about summary refs log tree commit homepage
path: root/lib/PublicInbox
diff options
context:
space:
mode:
authorEric Wong <e@80x24.org>2020-11-23 07:05:53 +0000
committerEric Wong <e@80x24.org>2020-11-24 05:03:55 +0000
commitc118d783a22a2080964899266c91c1a1ac69f481 (patch)
tree1d1850f0d06ca1eeb62ce5047f6da0fbf203f897 /lib/PublicInbox
parentf1bdfad51c98825e7e2569aa34db7d3879fa9ef4 (diff)
downloadpublic-inbox-c118d783a22a2080964899266c91c1a1ac69f481.tar.gz
We'll be using this for MiscIdx and pre-generating the necessary
JSON for manifest.js.gz, so make it easier to share code for
generating per-repo JSON entries for grokmirror.
Diffstat (limited to 'lib/PublicInbox')
-rw-r--r--lib/PublicInbox/Git.pm53
-rw-r--r--lib/PublicInbox/ManifestJsGz.pm59
2 files changed, 56 insertions, 56 deletions
diff --git a/lib/PublicInbox/Git.pm b/lib/PublicInbox/Git.pm
index 86343ac9..917fa4a1 100644
--- a/lib/PublicInbox/Git.pm
+++ b/lib/PublicInbox/Git.pm
@@ -14,10 +14,12 @@ use POSIX ();
 use IO::Handle; # ->autoflush
 use Errno qw(EINTR);
 use File::Glob qw(bsd_glob GLOB_NOSORT);
+use File::Spec ();
 use Time::HiRes qw(stat);
 use PublicInbox::Spawn qw(popen_rd);
 use PublicInbox::Tmpfile;
 use Carp qw(croak);
+use Digest::SHA ();
 our @EXPORT_OK = qw(git_unquote git_quote);
 our $PIPE_BUFSIZ = 65536; # Linux default
 our $in_cleanup;
@@ -475,6 +477,57 @@ sub modified ($) {
         $modified || time;
 }
 
+# for grokmirror, which doesn't read gitweb.description
+# templates/hooks--update.sample and git-multimail in git.git
+# only match "Unnamed repository", not the full contents of
+# templates/this--description in git.git
+sub manifest_entry {
+        my ($self, $epoch, $default_desc) = @_;
+        my ($fh, $pid) = $self->popen('show-ref');
+        my $dig = Digest::SHA->new(1);
+        while (read($fh, my $buf, 65536)) {
+                $dig->add($buf);
+        }
+        close $fh;
+        waitpid($pid, 0);
+        return if $?; # empty, uninitialized git repo
+        my $git_dir = $self->{git_dir};
+        my $ent = {
+                fingerprint => $dig->hexdigest,
+                reference => undef,
+                modified => modified($self),
+        };
+        chomp(my $owner = $self->qx('config', 'gitweb.owner'));
+        utf8::decode($owner);
+        $ent->{owner} = $owner eq '' ? undef : $owner;
+        my $desc = '';
+        if (open($fh, '<', "$git_dir/description")) {
+                local $/ = "\n";
+                chomp($desc = <$fh>);
+                utf8::decode($desc);
+        }
+        $desc = 'Unnamed repository' if $desc eq '';
+        if (defined $epoch && $desc =~ /\AUnnamed repository/) {
+                $desc = "$default_desc [epoch $epoch]";
+        }
+        $ent->{description} = $desc;
+        if (open($fh, '<', "$git_dir/objects/info/alternates")) {
+                # n.b.: GitPython doesn't seem to handle comments or C-quoted
+                # strings like native git does; and we don't for now, either.
+                local $/ = "\n";
+                chomp(my @alt = <$fh>);
+
+                # grokmirror only supports 1 alternate for "reference",
+                if (scalar(@alt) == 1) {
+                        my $objdir = "$git_dir/objects";
+                        my $ref = File::Spec->rel2abs($alt[0], $objdir);
+                        $ref =~ s!/[^/]+/?\z!!; # basename
+                        $ent->{reference} = $ref;
+                }
+        }
+        $ent;
+}
+
 1;
 __END__
 =pod
diff --git a/lib/PublicInbox/ManifestJsGz.pm b/lib/PublicInbox/ManifestJsGz.pm
index ab1478af..3d8a38ae 100644
--- a/lib/PublicInbox/ManifestJsGz.pm
+++ b/lib/PublicInbox/ManifestJsGz.pm
@@ -6,15 +6,12 @@ package PublicInbox::ManifestJsGz;
 use strict;
 use v5.10.1;
 use parent qw(PublicInbox::WwwListing);
-use Digest::SHA ();
-use File::Spec ();
 use bytes (); # length
 use PublicInbox::Inbox;
 use PublicInbox::Config;
 use PublicInbox::Git;
 use IO::Compress::Gzip qw(gzip);
 use HTTP::Date qw(time2str);
-*try_cat = \&PublicInbox::Inbox::try_cat;
 
