glossary
--------
Commit objects reference one tree, and zero or more parents.
Single parent commits can typically generate a patch in
unified diff format via `git format-patch'.
Multiple parents means the commit is a merge.
Root commits have no ancestor. Note that it is
possible to have multiple root commits when merging independent histories.
Every commit references one top-level tree object.
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 = PublicInbox::OnDestroy->new($$, \&cmt_fin, $ctx);
$ctx->{git} = $git;
$ctx->{oid} = $oid;
$qsp_h->psgi_qx($ctx->{env}, undef, \&cmt_hdr_prep, $ctx, $cmt_fin);
}
sub show_other ($$) { # 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);
$qsp->{qsp_err} = \($ctx->{-qsp_err} = '');
$qsp->psgi_qx($ctx->{env}, undef, \&show_other_result, $ctx);
}
sub show_tree_result ($$) {
my ($bref, $ctx) = @_;
if (my $qsp_err = delete $ctx->{-qsp_err}) {
return html_page($ctx, 500, dbg_log($ctx) .
"git ls-tree -z error:$qsp_err");
}
my @ent = split(/\0/, $$bref);
my $qp = delete $ctx->{qp};
my $l = $ctx->{-linkify} //= PublicInbox::Linkify->new;
my $pfx = $ctx->{-path} // $qp->{b}; # {-path} is from RepoTree
$$bref = "tree $ctx->{tree_oid}";
# $REPO/tree/$path already sets {-upfx}
my $upfx = $ctx->{-upfx} //= '../../';
if (defined $pfx) {
$pfx =~ s!/+\z!!s;
if (my $t = $ctx->{-obj}) {
my $t = ascii_html($t);
$$bref .= <path: $x\n);
}
} else {
$pfx = '';
$$bref .= qq[ (path unknown)\n];
}
my ($x, $m, $t, $oid, $sz, $f, $n, $gitlink);
$$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
$$bref .= "\ng\t\t$n @ commit$oid";
$gitlink = 1;
next;
}
my $q = 'b='.ascii_html(uri_escape_path($pfx.$f));
if ($m eq 'd') { $n .= '/' }
elsif ($m eq 'x') { $n = "$n" }
elsif ($m eq 'l') { $n = "$n" }
$$bref .= qq(\n$m\t$sz\t$n);
}
$$bref .= dbg_log($ctx);
$$bref .= <glossary
--------
Tree objects belong to commits or other tree objects. Trees may
reference blobs, sub-trees, or (rarely) commits of submodules.
Path 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
$$bref .= <Commit objects may be stored in trees to reference submodules.
EOM
chop $$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($$bref);
my $l = PublicInbox::Linkify->new;
$$bref = $l->to_html($$bref);
$$bref =~ s!^object ([a-f0-9]+)!object $1!;
$$bref =~ s/^(tagger .*> )([0-9]+) ([\-+]?[0-9]+)/$1.strftime(
'%Y-%m-%d %H:%M:%S', gmtime($2 + tz_adj($3)))." $3"/sme;
# TODO: download link
html_page($ctx, 200, '', $$bref, '
', dbg_log($ctx));
}
sub show_tag ($$) {
my ($ctx, $res) = @_;
my ($git, $oid) = @$res;
$ctx->{git} = $git;
do_cat_async($ctx, \&show_tag_result, $oid);
}
# user_cb for SolverGit, called as: user_cb->($result_or_error, $uarg)
sub solve_result {
my ($res, $ctx) = @_;
my $hints = delete $ctx->{hints};
$res or return html_page($ctx, 404, 'Not found', dbg_log($ctx));
ref($res) eq 'ARRAY' or
return html_page($ctx, 500, 'Internal error', dbg_log($ctx));
my ($git, $oid, $type, $size, $di) = @$res;
return show_commit($ctx, $res) if $type eq 'commit';
return show_tree($ctx, $res) if $type eq 'tree';
return show_tag($ctx, $res) if $type eq 'tag';
return show_other($ctx, $res) if $type ne 'blob';
my $fn = $di->{path_b} // $hints->{path_b};
my $paths = $ctx->{-paths} //= do {
my $path = to_filename($fn // 'blob') // 'blob';
my $raw_more = qq[(raw)];
my @def;
# XXX not sure if this is the correct wording
if (defined($fn)) {
$raw_more .= qq(
name: ${\ascii_html($fn)} \t # note: path name is non-authoritative(*));
$def[0] = "
" .
'(*) Git path names are given by the tree(s) the blob belongs to.
Blobs themselves have no identifier aside from the hash of its contents.'.
qq(^
);
}
[ $path, $raw_more, @def ];
};
$ctx->{-q_value_html} //= do {
my $s = defined($fn) ? 'dfn:'.ascii_html($fn).' ' : '';
$s.'dfpost:'.substr($oid, 0, 7);
};
if ($size > $MAX_SIZE) {
return stream_large_blob($ctx, $res) if defined $ctx->{fn};
return html_page($ctx, 200, <Too big to show, download available
blob $oid $size bytes $paths->[1]
EOM
}
bless $ctx, 'PublicInbox::WwwStream'; # for DESTROY
$ctx->{git} = $git;
do_cat_async($ctx, \&show_blob, $oid);
}
sub show_blob { # git->cat_async callback
my ($blob, $oid, $type, $size, $ctx) = @_;
if (!$blob) {
my $e = "Failed to retrieve generated blob ($oid)";
warn "$e ($ctx->{git}->{git_dir}) type=$type";
return html_page($ctx, 500, "$e
".dbg_log($ctx))
}
my $bin = index(substr($$blob, 0, $BIN_DETECT), "\0") >= 0;
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, [ $$blob ]]);
}
my ($path, $raw_more, @def) = @{delete $ctx->{-paths}};
$bin and return html_page($ctx, 200,
"blob $oid $size bytes (binary)" .
" $raw_more
".dbg_log($ctx));
# TODO: detect + convert to ensure validity
utf8_maybe($$blob);
my $nl = ($$blob =~ s/\r?\n/\n/sg);
my $pad = length($nl);
($ctx->{-linkify} //= PublicInbox::Linkify->new)->linkify_1($$blob);
my $ok = $hl->do_hl($blob, $path) if $hl;
if ($ok) {
$blob = $ok;
} else {
$$blob = ascii_html($$blob);
}
# using some of the same CSS class names and ids as cgit
my $x = "blob $oid $size bytes $raw_more
" .
"
".
"";
# scratchpad in this loop is faster here than `printf $zfh':
$x .= sprintf("% ${pad}u\n", $_) for (1..$nl);
$x .= ' | | '. # pad for non-CSS users
"";
html_page($ctx, 200, $x, $ctx->{-linkify}->linkify_2($$blob),
'
|
'.dbg_log($ctx), @def);
}
sub start_solver ($) {
my ($ctx) = @_;
while (my ($from, $to) = each %QP_MAP) {
my $v = $ctx->{qp}->{$from} // next;
$ctx->{hints}->{$to} = $v if $v ne '';
}
$ctx->{-next_solver} = PublicInbox::OnDestroy->new($$, \&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
may_start_solver $ctx;
};
}
1;