--- trunk/run.pl 2006/07/01 20:29:09 556 +++ trunk/run.pl 2006/07/04 11:36:03 580 @@ -13,13 +13,12 @@ use WebPAC::Normalize; use WebPAC::Output::TT; use WebPAC::Validate; +use WebPAC::Output::MARC; use YAML qw/LoadFile/; use Getopt::Long; use File::Path; use Time::HiRes qw/time/; use File::Slurp; -use MARC::Record 2.0; # need 2.0 for utf-8 encoding see marcpm.sf.net -use MARC::Lint; use Data::Dump qw/dump/; =head1 NAME @@ -75,7 +74,11 @@ =item --marc-lint By default turned on if C<--marc-normalize> is used. You can disable lint -messages with C<--no-arc-lint>. +messages with C<--no-marc-lint>. + +=item --marc-dump + +Force dump or input and marc record for debugging. =back @@ -92,6 +95,7 @@ my $validate_path; my ($marc_normalize, $marc_output); my $marc_lint = 1; +my $marc_dump = 0; GetOptions( "limit=i" => \$limit, @@ -100,12 +104,13 @@ "one=s" => \$only_filter, "only=s" => \$only_filter, "config" => \$config, - "debug" => \$debug, + "debug+" => \$debug, "stats" => \$stats, "validate=s" => \$validate_path, "marc-normalize=s" => \$marc_normalize, "marc-output=s" => \$marc_output, "marc-lint!" => \$marc_lint, + "marc-dump!" => \$marc_dump, ); $config = LoadFile($config); @@ -139,8 +144,6 @@ my @links; my $indexer; -my $lint = new MARC::Lint if ($marc_lint); - while (my ($database, $db_config) = each %{ $config->{databases} }) { my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); @@ -265,13 +268,11 @@ $log->info("Using $normalize_path for normalization..."); - my $marc_fh; - if (my $path = $normalize->{output}) { - open($marc_fh, '>', $path) || - $log->logdie("can't open MARC output $path: $!"); - - $log->info("Creating MARC export file $path", $marc_lint ? ' (with lint)' : '', "\n"); - } + my $marc = new WebPAC::Output::MARC( + path => $normalize->{output}, + lint => $marc_lint, + dump => $marc_dump, + ) if ($normalize->{output}); # reset position in database $input_db->seek(1); @@ -294,7 +295,6 @@ $log->error( "MFN $mfn validation errors:\n", join("\n", @errors) ) if (@errors); } - my $ds = WebPAC::Normalize::data_structure( row => $row, rules => $rules, @@ -314,25 +314,20 @@ type => $config->{$use_indexer}->{type}, ) if ($indexer && $ds); - if ($marc_fh) { - my $marc = new MARC::Record; - $marc->encoding( 'utf-8' ); - my @marc_fields = WebPAC::Normalize::_get_marc_fields(); - if (! @marc_fields) { - $log->warn("MARC record $mfn is empty, skipping"); - } else { - $marc->add_fields( @marc_fields ); - if ($marc_lint) { - $lint->check_record( $marc ); - my $err = join( "\n", $lint->warnings ); - $log->error("MARC lint detected warning on MFN $mfn\n", - "Original imput row: ",dump($row), "\n", - "Normalized MARC row: ",dump(@marc_fields), "\n", - "MARC lint warnings: ",$err,"\n" - ) if ($err); - } - print $marc_fh $marc->as_usmarc; + if ($marc) { + my $i = 0; + + while (my $fields = WebPAC::Normalize::_get_marc_fields( fetch_next => 1 ) ) { + $marc->add( + id => $mfn . ( $i ? "/$i" : '' ), + fields => $fields, + leader => WebPAC::Normalize::marc_leader(), + row => $row, + ); + $i++; } + + $log->info("Created $i instances of MFN $mfn\n") if ($i > 1); } $total_rows++; @@ -341,7 +336,7 @@ $log->info("statistics of fields usage:\n", $input_db->stats) if ($stats); # close MARC file - close($marc_fh) if ($marc_fh); + $marc->finish if ($marc); }