Merge pull request #3 from jorol/issue2

This closes issue #2. For customization of PICA::Writer::Plus I created issue #4.

lib/Catmandu/Importer/PICA.pm

@@ -71,24 +71,18 @@ Parse PICA XML to native Perl hash containing two keys: '_id' and 'record'.
                   [
                     '001@',
                     '',
-                    '_',
-                    '',
                     '0',
                     '703'
                   ],
                   [
                     '001A',
                     '',
-                    '_',
-                    '',
                     '0',
                     '2045:10-03-11'
                   ],
                   [
                     '028B',
                     '01',
-                    '_',
-                    '',
                     'd',
                     'Thomas',
                     'a',

lib/PICA/Parser/Plus.pm

@@ -35,7 +35,7 @@ sub new {
     my $self = {
         filename   => undef,
         rec_number => 0,
-        xml_reader => undef,
+        reader => undef,
     };
 
     # check for file or filehandle
@@ -87,7 +87,8 @@ sub _decode {
     my @record;
 
     if (length($fields[0]) == LEADER_LEN-1 && $fields[0] !~ m/.*SUBFIELD_INDICATOR/){
-        push( @record, [ 'LDR', undef, undef, shift(@fields) ] );
+        # drop leader because usage is unclear
+        shift(@fields);
     }
 
     for my $field (@fields) {
@@ -103,7 +104,7 @@ sub _decode {
         }
         my @subfields = map { substr( $_, 0, 1 ), substr( $_, 1 ) }
             split( SUBFIELD_INDICATOR, substr( $data, 1 ) );
-        push( @record, [ $tag, $occurence, '_', '', @subfields ] );
+        push( @record, [ $tag, $occurence, @subfields ] );
     }
     return \@record;
 }

lib/PICA/Parser/XML.pm

@@ -87,7 +87,7 @@ sub _decode {
         # get field tag number
         my $tag = $field_node->getAttribute('tag');
         my $occurrence = $field_node->getAttribute('occurrence') // '';
-        push(@field, ($tag, $occurrence, '_', ''));
+        push(@field, ($tag, $occurrence));
 
             # get all subfield nodes
             foreach my $subfield_node ( $field_node->getChildrenByTagName('*') ) {

lib/PICA/Writer/Plain.pm

@@ -21,8 +21,7 @@ sub _write_record {
             print $fh "/".$field->[1]; # TODO: fix one-digit occ??
         }
         print $fh ' ';
-        # ignore $field->[2,3] ...
-        for (my $i=4; $i<scalar @$field; $i+=2) {
+        for (my $i=2; $i<scalar @$field; $i+=2) {
             my $value = $field->[$i+1];
             $value =~ s/\$/\$\$/g;
             print $fh SUBFIELD_INDICATOR . $field->[$i] . $value;

lib/PICA/Writer/Plus.pm

@@ -6,7 +6,7 @@ use strict;
 use charnames qw(:full);
 use constant SUBFIELD_INDICATOR => "\N{INFORMATION SEPARATOR ONE}";
 use constant END_OF_FIELD       => "\N{INFORMATION SEPARATOR TWO}";
-use constant END_OF_RECORD      => "\x1D\x1A"; # TODO: check
+use constant END_OF_RECORD      => "\N{LINE FEED}"; # or \N{INFORMATION SEPARATOR THREE}? I would prefer newline separated format
 
 use Moo;
 with 'PICA::Writer::Handle';
@@ -21,7 +21,7 @@ sub _write_record {
             print $fh "/".$field->[1];
         }
         print $fh ' ';
-        for (my $i=4; $i<scalar @$field; $i+=2) {
+        for (my $i=2; $i<scalar @$field; $i+=2) {
             print $fh SUBFIELD_INDICATOR . $field->[$i] . $field->[$i+1];
         }
         print $fh END_OF_FIELD;

lib/PICA/Writer/XML.pm

@@ -29,7 +29,7 @@ sub _write_record {
                 defined $field->[1] && $field->[1] ne '' ?
                 " occurrence=\"$field->[1]\"" : ""
             ) . ">\n";
-            for (my $i=4; $i<scalar @$field; $i+=2) {
+            for (my $i=2; $i<scalar @$field; $i+=2) {
                 my $value = $field->[$i+1];
                 $value =~ s/</&lt;/g;
                 $value =~ s/&/&amp;/g;

t/01-parser.t

@@ -10,16 +10,16 @@ isa_ok( $parser, 'PICA::Parser::XML' );
 my $record = $parser->next();
 ok($record->{_id} eq '658700774', 'record _id' );
 ok($record->{record}->[0][0] eq '001@', 'tag from first field' );
-is_deeply($record->{record}->[1], ['001A', '', '_', '', '0', '2045:10-03-11'], 'second field');
+is_deeply($record->{record}->[1], ['001A', '', '0', '2045:10-03-11'], 'second field');
 ok($parser->next()->{_id} eq '65869538X', 'next record');
 
 use PICA::Parser::Plus;
 $parser = PICA::Parser::Plus->new( './t/picaplus.dat' );
 isa_ok( $parser, 'PICA::Parser::Plus' );
 $record = $parser->next();
 ok($record->{_id} eq '1041318383', 'record _id' );
-ok($record->{record}->[0][0] eq 'LDR', 'tag from first field' );
-is_deeply($record->{record}->[1], ['001A', '', '_', '', '0', '1240:04-09-13'], 'second field');
+ok($record->{record}->[0][0] eq '001A', 'tag from first field' );
+is_deeply($record->{record}->[0], ['001A', '', '0', '1240:04-09-13'], 'first field');
 ok($parser->next()->{_id} eq '1041318464', 'next record');
 
 done_testing();

t/02-importer.t

@@ -16,7 +16,7 @@ $importer->each(
 );
 ok(scalar @records == 5, 'records');
 ok( $records[0]->{'_id'} eq '658700774', 'record _id' );
-is_deeply( $records[0]->{'record'}->[7], ['003@', '', '_', '', '0', '658700774'],
+is_deeply( $records[0]->{'record'}->[7], ['003@', '', '0', '658700774'],
     'record field'
 );
 
@@ -29,7 +29,7 @@ $importer->each(
 );
 ok(scalar @records == 10, 'records');
 ok( $records[0]->{'_id'} eq '1041318383', 'record _id' );
-is_deeply( $records[0]->{'record'}->[6], ['003@', '', '_', '', '0', '1041318383'],,
+is_deeply( $records[0]->{'record'}->[5], ['003@', '', '0', '1041318383'],,
     'record field'
 );
 

t/05-writer.t

@@ -11,12 +11,12 @@ my $writer = PICA::Writer::Plain->new( fh => $fh );
 
 my @pica_records = (
     [
-      ['003@', '', '_', '', '0', '1041318383'],
-      ['021A', '', '_', '', 'a', encode('UTF-8',"Hello \$\N{U+00A5}!")],
+      ['003@', '', '0', '1041318383'],
+      ['021A', '', 'a', encode('UTF-8',"Hello \$\N{U+00A5}!")],
     ],
     {
       record => [
-        ['028C', '01', '_', '', d => 'Emma', a => 'Goldman']
+        ['028C', '01', d => 'Emma', a => 'Goldman']
       ]
     }
 );