AuthoritiesMarc.pm Improvements
[koha.git] / C4 / Breeding.pm
index cccbdea..79f2e2f 100644 (file)
@@ -19,183 +19,202 @@ package C4::Breeding;
 
 use strict;
 use C4::Biblio;
-use C4::Search;
+use C4::Koha;
+use C4::Charset;
 use MARC::File::USMARC;
-use MARC::Record;
-use Encode;
-require Exporter;
+use C4::ImportBatch;
+
 use vars qw($VERSION @ISA @EXPORT @EXPORT_OK %EXPORT_TAGS);
 
-# set the version for version checking
-$VERSION = 0.01;
+BEGIN {
+       # set the version for version checking
+       $VERSION = 0.02;
+       require Exporter;
+       @ISA = qw(Exporter);
+       @EXPORT = qw(&ImportBreeding &BreedingSearch);
+}
 
 =head1 NAME
 
-C4::Breeding : script to add a biblio in marc_breeding table.
+C4::Breeding : module to add biblios to import_records via
+               the breeding/reservoir API.
 
 =head1 SYNOPSIS
-       &ImportBreeding($marcrecords,$overwrite_biblio,$filename,$z3950random);
 
-       C<$marcrecord> => the MARC::Record
-       C<$overwrite_biblio> => if set to 1 a biblio with the same ISBN will be overwritted.
-                                                               if set to 0 a biblio with the same isbn will be ignored (the previous will be kept)
-                                                               if set to -1 the biblio will be added anyway (more than 1 biblio with the same ISBN possible in the breeding
-       C<$encoding> => USMARC
-                                               or UNIMARC. used for char_decoding.
-                                               If not present, the parameter marcflavour is used instead
-       C<$z3950random> => the random value created during a z3950 search result.
+    use C4::Scan;
+    &ImportBreeding($marcrecords,$overwrite_biblio,$filename,$z3950random,$batch_type);
+
+    C<$marcrecord> => the MARC::Record
+    C<$overwrite_biblio> => if set to 1 a biblio with the same ISBN will be overwritted.
+                                if set to 0 a biblio with the same isbn will be ignored (the previous will be kept)
+                                if set to -1 the biblio will be added anyway (more than 1 biblio with the same ISBN 
+                                possible in the breeding
+    C<$encoding> => USMARC
+                        or UNIMARC. used for char_decoding.
+                        If not present, the parameter marcflavour is used instead
+    C<$z3950random> => the random value created during a z3950 search result.
 
 =head1 DESCRIPTION
 
-This is for depository of records coming from z3950 or directly imported.
+    ImportBreeding import MARC records in the reservoir (import_records/import_batches tables).
+    the records can be properly encoded or not, we try to reencode them in utf-8 if needed.
+    works perfectly with BNF server, that sends UNIMARC latin1 records. Should work with other servers too.
+
+=head2 ImportBreeding
+
+       ImportBreeding($marcrecords,$overwrite_biblio,$filename,$encoding,$z3950random,$batch_type);
+
+       TODO description
 
