--- trunk/run.pl 2006/10/08 20:28:17 755 +++ trunk/run.pl 2007/05/27 16:49:15 857 @@ -8,11 +8,11 @@ use WebPAC::Common 0.02; use WebPAC::Parser 0.08; -use WebPAC::Input 0.13; -use WebPAC::Store 0.11; +use WebPAC::Input 0.16; +use WebPAC::Store 0.14; use WebPAC::Normalize 0.22; use WebPAC::Output::TT; -use WebPAC::Validate 0.06; +use WebPAC::Validate 0.11; use WebPAC::Output::MARC; use WebPAC::Config; use Getopt::Long; @@ -21,6 +21,7 @@ use File::Slurp; use Data::Dump qw/dump/; use Storable qw/dclone/; +use Pod::Usage qw/pod2usage/; use Proc::Queue size => 1; use POSIX ":sys_wait_h"; # imports WNOHANG @@ -31,7 +32,7 @@ B -Options: +=head1 OPTIONS =over 4 @@ -67,6 +68,19 @@ turn on extra validation of imput records, see L +You can use special variables C<$database> and $C<$input> in this parametar +like C<--validate 'conf/validate/$database-$input'> to construct filename + +=item --validate-delimiters path/to/validate_delimiters_file + +this option is used with C<--validate> to turn on extra validation of +delimiters. If file is non existant, it will be created on first run. + +=item --marc-generate + +Generate MARC file. This will automatically be on if file contains C directives. +You can use this option as C<--no-marc-generate> to disable MARC generation. + =item --marc-lint By default turned on if normalisation file has C directives. You can disable lint @@ -102,11 +116,14 @@ my $only_filter; my $stats = 0; my $validate_path; +my $validate_delimiters_path; +my $marc_generate = 1; my $marc_lint = 1; my $marc_dump = 0; my $parallel = 0; my $only_links = 0; my $merge = 0; +my $help; my $log = _new WebPAC::Common()->_get_logger(); @@ -120,13 +137,20 @@ "debug+" => \$debug, "stats" => \$stats, "validate=s" => \$validate_path, + "validate-delimiters=s" => \$validate_delimiters_path, + "marc-generate!" => \$marc_generate, "marc-lint!" => \$marc_lint, "marc-dump!" => \$marc_dump, "parallel=i" => \$parallel, "only-links!" => \$only_links, "merge" => \$merge, + "help" => \$help, ); +$marc_generate = 0 if ( $validate_delimiters_path ); + +pod2usage(-verbose => 2) if ($help); + my $config = new WebPAC::Config( path => $config_path ); #print "config = ",dump($config) if ($debug); @@ -135,6 +159,12 @@ $log->info( "-" x 79 ); +my $log_file = 'log'; + +if (-e $log_file ) { # && -s $log_file > 5 * 1024 * 1024) { + $log->info("moved old log with ", -s $log_file, " bytes to '${log_file}.old'"); + rename $log_file, "${log_file}.old" || $log->logwarn("can't rename $log_file to ${log_file}.old: $!"); +} my $estcmd_fh; my $estcmd_path = './estcmd-merge.sh'; @@ -145,16 +175,15 @@ $log->info("created merge batch file $estcmd_path"); } - my $validate; $validate = new WebPAC::Validate( - path => $validate_path, -) if ($validate_path); - + delimiters => $config->webpac('delimiters'), +) if ($validate_path || $validate_delimiters_path); my $use_indexer = $config->use_indexer; +$stats ||= $validate; if ($stats) { - $log->debug("option --stats disables update of indexing engine..."); + $log->debug("disabled indexing for stats collection"); $use_indexer = undef; } else { $log->info("using $use_indexer indexing engine..."); @@ -183,7 +212,8 @@ return $c; } -while (my ($database, $db_config) = each %{ $config->databases }) { +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); @@ -350,13 +380,6 @@ my $rec = shift || die "need rec!"; my $mfn = $rec->{'000'}->[0] || die "need mfn in 000"; - $store->save_row( - database => $database, - input => $input_name, - id => $mfn, - row => $rec, - ); - WebPAC::Normalize::data_structure( row => $rec, rules => $rules, @@ -380,12 +403,30 @@ lookup_coderef => $lookup_coderef, lookup => $lookup_jar, %{ $input }, + load_row => sub { + my $a = shift; + return $store->load_row( + database => $database, + input => $input_name, + id => $a->{id}, + ); + }, + save_row => sub { + my $a = shift; + return $store->save_row( + database => $database, + input => $input_name, + id => $a->{id}, + row => $a->{row}, + ); + }, + ); my $lookup_data = WebPAC::Normalize::_get_lookup(); if (defined( $lookup_data->{$database}->{$input_name} )) { - $log->debug("created following lookups: ", dump( $lookup_data )); + $log->debug("created following lookups: ", sub { dump( $lookup_data ) } ); foreach my $key (keys %{ $lookup_data->{$database}->{$input_name} }) { $store->save_lookup( @@ -405,10 +446,15 @@ 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"); + + if ( $validate ) { + $validate->read_validate_file( $validate->fill_in( $validate_path, database => $database, input => $input_name ) ); + $validate->read_validate_delimiters_file( $validate->fill_in( $validate_delimiters_path, database => $database, input => $input_name ) ); + } } my $marc; - if ($parser->have_rules( 'marc', $database, $input_name )) { + if ($marc_generate && $parser->have_rules( 'marc', $database, $input_name )) { $marc = new WebPAC::Output::MARC( path => "out/marc/${database}-${input_name}.marc", lint => $marc_lint, @@ -436,17 +482,19 @@ foreach my $db (keys %$depends) { foreach my $i (keys %{$depends->{$db}}) { foreach my $k (keys %{$depends->{$db}->{$i}}) { + my $t = time(); $log->debug("loading lookup $db/$i"); $lookup_hash->{$db}->{$i}->{$k} = $store->load_lookup( database => $db, input => $i, key => $k, ); + $log->debug(sprintf("lookup $db/$i took %.2fs", time() - $t)); } } } - $log->debug("lookup_hash = ", dump( $lookup_hash )); + $log->debug("lookup_hash = ", sub { dump( $lookup_hash ) }); } @@ -454,6 +502,8 @@ my $row = $input_db->fetch || next; + $total_rows++; + my $mfn = $row->{'000'}->[0]; if (! $mfn || $mfn !~ m#^\d+$#) { @@ -464,11 +514,12 @@ if ($validate) { - if ( my $errors = $validate->validate_errors( $row, $input_db->dump ) ) { + if ( my $errors = $validate->validate_rec( $row, $input_db->dump_ascii ) ) { $log->error( "MFN $mfn validation error:\n", $validate->report_error( $errors ) ); } + next; # validation doesn't create any output } my $ds = WebPAC::Normalize::data_structure( @@ -487,7 +538,7 @@ }, ); - $log->debug("ds = ",dump($ds)) if ($ds); + $log->debug("ds = ", sub { dump($ds) }) if ($ds); $store->save_ds( database => $database, @@ -509,7 +560,7 @@ $marc->add( id => $mfn . ( $i ? "/$i" : '' ), fields => $fields, - leader => WebPAC::Normalize::marc_leader(), + leader => WebPAC::Normalize::_get_marc_leader(), row => $row, ); $i++; @@ -517,8 +568,6 @@ $log->info("Created $i instances of MFN $mfn\n") if ($i > 1); } - - $total_rows++; } if ($validate) { @@ -527,6 +576,11 @@ $log->info("validation errors:\n$errors\n" ); print $report_fh "$errors\n" if ($report_fh); } + + print $report_fh "\nAll possible subfields/delimiter templates:\n", $validate->delimiters_templates( report => 1, current_input => 1 ), "\n\n"; + + # must be last thing that touches $validate for this input + $validate->reset; } if ($stats) { @@ -539,8 +593,7 @@ $marc->finish if ($marc); # close report - close($report_fh) if ($report_fh) - + close($report_fh) if ($report_fh); } eval { $indexer->finish } if ($indexer && $indexer->can('finish')); @@ -567,6 +620,9 @@ $log->info("all parallel processes finished"); } +# save new delimiters if needed +$validate->save_delimiters_templates if ( $validate ); + # # handle links or merge after indexing #