 our $json = PublicInbox::Config::json();
 
@@ -26,21 +23,6 @@ sub url_regexp {
         $ctx->SUPER::url_regexp('publicInbox.grokManifest', 'match=domain');
 }
 
-sub fingerprint ($) {
-        my ($git) = @_;
-        # TODO: convert to qspawn for fairness when there's
-        # thousands of repos
-        my ($fh, $pid) = $git->popen('show-ref');
-        my $dig = Digest::SHA->new(1);
-        while (read($fh, my $buf, 65536)) {
-                $dig->add($buf);
-        }
-        close $fh;
-        waitpid($pid, 0);
-        return if $?; # empty, uninitialized git repo
-        $dig->hexdigest;
-}
-
 sub manifest_add ($$;$$) {
         my ($ctx, $ibx, $epoch, $default_desc) = @_;
         my $url_path = "/$ibx->{name}";
@@ -51,48 +33,13 @@ sub manifest_add ($$;$$) {
         }
         return unless -d $git_dir;
         my $git = PublicInbox::Git->new($git_dir);
-        my $fingerprint = fingerprint($git) or return; # no empty repos
-
-        chomp(my $owner = $git->qx('config', 'gitweb.owner'));
-        chomp(my $desc = try_cat("$git_dir/description"));
-        utf8::decode($owner);
-        utf8::decode($desc);
-        $owner = undef if $owner eq '';
-        $desc = 'Unnamed repository' if $desc eq '';
-
-        # templates/hooks--update.sample and git-multimail in git.git
-        # only match "Unnamed repository", not the full contents of
-        # templates/this--description in git.git
-        if ($desc =~ /\AUnnamed repository/) {
-                $desc = "$default_desc [epoch $epoch]" if defined($epoch);
-        }
-
-        my $reference;
-        chomp(my $alt = try_cat("$git_dir/objects/info/alternates"));
-        if ($alt) {
-                # n.b.: GitPython doesn't seem to handle comments or C-quoted
-                # strings like native git does; and we don't for now, either.
-                my @alt = split(/\n+/, $alt);
-
-                # grokmirror only supports 1 alternate for "reference",
-                if (scalar(@alt) == 1) {
-                        my $objdir = "$git_dir/objects";
-                        $reference = File::Spec->rel2abs($alt[0], $objdir);
-                        $reference =~ s!/[^/]+/?\z!!; # basename
-                }
-        }
+        my $ent = $git->manifest_entry($epoch, $default_desc) or return;
         $ctx->{-abs2urlpath}->{$git_dir} = $url_path;
-        my $modified = $git->modified;
+        my $modified = $ent->{modified};
         if ($modified > ($ctx->{-mtime} // 0)) {
                 $ctx->{-mtime} = $modified;
         }
-        $ctx->{manifest}->{$url_path} = {
-                owner => $owner,
-                reference => $reference,
-                description => $desc,
-                modified => $modified,
-                fingerprint => $fingerprint,
-        };
+        $ctx->{manifest}->{$url_path} = $ent;
 }
 
 sub ibx_entry {