 =cut
 
-@ISA = qw(Exporter);
-@EXPORT = qw(&ImportBreeding &BreedingSearch);
-
-sub  ImportBreeding {
-       my ($marcrecords,$overwrite_biblio,$filename,$encoding,$z3950random) = @_;
-## use marc:batch send them in one by one
-#      my @marcarray = split /\x1D/, $marcrecords;
-       my $dbh = C4::Context->dbh;
-my @kohafields;
-my @values;
-my @relations;
-my $sort;
-my @and_or;
-my @results;
-my $count;
-       my $searchbreeding = $dbh->prepare("select id from marc_breeding where isbn=? and title=?");
-my $findbreedingid = $dbh->prepare("select max(id) from marc_breeding");
-
-       my $insertsql = $dbh->prepare("insert into marc_breeding (file,isbn,title,author,marc,encoding,z3950random,classification,subclass) values(?,?,?,?,?,?,?,?,?)");
-       my $replacesql = $dbh->prepare("update marc_breeding set file=?,isbn=?,title=?,author=?,marc=?,encoding=?,z3950random=?,classification=?,subclass=? where id=?");
-       $encoding = C4::Context->preference("marcflavour") unless $encoding;
-       # fields used for import results
-       my $imported=0;
-       my $alreadyindb = 0;
-       my $alreadyinfarm = 0;
-       my $notmarcrecord = 0;
-       my $breedingid;
-#      for (my $i=0;$i<=$#marcarray;$i++) {
-               my $marcrecord = MARC::File::USMARC::decode($marcrecords);
-               my $marcxml=$marcrecord->as_xml_record($marcrecord);
-               $marcxml=Encode::encode('utf8',$marcxml);
-               my @warnings = $marcrecord->warnings();
-               if (scalar($marcrecord->fields()) == 0) {
-                       $notmarcrecord++;
-               } else {
-                       my $xmlhash=XML_xml2hash_onerecord($marcxml);   
-                       my $oldbiblio = XMLmarc2koha_onerecord($dbh,$xmlhash,'biblios');
-                       # if isbn found and biblio does not exist, add it. If isbn found and biblio exists, overwrite or ignore depending on user choice
-                       # drop every "special" char : spaces, - ...
-                       $oldbiblio->{isbn} =~ s/ |-|\.//g,
-                       $oldbiblio->{isbn} = substr($oldbiblio->{isbn},0,10);
-                       $oldbiblio->{issn} =~ s/ |-|\.//g,
-                       $oldbiblio->{issn} = substr($oldbiblio->{issn},0,10);
-                       # search if biblio exists
-                       my $biblioitemnumber;
-                       my $facets;
-                   if ( !$z3950random){
-                       if ($oldbiblio->{isbn}) {
-                       push @kohafields,"isbn";
-                       push @values,$oldbiblio->{isbn};
-                       push @relations,"";
-                       push @and_or,"";
-                       
-                       ($count,$facets,@results)=ZEBRAsearch_kohafields(\@kohafields,\@values,\@relations);
-                       } else {
-                       push @kohafields,"issn";
-                       push @values,$oldbiblio->{issn};
-                       push @relations,"";
-                       push @and_or,"";
-                       $sort="";
-                       ($count,$facets,@results)=ZEBRAsearch_kohafields(\@kohafields,\@values,\@relations);
-                       }
-                    }
-                       if ($count>0 && !$z3950random) {
-                               $alreadyindb++;
-                       } else {
-                               # search in breeding farm
-                               
-                               if ($oldbiblio->{isbn}) {
-                                       $searchbreeding->execute($oldbiblio->{isbn},$oldbiblio->{title});
-                                       ($breedingid) = $searchbreeding->fetchrow;
-                               } elsif ($oldbiblio->{issn}){
-                                       $searchbreeding->execute($oldbiblio->{issn},$oldbiblio->{title});
-                                       ($breedingid) = $searchbreeding->fetchrow;
-                               }
-                               if ($breedingid && $overwrite_biblio eq 0) {
-                                       $alreadyinfarm++;
-                               } else {
-                                       my $recoded=MARC::Record->new_from_xml($marcxml,"UTF-8");
-                                       $recoded->encoding('UTF-8');
-                                       
-                                       if ($breedingid && $overwrite_biblio eq 1) {
-                                               $replacesql ->execute($filename,substr($oldbiblio->{isbn}.$oldbiblio->{issn},0,10),$oldbiblio->{title},$oldbiblio->{author},$recoded->as_usmarc,$encoding,$z3950random,$oldbiblio->{classification},$oldbiblio->{subclass},$breedingid);
