JBAS-1437 Auth2bib links honors authority 008 #14/#15

Only link 1XX and 7XX bib fields to authority records whose 008 #14
field is 'a'.

Only link 6XX bib fields to authority records whose 008 #15 field is 'a'.

https://www.loc.gov/marc/authority/ad008.html

Signed-off-by: Bill Erickson <berickxx@gmail.com>

diff --git a/KCLS/linking/authority_control_fields.pl b/KCLS/linking/authority_control_fields.pl
index 1c0adf4..414ad7f 100755 (executable)
--- a/KCLS/linking/authority_control_fields.pl
+++ b/KCLS/linking/authority_control_fields.pl
@@ -592,11 +592,44 @@ if($input_file) {
        announce("Start $start_time for all records");
 }
 
+# Fetch leader/008 values for authority records.  Filter out any whose
+# 008 14 or 15 field are not appropriate for the requested bib tag.
+# https://www.loc.gov/marc/authority/ad008.html
+sub authority_leaders_008_14_15 {
+    my ($e, $bib_tag, $auth_ids) = @_;
+
+    my $auth_leaders = $e->json_query({
+        select => {afr => ['record', 'value']},
+        from => 'afr',
+        where => {'+afr' => {tag => '008', record => $auth_ids}}
+    });
+
+    my $index;
+    $index = 14 if $bib_tag =~ /^[17]/; # author/name record
+    $index = 15 if $bib_tag =~ /^6/; # subject record
+
+    # avoid checking any other types of authority records.
+    return $auth_leaders unless $index;
+
+    my @keepers;
+    for my $leader (@$auth_leaders) {
+        my $value = $leader->{value} || '';
+        if (substr($value, $index, 1) eq 'a') {
+            push(@keepers, $leader);
+        } else {
+            announce("Skipping authority record ".$leader->{record}.
+                " on bib $bib_tag match; 008/#14|#15 not appropriate");
+        }
+    }
+
+    return \@keepers;
+}
+
 # given a set of authority record ID's and a controlled bib field,
 # returns the ID of the first authority record in the set that 
 # matches the thesaurus spec of the bib record.
 sub find_matching_auth_for_thesaurus {
-    my ($e, $bib_field, $auth_ids) = @_;
+    my ($e, $bib_field, $auth_leaders) = @_;
 
     # bib field thesaurus spec
     my $cfield_ind2 = $bib_field->indicator(2);
@@ -620,12 +653,6 @@ sub find_matching_auth_for_thesaurus {
             "remapped to ind2 value '$cfield_ind2'");
     }
 
-    my $auth_leaders = $e->json_query({
-        select => {afr => ['record', 'value']},
-        from => 'afr',
-        where => {'+afr' => {tag => '008', record => $auth_ids}}
-    });
-
     my $authz_found = undef;
     for my $leader (@$auth_leaders) {
         my $value = $leader->{value};
@@ -731,6 +758,19 @@ foreach my $rec_id (@records) {
 
                 # Find the best authority record to use for linking.
 
+                my $auth_leaders;
+                if ($bib_field->tag =~ /^[167]/) {
+                    # For 1XX, 6XX, and 7XX bib fields, only link to 
+                    # authority records whose leader/008 positions
+                    # 14 and 15 are coded to allow use as a name/author 
+                    # or subject record, depending.
+
+                    $auth_leaders = authority_leaders_008_14_15(
+                        $e, $bib_field->tag, $validates);
+
+                    $validates = [map {$_->{record}} @$auth_leaders];
+                }
+
                 my $auth_id;
 
                 if ($bib_field->tag() =~ /^65[015]/) {
@@ -742,9 +782,10 @@ foreach my $rec_id (@records) {
                     # validation API search call above.
                    
                     $auth_id = find_matching_auth_for_thesaurus(
-                        $e, $bib_field, $validates) || '';
+                        $e, $bib_field, $auth_leaders) || '';
 
                 } else {
+
                     # For all other controlled fields, use the first 
                     # authority record in the result set.
                     $auth_id = $validates->[0];