--- trunk/run.pl 2006/09/25 13:08:17 702 +++ trunk/run.pl 2006/09/26 12:42:49 713 @@ -8,9 +8,8 @@ use WebPAC::Common 0.02; use WebPAC::Parser 0.04; -use WebPAC::Lookup 0.03; -use WebPAC::Input 0.11; -use WebPAC::Store 0.03; +use WebPAC::Input 0.13; +use WebPAC::Store 0.11; use WebPAC::Normalize 0.11; use WebPAC::Output::TT; use WebPAC::Validate 0.06; @@ -189,6 +188,15 @@ Proc::Queue::size($parallel); } +sub create_ds_config { + my ($db_config, $database, $input, $mfn) = @_; + my $c = dclone( $db_config ); + $c->{_} = $database || $log->logconfess("need database"); + $c->{_mfn} = $mfn || $log->logconfess("need mfn"); + $c->{input} = $input || $log->logconfess("need input"); + return $c; +} + while (my ($database, $db_config) = each %{ $config->databases }) { my ($only_database,$only_input) = split(m#/#, $only_filter) if ($only_filter); @@ -277,7 +285,7 @@ my $abs_path = abs_path($0); $abs_path =~ s#/[^/]*$#/#; - my $db_path = $config->get('webpac')->{db_path} . '/' . $database; + my $db_path = $config->webpac('db_path'); if ($clean) { $log->info("creating new database '$database' in $db_path"); @@ -286,9 +294,8 @@ $log->info("working on database '$database' in $db_path"); } - my $db = new WebPAC::Store( + my $store = new WebPAC::Store( path => $db_path, - database => $database, debug => $debug, ); @@ -308,7 +315,9 @@ foreach my $input (@inputs) { - next if ($only_input && ($input->{name} !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); + my $input_name = $input->{name} || $log->logdie("input without a name isn't valid: ",dump($input)); + + next if ($only_input && ($input_name !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); my $type = lc($input->{type}); @@ -316,8 +325,10 @@ my $input_module = $config->webpac('inputs')->{$type}; - $log->info("working on input '$input->{name}' in $input->{path} [type: $input->{type}] using $input_module", - $input->{lookup} ? "lookup '$input->{lookup}'" : "" + my @lookups = $parser->have_lookup_create($database, $input); + + $log->info("working on input '$input_name' in $input->{path} [type: $input->{type}] using $input_module", + @lookups ? " creating lookups: ".join(", ", @lookups) : "" ); if ($stats) { @@ -326,22 +337,11 @@ delete($input->{modify_file}); } - warn "parser->depends = ", dump( $parser->{depends} ); - warn "depends on: ", dump( $parser->depends($database, $input->{name}) ); - warn "lookup_create_rules = ", dump( $parser->lookup_create_rules($database, $input->{name}) ); - warn "parser->_lookup_create = ", dump( $parser->{_lookup_create} ); - - my $lookup; - my $input_db = new WebPAC::Input( module => $input_module, encoding => $config->webpac('webpac_encoding'), limit => $limit || $input->{limit}, offset => $offset, - lookup_coderef => sub { - my $rec = shift || return; - $lookup->add( $rec ); - }, recode => $input->{recode}, stats => $stats, modify_records => $input->{modify_records}, @@ -350,22 +350,65 @@ $log->logdie("can't create input using $input_module") unless ($input); if (defined( $input->{lookup} )) { - $log->warn("$database/", $input->{name}, " has depriciated lookup definition, removing it..."); + $log->warn("$database/$input_name has depriciated lookup definition, removing it..."); delete( $input->{lookup} ); } + my $lookup; + my $lookup_coderef; + + if (@lookups) { + + my $rules = $parser->lookup_create_rules($database, $input) || $log->logdie("no rules found for $database/$input"); + + $lookup_coderef = sub { + my $rec = shift || die "need rec!"; + my $mfn = $rec->{'000'}->[0] || die "need mfn in 000"; + + WebPAC::Normalize::data_structure( + row => $rec, + rules => $rules, + lookup => $lookup, + config => create_ds_config( $db_config, $database, $input, $mfn ), + ); + + warn "current lookup = ", dump($lookup) if ($lookup); + }; + + WebPAC::Normalize::_set_lookup( undef ); + + $log->debug("created lookup_coderef using:\n$rules"); + + }; + my $maxmfn = $input_db->open( path => $input->{path}, code_page => $input->{encoding}, # database encoding + lookup_coderef => $lookup_coderef, %{ $input }, ); + my $lookup_data = WebPAC::Normalize::_get_lookup(); + + if (defined( $lookup_data->{$database}->{$input_name} )) { + $log->debug("created following lookups: ", dump( $lookup_data )); + + foreach my $key (keys %{ $lookup_data->{$database}->{$input_name} }) { + $store->save_lookup( + database => $database, + input => $input_name, + key => $key, + data => $lookup_data->{$database}->{$input_name}->{$key}, + ); + } + } + my $report_fh; if ($stats || $validate) { - my $path = "out/report/" . $database . '-' . $input->{name} . '.txt'; + my $path = "out/report/${database}-${input_name}.txt"; open($report_fh, '>', $path) || $log->logdie("can't open $path: $!"); - print $report_fh "Report for database '$database' input '$input->{name}' records ", + print $report_fh "Report for database '$database' input '$input_name' records ", $offset || 1, "-", $limit || $input->{limit} || $maxmfn, "\n\n"; $log->info("Generating report file $path"); } @@ -376,7 +419,7 @@ if ($marc_normalize) { @norm_array = ( { path => $marc_normalize, - output => $marc_output || 'out/marc/' . $database . '-' . $input->{name} . '.marc', + output => $marc_output || "out/marc/${database}-${input_name}.marc", } ); } @@ -424,33 +467,23 @@ } } - my $ds_config = dclone($db_config); - - # default values -> database key - $ds_config->{_} = $database; - - # current mfn - $ds_config->{_mfn} = $mfn; - - # attach current input - $ds_config->{input} = $input; - my $ds = WebPAC::Normalize::data_structure( row => $row, rules => $rules, lookup => $lookup ? $lookup->lookup_hash : undef, - config => $ds_config, + config => create_ds_config( $db_config, $database, $input, $mfn ), marc_encoding => 'utf-8', ); - $db->save_ds( + $store->save_ds( + database => $database, + input => $input_name, id => $mfn, ds => $ds, - prefix => $input->{name}, ) if ($ds && !$stats); $indexer->add( - id => $input->{name} . "/" . $mfn, + id => "${input_name}/${mfn}", ds => $ds, type => $config->get($indexer_config)->{type}, ) if ($indexer && $ds);