-                                       } else {
-                                               $insertsql ->execute($filename,substr($oldbiblio->{isbn}.$oldbiblio->{issn},0,10),$oldbiblio->{title},$oldbiblio->{author},$recoded->as_usmarc,$encoding,$z3950random,$oldbiblio->{classification},$oldbiblio->{subclass});
-                                       $findbreedingid->execute;
-                                       $breedingid=$findbreedingid->fetchrow;
-                                       }
-                                       $imported++;
-                               }
-                       }
-               }
-       #}
-       return ($notmarcrecord,$alreadyindb,$alreadyinfarm,$imported,$breedingid);
+sub ImportBreeding {
+    my ($marcrecords,$overwrite_biblio,$filename,$encoding,$z3950random,$batch_type) = @_;
+    my @marcarray = split /\x1D/, $marcrecords;
+    
+    my $dbh = C4::Context->dbh;
+    
+    my $batch_id = 0;
+    if ($batch_type eq 'z3950') {
+        $batch_id = GetZ3950BatchId($filename);
+    } else {
+        # create a new one
+        $batch_id = AddImportBatch('create_new', 'staging', 'batch', $filename, '');
+    }
+    my $searchisbn = $dbh->prepare("select biblioitemnumber from biblioitems where isbn=?");
+    my $searchissn = $dbh->prepare("select biblioitemnumber from biblioitems where issn=?");
+    # FIXME -- not sure that this kind of checking is actually needed
+    my $searchbreeding = $dbh->prepare("select import_record_id from import_biblios where isbn=? and title=?");
+    
+#     $encoding = C4::Context->preference("marcflavour") unless $encoding;
+    # fields used for import results
+    my $imported=0;
+    my $alreadyindb = 0;
+    my $alreadyinfarm = 0;
+    my $notmarcrecord = 0;
+    my $breedingid;
+    for (my $i=0;$i<=$#marcarray;$i++) {
+        my ($marcrecord, $charset_result, $charset_errors);
+        ($marcrecord, $charset_result, $charset_errors) = 
+            MarcToUTF8Record($marcarray[$i]."\x1D", C4::Context->preference("marcflavour"), $encoding);
+        
+#         warn "$i : $marcarray[$i]";
+        # FIXME - currently this does nothing 
+        my @warnings = $marcrecord->warnings();
+        
+        if (scalar($marcrecord->fields()) == 0) {
+            $notmarcrecord++;
+        } else {
+            my $oldbiblio = TransformMarcToKoha($dbh,$marcrecord,'');
+            # if isbn found and biblio does not exist, add it. If isbn found and biblio exists, 
+            # overwrite or ignore depending on user choice
+            # drop every "special" char : spaces, - ...
+            $oldbiblio->{isbn} =~ s/\(.*$//;
+            $oldbiblio->{isbn} =~ tr/ -_//;
+            $oldbiblio->{isbn} = uc $oldbiblio->{isbn}; 
+            # search if biblio exists
+            my $biblioitemnumber;
+            if ($oldbiblio->{isbn}) {
+                $searchisbn->execute($oldbiblio->{isbn});
+                ($biblioitemnumber) = $searchisbn->fetchrow;
+            } else {
+                if ($oldbiblio->{issn}) {
+                    $searchissn->execute($oldbiblio->{issn});
+                       ($biblioitemnumber) = $searchissn->fetchrow;
+                }
+            }
+            if ($biblioitemnumber && $overwrite_biblio ne 2) {
+                $alreadyindb++;
+            } else {
+                # FIXME - in context of batch load,
+                # rejecting records because already present in the reservoir
+                # not correct in every case.
+                # search in breeding farm
+                if ($oldbiblio->{isbn}) {
+                    $searchbreeding->execute($oldbiblio->{isbn},$oldbiblio->{title});
+                    ($breedingid) = $searchbreeding->fetchrow;
+                } elsif ($oldbiblio->{issn}){
+                    $searchbreeding->execute($oldbiblio->{issn},$oldbiblio->{title});
+                    ($breedingid) = $searchbreeding->fetchrow;
+                }
+                if ($breedingid && $overwrite_biblio eq '0') {
+                    $alreadyinfarm++;
+                } else {
+                    if ($breedingid && $overwrite_biblio eq '1') {
+                        ModBiblioInBatch($breedingid, $marcrecord);
+                    } else {
+                        my $import_id = AddBiblioToBatch($batch_id, $imported, $marcrecord, $encoding, $z3950random);
+                        $breedingid = $import_id;
+                    }
+                    $imported++;
+                }
+            }
+        }
+    }
+    return ($notmarcrecord,$alreadyindb,$alreadyinfarm,$imported,$breedingid);
 }
 
