From 4fd847b7ae12f4efdbf8c4740fd20a7dccee0849 Mon Sep 17 00:00:00 2001
From: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Date: Wed, 31 May 2023 11:44:26 +0000
Subject: [PATCH] Bug 33871: Use Koha object in Sitemapper, add optional filter

Test plan:
Run misc/cronjobs/sitemap.pl -where "biblionumber>X" (replace X
by some clever value)
Verify results by browsing sitemap files.

Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
Signed-off-by: Pedro Amorim <pedro.amorim@ptfs-europe.com>

Signed-off-by: Jonathan Druart <jonathan.druart@bugs.koha-community.org>
Signed-off-by: Tomas Cohen Arazi <tomascohen@theke.io>
---
 Koha/Sitemapper.pm       | 23 +++++++++++------------
 misc/cronjobs/sitemap.pl | 11 +++++++++--
 2 files changed, 20 insertions(+), 14 deletions(-)

diff --git a/Koha/Sitemapper.pm b/Koha/Sitemapper.pm
index eab42f35ba..6c2520781d 100644
--- a/Koha/Sitemapper.pm
+++ b/Koha/Sitemapper.pm
@@ -18,10 +18,11 @@ package Koha::Sitemapper;
 # You should have received a copy of the GNU General Public License
 # along with Koha; if not, see <http://www.gnu.org/licenses>.
 
-use Moo;
 use Modern::Perl;
+use Moo;
+
+use Koha::Biblios;
 use Koha::Sitemapper::Writer;
-use C4::Context;
 
 
 has url => ( is => 'rw', );
@@ -50,18 +51,16 @@ has count => ( is => 'rw', default => sub { 0 } );
 
 
 sub run {
-    my $self = shift;
+    my ( $self, $where ) = @_;
+    my $filter = $where ? \$where : {};
 
     say "Creation of Sitemap files in '" . $self->dir . "' directory"
         if $self->verbose;
 
     $self->writer( Koha::Sitemapper::Writer->new( sitemapper => $self ) );
-    my $sth = C4::Context->dbh->prepare(
-         "SELECT biblionumber, timestamp FROM biblio" );
-    $sth->execute();
-    $self->sth($sth);
+    my $rs = Koha::Biblios->search( $filter, { columns => [ qw/biblionumber timestamp/ ] });
 
-    while ( $self->process() ) {
+    while ( $self->process($rs) ) {
         say "..... ", $self->count
             if $self->verbose && $self->count % 10000 == 0;
     }
@@ -69,10 +68,10 @@ sub run {
 
 
 sub process {
-    my $self = shift;
+    my ( $self, $rs ) = @_;
 
-    my ($biblionumber, $timestamp) = $self->sth->fetchrow;
-    unless ($biblionumber) {
+    my $biblio = $rs->next;
+    unless( $biblio ) {
         $self->writer->end();
         say "Number of biblio records processed: ", $self->count, "\n" .
             "Number of Sitemap files:            ", $self->writer->count
@@ -80,7 +79,7 @@ sub process {
         return;
     }
 
-    $self->writer->write($biblionumber, $timestamp);
+    $self->writer->write( $biblio->biblionumber, $biblio->timestamp );
     $self->count( $self->count + 1 );
     return $self->count;
 }
diff --git a/misc/cronjobs/sitemap.pl b/misc/cronjobs/sitemap.pl
index 93abb02ed0..4b4576a567 100755
--- a/misc/cronjobs/sitemap.pl
+++ b/misc/cronjobs/sitemap.pl
@@ -29,12 +29,14 @@ use Koha::Sitemapper;
 
 
 my ($verbose, $help, $url, $dir, $short) = (0, 0, '', '.', 1);
+my $where;
 GetOptions(
     'verbose'   => \$verbose,
     'help'      => \$help,
     'url=s'     => \$url,
     'dir=s'     => \$dir,
     'short!'    => \$short,
+    'where=s'   => \$where,
 );
 
 sub usage {
@@ -59,14 +61,14 @@ my $sitemapper = Koha::Sitemapper->new(
     dir     => $dir,
     short   => $short,
 );
-$sitemapper->run();
+$sitemapper->run($where);
 
 
 =head1 USAGE
 
 =over
 
-=item sitemap.pl [--verbose|--help|--short|--noshort|--url|--dir]
+=item sitemap.pl [--verbose|--help|--short|--noshort|--url|--dir|--where ]
 
 =back
 
@@ -75,6 +77,7 @@ $sitemapper->run();
   sitemap.pl --verbose
   sitemap.pl --noshort --dir /home/koha/mylibrary/www
   sitemap.pl --url opac.myDNSname.org
+  sitemap.pl --where 'biblionumber<100'
 
 =head1 DESCRIPTION
 
@@ -123,6 +126,10 @@ records processed.
 
 Print this help page.
 
+=item B<--where>
+
+Add a filter to limit the selection of biblio records. May be useful when testing the feature.
+
 =back
 
 =cut
-- 
2.39.2