From 6e22c9a072f159d3aee3192b545efeb822cf5916 Mon Sep 17 00:00:00 2001
From: Nick Clemens <nick@bywatersolutions.com>
Date: Fri, 23 Jun 2023 13:08:31 +0000
Subject: [PATCH] Bug 33270: Add record_strip_nonxml routine to
 Koha::Biblio::Metadata
Content-Type: text/plain; charset=utf-8

This adds a routine that can strip non xml characters form a record.
It is intended for cases where we do not wish to throw an exception,
but rather need to process a record to allow other work to continue

To test:
prove -v t/db_dependent/Koha/Biblio/Metadata.t

Signed-off-by: Sam Lau <samalau@gmail.com>

Signed-off-by: Marcel de Rooy <m.de.rooy@rijksmuseum.nl>
---
 Koha/Biblio/Metadata.pm               | 37 +++++++++++++++++
 t/db_dependent/Koha/Biblio/Metadata.t | 57 ++++++++++++++++++++++++++-
 2 files changed, 93 insertions(+), 1 deletion(-)

diff --git a/Koha/Biblio/Metadata.pm b/Koha/Biblio/Metadata.pm
index b6d532ea40..474049f75e 100644
--- a/Koha/Biblio/Metadata.pm
+++ b/Koha/Biblio/Metadata.pm
@@ -21,6 +21,7 @@ use MARC::File::XML;
 use Scalar::Util qw( blessed );
 
 use C4::Biblio qw( GetMarcFromKohaField );
+use C4::Charset qw( StripNonXmlChars );
 use C4::Items qw( GetMarcItem );
 use Koha::Database;
 use Koha::Exceptions::Metadata;
@@ -133,6 +134,42 @@ sub record {
     return $record;
 }
 
+=head3 record_strip_nonxml
+
+my $record = $metadata->record_strip_nonxml;
+
+This subroutine is intended for cases where we encounter a record that cannot be parsed, but want
+to make a good effort to present the record (for harvesting, deletion, editing) rather than throwing
+an exception
+
+Will return undef if the record cannot be built
+
+=cut
+
+sub record_strip_nonxml {
+
+    my ($self, $params) = @_;
+
+    my $record;
+    my $marcxml_error;
+
+    eval {
+        $record = MARC::Record->new_from_xml(
+            StripNonXmlChars( $self->metadata ), 'UTF-8',
+            $self->schema
+        );
+    };
+    if( $@ ){
+        $marcxml_error = $@;
+        chomp $marcxml_error;
+        warn $marcxml_error;
+        return;
+    }
+
+    $params->{record} = $record;
+    return $self->record( $params );
+}
+
 =head2 Internal methods
 
 =head3 _embed_items
diff --git a/t/db_dependent/Koha/Biblio/Metadata.t b/t/db_dependent/Koha/Biblio/Metadata.t
index d6a5b8b94e..5318ebc23f 100755
--- a/t/db_dependent/Koha/Biblio/Metadata.t
+++ b/t/db_dependent/Koha/Biblio/Metadata.t
@@ -17,7 +17,7 @@
 
 use Modern::Perl;
 
-use Test::More tests => 3;
+use Test::More tests => 4;
 use Test::Exception;
 use Test::Warn;
 
@@ -89,6 +89,61 @@ subtest 'record() tests' => sub {
     $schema->storage->txn_rollback;
 };
 
+subtest 'record_strip_nonxml() tests' => sub {
+
+    plan tests => 5;
+
+    $schema->storage->txn_begin;
+
+    my $title = 'Oranges and' . chr(31) . ' Peaches';
+
+    # Create a valid record
+    my $record = MARC::Record->new();
+    my $field  = MARC::Field->new( '245', '', '', 'a' => $title );
+    $record->append_fields($field);
+    my ($biblio_id) = C4::Biblio::AddBiblio( $record, '' );
+
+    my $metadata = Koha::Biblios->find($biblio_id)->metadata;
+    my $record2  = $metadata->record_strip_nonxml;
+
+    is( ref $record2, 'MARC::Record', 'Method record() returned a MARC::Record object' );
+    is(
+        $record2->field('245')->subfield("a"),
+        "Oranges and Peaches", 'Title in 245$a matches title with control character removed'
+    );
+
+    my $bad_data = $builder->build_object(
+        {
+            class => 'Koha::Biblio::Metadatas',
+            value => { format => 'marcxml', schema => 'MARC21', metadata => 'this_is_not_marcxml' }
+        }
+    );
+
+    warning_like { $record2 = $bad_data->record_strip_nonxml; }
+    qr/parser error : Start tag expected, '<' not found/,
+        'Warning thrown explicitly';
+
+    is(
+        $record2, undef,
+        "record_strip_nonxml returns undef when the record cannot be parsed after removing nonxml characters"
+    );
+
+    my $builder = t::lib::TestBuilder->new;
+    my $item    = $builder->build_sample_item( { biblionumber => $metadata->biblionumber } );
+
+    # Emptied the OpacHiddenItems pref
+    t::lib::Mocks::mock_preference( 'OpacHiddenItems', '' );
+    my ($itemfield) = C4::Biblio::GetMarcFromKohaField('items.itemnumber');
+
+    $record2 = $metadata->record_strip_nonxml( { embed_items => 1 } );
+
+    my @items = $record2->field($itemfield);
+
+    is( scalar @items, 1, "We got back our item" );
+
+    $schema->storage->txn_rollback;
+};
+
 subtest '_embed_items' => sub {
     plan tests => 10;
 
-- 
2.30.2