Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
78 changes: 78 additions & 0 deletions bin/mirrors.pl
Original file line number Diff line number Diff line change
@@ -0,0 +1,78 @@
use strict;
use warnings;
use v5.36;

use Getopt::Long;
use MetaCPAN::Logger qw< :log :dlog >;
use Cpanel::JSON::XS ();

use MetaCPAN::ES;
use MetaCPAN::Ingest qw< cpan_dir >;

# args
my ( $distribution, $files_only, $undo );
GetOptions(
"distribution=s" => \$distribution,
"files_only" => \$files_only,
"undo" => \$undo,
);

# setup
my $cpan = cpan_dir();
my $es = MetaCPAN::ES->new( type => "mirror" );


index_mirrors();

$es->index_refresh;

# TODO:
# cdn_purge_now( { keys => ['MIRRORS'], } );

log_info {"done"};

###

sub index_mirrors () {
log_info { 'Getting mirrors.json file from ' . $cpan };

my $json = $cpan->child( 'indices', 'mirrors.json' )->slurp;

# Clear out everything in the index
# so don't end up with old mirrors
$es->clear_type;

my $mirrors = Cpanel::JSON::XS::decode_json($json);
foreach my $mirror (@$mirrors) {
$mirror->{location} = {
lon => delete $mirror->{longitude},
lat => delete $mirror->{latitude}
};

#Dlog_trace {"Indexing $_"} $mirror;
log_debug {sprintf("Indexing %s", $mirror->{name})};

my @doc =
map { $_ => $mirror->{$_} }
grep { defined $mirror->{$_} }
keys %$mirror;

$es->index( body => { @doc } );
}
}

1;

__END__

=pod

=head1 SYNOPSIS

$ bin/mirrors.pl

=head1 SOURCE

L<http://www.cpan.org/indices/mirrors.json>

=cut
22 changes: 22 additions & 0 deletions lib/MetaCPAN/ES.pm
Original file line number Diff line number Diff line change
Expand Up @@ -4,6 +4,7 @@ use strict;
use warnings;
use v5.36;

use MetaCPAN::Logger qw< :log :dlog >;
use Search::Elasticsearch;

use MetaCPAN::Ingest qw< config >;
Expand Down Expand Up @@ -108,4 +109,25 @@ sub count ( $self, %args ) {
);
}

sub clear_type ( $self ) {
my $bulk = $self->bulk;
my $scroll = $self->scroll(
query => { match_all => {} },
sort => '_doc',
);

my @ids;
while ( my $search = $scroll->next ) {
push @ids => $search->{_id};
log_debug { "deleting id=" . $search->{_id} };
if ( @ids == 500 ) {
$bulk->delete_ids(@ids);
@ids = ();
}
}
$bulk->delete_ids(@ids);

$bulk->flush;
}

1;