Z39 Batch ML

Signed-off-by: Bill Erickson <berick@esilibrary.com>

diff --git a/Open-ILS/src/perlmods/lib/OpenILS/Application/Search/Z3950.pm b/Open-ILS/src/perlmods/lib/OpenILS/Application/Search/Z3950.pm
index a217f7b..c508fd5 100644 (file)
--- a/Open-ILS/src/perlmods/lib/OpenILS/Application/Search/Z3950.pm
+++ b/Open-ILS/src/perlmods/lib/OpenILS/Application/Search/Z3950.pm
@@ -666,9 +666,6 @@ sub send_and_queue_bucket_searches {
         my $bre_id = delete $search->{bre_id};
         $search->{limit} = $search_limit;
 
-        $logger->info("z39 firing bucket search " .
-            OpenSRF::Utils::JSON->perl2JSON($search));
-
         # toss it onto the multi-pile
         my $req = $multi_ses->request(
             'open-ils.search.z3950.search_class', $e->authtoken, $search);
@@ -794,12 +791,63 @@ sub compile_bucket_zsearch {
         }
     }
 
+    # To avoid searching on clumped ISBN values and/or ISBNs with extra
+    # junk [e.g. 9781405144094 (hardcover : alk. paper)], send ISBN values 
+    # through a normalizer which returns a clean list of values.  For each 
+    # extracted ISBN, clone the search so that every ISBN is covered.
+    # TODO: do we need similar treatment for other ident fields?
+    for my $bre_id (keys %z_searches) {
+        for my $src (keys %{$z_searches{$bre_id}}) {
+
+            my $blob = $z_searches{$bre_id}{$src};
+
+            if (my $isbn = $blob->{search}{isbn}) {
+                next if $isbn =~ /^[A-Za-z0-9]+$/; # clean and single
+
+                my $clean_str = $e->json_query({
+                    from => ['translate_isbn1013', $isbn]})->[0];
+
+                next unless ref $clean_str and 
+                    $clean_str = $clean_str->{translate_isbn1013};
+
+                my @isbn_list = split(/\s/, $clean_str);
+
+                if (@isbn_list == 1) {
+                    $blob->{search}{isbn} = $isbn_list[0];
+                    next;
+                }
+
+                # create a new container for the set of searches
+                $z_searches{$bre_id}{$src} = [];
+
+                my $cur_blob;
+                for my $idx (0..$#isbn_list) {
+                    my $clean_isbn = $isbn_list[$idx];
+                    if ($idx == 0) {
+                        $cur_blob = $blob;
+                    } else {
+                        # clone
+                        $logger->info("z30 cloning search for clean ISBN $clean_isbn");
+                        $cur_blob = OpenSRF::Utils::JSON->JSON2perl(      
+                            OpenSRF::Utils::JSON->perl2JSON($blob));  
+                    }
+                    $cur_blob->{search}{isbn} = $clean_isbn;
+                    push(@{$z_searches{$bre_id}{$src}}, $cur_blob);
+                    $logger->info("z30 clean ISBN $clean_isbn");
+                }
+            }
+        }
+    }
+
     # let's turn this into something slightly more digestable
     my @searches;
     for my $bre_id (keys %z_searches) {
-        for my $blob (values %{$z_searches{$bre_id}}) {
-            $blob->{bre_id} = $bre_id;
-            push(@searches, $blob);
+        for my $blobset (values %{$z_searches{$bre_id}}) {
+            $blobset = [$blobset] unless ref $blobset eq 'ARRAY';
+            for my $blob (@$blobset) {
+                $blob->{bre_id} = $bre_id;
+                push(@searches, $blob);
+            }
         }
     }