--- trunk/run.pl 2007/11/02 10:54:44 957 +++ trunk/run.pl 2007/11/12 12:35:32 1043 @@ -133,7 +133,7 @@ "clean" => \$clean, "one=s" => \$only_filter, "only=s" => \$only_filter, - "config" => \$config_path, + "config=s" => \$config_path, "debug+" => \$debug, "stats" => \$stats, "validate=s" => \$validate_path, @@ -153,6 +153,8 @@ my $config = new WebPAC::Config( path => $config_path ); +WebPAC::Normalize::_debug( $debug - 1 ) if $debug > 1; + #print "config = ",dump($config) if ($debug); die "no databases in config file!\n" unless ($config->databases); @@ -191,7 +193,13 @@ # parse normalize files and create source files for lookup and normalization -my $parser = new WebPAC::Parser( config => $config ); +my ($only_database,$only_input) = split(m#/#, $only_filter) if $only_filter; + +my $parser = new WebPAC::Parser( + config => $config, + only_database => $only_database, + only_input => $only_input, +); my $total_rows = 0; my $start_t = time(); @@ -215,7 +223,6 @@ foreach my $database ( sort keys %{ $config->databases } ) { my $db_config = $config->databases->{$database}; - my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); next if ($only_database && $database !~ m/$only_database/i); if ($parallel) { @@ -347,9 +354,16 @@ my $type = lc($input->{type}); - die "I know only how to handle input types ", join(",", $config->webpac('inputs') ), " not '$type'!\n" unless (grep(/$type/, $config->webpac('inputs'))); + # FIXME check if input module exists + my $input_module = $input->{module}; - my $input_module = $config->webpac('inputs')->{$type}; + if ( ! $input_module ) { + if ( grep(/$type/, $config->webpac('inputs')) ) { + $input_module = $config->webpac('inputs')->{$type}; + } else { + $log->logdie("I know only how to handle input types ", join(",", $config->webpac('inputs') ), " not '$type'!" ); + } + } my @lookups = $parser->have_lookup_create($database, $input); @@ -543,7 +557,7 @@ next; # validation doesn't create any output } - if ( my $ds = WebPAC::Normalize::data_structure( + my $ds = WebPAC::Normalize::data_structure( row => $row, rules => $rules, lookup => $lookup_hash, @@ -551,15 +565,18 @@ marc_encoding => 'utf-8', load_row_coderef => sub { my ($database,$input,$mfn) = @_; +#warn "### load_row($database,$input,$mfn) from data_structure\n"; return $store->load_row( database => $database, input => $input, id => $mfn, ); }, - ) ) { # if + ); + + $log->debug("ds = ", sub { dump($ds) }); - $log->debug("ds = ", sub { dump($ds) }); + if ( $ds ) { $store->save_ds( database => $database, @@ -574,28 +591,28 @@ type => $config->get($indexer_config)->{type}, ) if $indexer; - if ($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::_get_marc_leader(), - row => $row, - ); - $i++; - } - - $log->info("Created $i instances of MFN $mfn\n") if ($i > 1); - } - foreach my $out ( @output_modules ) { $out->add( $mfn, $ds ) if $out->can('add'); } } else { - $log->warn("record $pos didn't produce any output after normalization rules!"); + $log->warn("record $pos didn't produce any output after normalization rules!") unless $marc; + } + + if ($marc) { + my $i = 0; + + while (my $fields = WebPAC::Normalize::MARC::_get_marc_fields( fetch_next => 1 ) ) { + $marc->add( + id => $mfn . ( $i ? "/$i" : '' ), + fields => $fields, + leader => WebPAC::Normalize::MARC::_get_marc_leader(), + row => $row, + ); + $i++; + } + + $log->info("Created $i instances of MFN $mfn\n") if ($i > 1); } }