$log); + "<tr><td\nclass=linenumbers><pre>"; + # scratchpad in this loop is faster here than `printf $zfh': + $x .= sprintf("<a id=n$_ href=#n$_>% ${pad}u</a>\n", $_) for (1..$nl); + $x .= '</pre></td><td><pre> </pre></td>'. # pad for non-CSS users + "<td\nclass=lines><pre\nstyle='white-space:pre'><code>"; + html_page($ctx, 200, $x, $ctx->{-linkify}->linkify_2($$blob), + '</code></pre></td></tr></table>'.dbg_log($ctx), @def); } -# GET /$INBOX/$GIT_OBJECT_ID/s/ -# GET /$INBOX/$GIT_OBJECT_ID/s/$FILENAME -sub show ($$;$) { - my ($ctx, $oid_b, $fn) = @_; - my $qp = $ctx->{qp}; - my $hints = $ctx->{hints} = {}; +sub start_solver ($) { + my ($ctx) = @_; while (my ($from, $to) = each %QP_MAP) { - defined(my $v = $qp->{$from}) or next; - $hints->{$to} = $v if $v ne ''; + my $v = $ctx->{qp}->{$from} // next; + $ctx->{hints}->{$to} = $v if $v ne ''; } - - $ctx->{'log'} = tmpfile("solve.$oid_b") // die "tmpfile: $!"; - $ctx->{fn} = $fn; + $ctx->{-next_solver} = on_destroy \&next_solver; + ++$solver_nr; + $ctx->{-tmp} = File::Temp->newdir("solver.$ctx->{oid_b}-XXXX", + TMPDIR => 1); + $ctx->{lh} or open $ctx->{lh}, '+>>', "$ctx->{-tmp}/solve.log"; my $solver = PublicInbox::SolverGit->new($ctx->{ibx}, \&solve_result, $ctx); + $solver->{limiter} = $solver_lim; + $solver->{gits} //= [ $ctx->{git} ]; + $solver->{tmp} = $ctx->{-tmp}; # share tmpdir # PSGI server will call this immediately and give us a callback (-wcb) + $solver->solve(@$ctx{qw(env lh oid_b hints)}); +} + +# run the next solver job when done and DESTROY-ed +sub next_solver { + --$solver_nr; + # XXX FIXME: client may've disconnected if it waited a long while + start_solver(shift(@solver_q) // return); +} + +sub may_start_solver ($) { + my ($ctx) = @_; + $solver_lim //= $ctx->{www}->{pi_cfg}->limiter('codeblob'); + if ($solver_nr >= $solver_lim->{max}) { + @solver_q > 128 ? html_page($ctx, 503, 'too busy') + : push(@solver_q, $ctx); + } else { + start_solver($ctx); + } +} + +# GET /$INBOX/$GIT_OBJECT_ID/s/ +# GET /$INBOX/$GIT_OBJECT_ID/s/$FILENAME +sub show ($$;$) { + my ($ctx, $oid_b, $fn) = @_; + @$ctx{qw(oid_b fn)} = ($oid_b, $fn); sub { $ctx->{-wcb} = $_[0]; # HTTP write callback - $solver->solve($ctx->{env}, $ctx->{log}, $oid_b, $hints); + may_start_solver $ctx; }; } |