From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net X-Spam-Level: X-Spam-ASN: X-Spam-Status: No, score=-4.0 required=3.0 tests=ALL_TRUSTED,BAYES_00 shortcircuit=no autolearn=ham autolearn_force=no version=3.4.2 Received: from localhost (dcvr.yhbt.net [127.0.0.1]) by dcvr.yhbt.net (Postfix) with ESMTP id 093F11F8C8 for ; Sat, 11 Sep 2021 23:30:47 +0000 (UTC) From: Eric Wong To: meta@public-inbox.org Subject: [PATCH 1/2] manifest.js.gz: avoid long-lived per-epoch cat-file processes Date: Sat, 11 Sep 2021 23:30:45 +0000 Message-Id: <20210911233046.23862-2-e@80x24.org> In-Reply-To: <20210911233046.23862-1-e@80x24.org> References: <20210911233046.23862-1-e@80x24.org> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit List-Id: When generating per-inbox manifests, we were forgetting to cleanup per-epoch "git cat-file --batch" processes. Our previous method of generating modified times was also stupidly inefficient, so replace the pipeline with a single "git for-each-ref" invocation. --- lib/PublicInbox/Git.pm | 25 +++++-------------------- lib/PublicInbox/ManifestJsGz.pm | 2 +- 2 files changed, 6 insertions(+), 21 deletions(-) diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm index e557f6d6..219a1732 100644 --- a/lib/PublicInbox/Git.pm +++ b/lib/PublicInbox/Git.pm @@ -16,7 +16,7 @@ use Errno qw(EINTR EAGAIN); use File::Glob qw(bsd_glob GLOB_NOSORT); use File::Spec (); use Time::HiRes qw(stat); -use PublicInbox::Spawn qw(popen_rd); +use PublicInbox::Spawn qw(popen_rd spawn); use PublicInbox::Tmpfile; use IO::Poll qw(POLLIN); use Carp qw(croak); @@ -466,28 +466,13 @@ sub cat_async ($$$;$) { push(@$inflight, $oid, $cb, $arg); } -sub extract_cmt_time { - my ($bref, undef, undef, undef, $modified) = @_; - - if ($$bref =~ /^committer .*?> ([0-9]+) [\+\-]?[0-9]+/sm) { - my $cmt_time = $1 + 0; - $$modified = $cmt_time if $cmt_time > $$modified; - } -} - # returns the modified time of a git repo, same as the "modified" field # of a grokmirror manifest sub modified ($) { - my ($self) = @_; - my $modified = 0; - my $fh = popen($self, qw(rev-parse --branches)); - local $/ = "\n"; - while (my $oid = <$fh>) { - chomp $oid; - cat_async($self, $oid, \&extract_cmt_time, \$modified); - } - cat_async_wait($self); - $modified || time; + # committerdate:unix is git 2.9.4+ (2017-05-05), so using raw instead + my $fh = popen($_[0], qw[for-each-ref --sort=-committerdate + --format=%(committerdate:raw) --count=1]); + (split(/ /, <$fh> // time))[0] + 0; # integerize for JSON } # for grokmirror, which doesn't read gitweb.description diff --git a/lib/PublicInbox/ManifestJsGz.pm b/lib/PublicInbox/ManifestJsGz.pm index 69d81fa1..cde60245 100644 --- a/lib/PublicInbox/ManifestJsGz.pm +++ b/lib/PublicInbox/ManifestJsGz.pm @@ -27,7 +27,7 @@ sub inject_entry ($$$;$) { $ctx->{manifest}->{$url_path} = $ent; } -sub manifest_add ($$;$$) { # slow path w/o extindex "all" +sub manifest_add ($$;$$) { # slow path w/o extindex "all" (or per-inbox) my ($ctx, $ibx, $epoch, $default_desc) = @_; my $url_path = "/$ibx->{name}"; my $git;