 
-=item BreedingSearch
+=head2 BreedingSearch
 
-  ($count, @results) = &BreedingSearch($title,$isbn,$random);
+($count, @results) = &BreedingSearch($title,$isbn,$random);
 C<$title> contains the title,
 C<$isbn> contains isbn or issn,
 C<$random> contains the random seed from a z3950 search.
 
 C<$count> is the number of items in C<@results>. C<@results> is an
-array of references-to-hash; the keys are the items from the C<marc_breeding> table of the Koha database.
+array of references-to-hash; the keys are the items from the C<import_records> and
+C<import_biblios> tables of the Koha database.
 
 =cut
 
 sub BreedingSearch {
-       my ($title,$isbn,$z3950random) = @_;
-       my $dbh   = C4::Context->dbh;
-       my $count = 0;
-       my ($query,@bind);
-       my $sth;
-       my @results;
-
-       $query = "Select id,file,isbn,title,author,classification,subclass from marc_breeding where ";
-       if ($z3950random) {
-               $query .= "z3950random = ?";
-               @bind=($z3950random);
-       } else {
-           @bind=();
-               if ($title) {
-                       $query .= "title like ?";
-                       push(@bind,"$title%");
-               }
-               if ($title && $isbn) {
-                       $query .= " and ";
-               }
-               if ($isbn) {
-                       $query .= "isbn like ?";
-                       push(@bind,"$isbn%");
-               }
-       }
-       $sth   = $dbh->prepare($query);
-       $sth->execute(@bind);
-       while (my $data = $sth->fetchrow_hashref) {
-                       $results[$count] = $data;
-                       $count++;
-       } # while
-
-       $sth->finish;
-       return($count, @results);
+    my ($title,$isbn,$z3950random) = @_;
+    my $dbh   = C4::Context->dbh;
+    my $count = 0;
+    my ($query,@bind);
+    my $sth;
+    my @results;
+
+    $query = "SELECT import_record_id, file_name, isbn, title, author
+              FROM  import_biblios 
+              JOIN import_records USING (import_record_id)
+              JOIN import_batches USING (import_batch_id)
+              WHERE ";
+    if ($z3950random) {
+        $query .= "z3950random = ?";
+        @bind=($z3950random);
+    } else {
+        @bind=();
+        if ($title) {
+            $query .= "title like ?";
+            push(@bind,"$title%");
+        }
+        if ($title && $isbn) {
+            $query .= " and ";
+        }
+        if ($isbn) {
+            $query .= "isbn like ?";
+            push(@bind,"$isbn%");
+        }
+    }
+    $sth   = $dbh->prepare($query);
+    $sth->execute(@bind);
+    while (my $data = $sth->fetchrow_hashref) {
+            $results[$count] = $data;
+            # FIXME - hack to reflect difference in name 
+            # of columns in old marc_breeding and import_records
+            # There needs to be more separation between column names and 
+            # field names used in the templates </soapbox>
+            $data->{'file'} = $data->{'file_name'};
+            $data->{'id'} = $data->{'import_record_id'};
+            $count++;
+    } # while
+
+    $sth->finish;
+    return($count, @results);
 } # sub breedingsearch
 
+1;
+__END__
 
-END { }       # module clean-up code here (global destructor)