package PublicInbox::ViewVCS;
use strict;
use warnings;
-use Encode qw(find_encoding);
+use bytes (); # only for bytes::length
use PublicInbox::SolverGit;
use PublicInbox::WwwStream;
use PublicInbox::Linkify;
+use PublicInbox::Tmpfile;
use PublicInbox::Hval qw(ascii_html to_filename);
my $hl = eval {
require PublicInbox::HlMod;
my %QP_MAP = ( A => 'oid_a', B => 'oid_b', a => 'path_a', b => 'path_b' );
my $max_size = 1024 * 1024; # TODO: configurable
-my $enc_utf8 = find_encoding('UTF-8');
+my $BIN_DETECT = 8000; # same as git
sub html_page ($$$) {
my ($ctx, $code, $strref) = @_;
my ($nr, undef) = @_;
$nr == 1 ? $$strref : undef;
});
- $wcb->($res);
+ $wcb ? $wcb->($res) : $res;
+}
+
+sub stream_large_blob ($$$$) {
+ my ($ctx, $res, $logref, $fn) = @_;
+ my ($git, $oid, $type, $size, $di) = @$res;
+ my $cmd = ['git', "--git-dir=$git->{git_dir}", 'cat-file', $type, $oid];
+ my $qsp = PublicInbox::Qspawn->new($cmd);
+ my @cl = ('Content-Length', $size);
+ my $env = $ctx->{env};
+ $env->{'public-inbox.tmpgit'} = $git; # for {-tmp}/File::Temp::Dir
+ $env->{'qspawn.wcb'} = delete $ctx->{-wcb};
+ $qsp->psgi_return($env, undef, sub {
+ my ($r, $bref) = @_;
+ if (!defined $r) { # error
+ html_page($ctx, 500, $logref);
+ } elsif (index($$bref, "\0") >= 0) {
+ my $ct = 'application/octet-stream';
+ [200, ['Content-Type', $ct, @cl ] ];
+ } else {
+ my $n = bytes::length($$bref);
+ if ($n >= $BIN_DETECT || $n == $size) {
+ my $ct = 'text/plain; charset=UTF-8';
+ return [200, ['Content-Type', $ct, @cl] ];
+ }
+ if ($r == 0) {
+ warn "premature EOF on $oid $$logref\n";
+ return html_page($ctx, 500, $logref);
+ }
+ undef; # bref keeps growing
+ }
+ });
+}
+
+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" . $$logref;
+ return html_page($ctx, 200, $logref);
+ }
+ my $cmd = ['git', "--git-dir=$git->{git_dir}",
+ qw(show --encoding=UTF-8 --no-color --no-abbrev), $oid ];
+ my $qsp = PublicInbox::Qspawn->new($cmd);
+ my $env = $ctx->{env};
+ $qsp->psgi_qx($env, undef, sub {
+ my ($bref) = @_;
+ if (my $err = $qsp->{err}) {
+ utf8::decode($$err);
+ $$logref .= "git show error: $err";
+ return html_page($ctx, 500, $logref);
+ }
+ my $l = PublicInbox::Linkify->new;
+ utf8::decode($$bref);
+ $l->linkify_1($$bref);
+ $$bref = '<pre>'. $l->linkify_2(ascii_html($$bref));
+ $$bref .= '</pre><hr>' . $$logref;
+ html_page($ctx, 200, $bref);
+ });
}
sub solve_result {
$ref eq 'ARRAY' or return html_page($ctx, 500, \$log);
my ($git, $oid, $type, $size, $di) = @$res;
+ return show_other($ctx, $res, \$log, $fn) if $type ne 'blob';
+ my $path = to_filename($di->{path_b} || $hints->{path_b} || 'blob');
+ my $raw_link = "(<a\nhref=$path>raw</a>)";
if ($size > $max_size) {
- # TODO: stream the raw file if it's gigantic, at least
- $log = '<pre><b>Too big to show</b></pre>' . $log;
+ return stream_large_blob($ctx, $res, \$log, $fn) if defined $fn;
+ $log = "<pre><b>Too big to show, download available</b>\n" .
+ "$oid $type $size bytes $raw_link</pre>" . $log;
return html_page($ctx, 500, \$log);
}
return html_page($ctx, 500, \$log);
}
- my $binary = index($$blob, "\0") >= 0;
- if ($fn) {
+ my $bin = index(substr($$blob, 0, $BIN_DETECT), "\0") >= 0;
+ if (defined $fn) {
my $h = [ 'Content-Length', $size, 'Content-Type' ];
- push(@$h, ($binary ? 'application/octet-stream' : 'text/plain'));
+ push(@$h, ($bin ? 'application/octet-stream' : 'text/plain'));
return delete($ctx->{-wcb})->([200, $h, [ $$blob ]]);
}
- my $path = to_filename($di->{path_b} || $hints->{path_b} || 'blob');
- my $raw_link = "(<a\nhref=$path>raw</a>)";
- if ($binary) {
+ if ($bin) {
$log = "<pre>$oid $type $size bytes (binary)" .
" $raw_link</pre>" . $log;
return html_page($ctx, 200, \$log);
}
- $$blob = $enc_utf8->decode($$blob);
+ # TODO: detect + convert to ensure validity
+ utf8::decode($$blob);
my $nl = ($$blob =~ tr/\n/\n/);
my $pad = length($nl);
$hints->{$to} = $v;
}
- open my $log, '+>', undef or die "open: $!";
+ my $log = tmpfile("solve.$oid_b");
my $solver = PublicInbox::SolverGit->new($ctx->{-inbox}, sub {
solve_result($ctx, $_[0], $log, $hints, $fn);
});