--- trunk/run.pl 2007/11/03 12:35:42 978 +++ trunk/run.pl 2009/05/19 11:55:30 1185 @@ -11,7 +11,7 @@ use WebPAC::Input 0.16; use WebPAC::Store 0.15; use WebPAC::Normalize 0.22; -use WebPAC::Output::TT; +#use WebPAC::Output::TT; use WebPAC::Validate 0.11; use WebPAC::Output::MARC; use WebPAC::Config; @@ -22,8 +22,8 @@ use Data::Dump qw/dump/; use Storable qw/dclone/; use Pod::Usage qw/pod2usage/; +use LWP::Simple qw//; -use Proc::Queue size => 1; use POSIX ":sys_wait_h"; # imports WNOHANG =head1 NAME @@ -103,6 +103,10 @@ Create merged index of databases which have links +=item --mirror http://www.example.com + +Tries to download input path files from mirror URI + =back =cut @@ -123,6 +127,7 @@ my $parallel = 0; my $only_links = 0; my $merge = 0; +my $mirror; my $help; my $log = _new WebPAC::Common()->_get_logger(); @@ -144,6 +149,7 @@ "parallel=i" => \$parallel, "only-links!" => \$only_links, "merge" => \$merge, + "mirror=s" => \$mirror, "help" => \$help, ); @@ -153,6 +159,8 @@ 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); @@ -205,6 +213,8 @@ my @links; if ($parallel) { + eval 'use Proc::Queue size => 1;'; + die $@ if $@; $log->info("Using $parallel processes for speedup"); Proc::Queue::size($parallel); } @@ -326,12 +336,15 @@ # add database to arugemnts for output filter $output->{database} = $database; + $output->{clean} = $clean; $log->debug("calling $module->new(",dump( $output ),")"); my $out = new $module->new( $output ); - $out->init; - - push @output_modules, $out; + if ( $out->init ) { + push @output_modules, $out; + } else { + $log->warn("SKIPPED $module"); + } } @@ -344,10 +357,32 @@ $log->info("database $database doesn't have inputs defined"); } ); + if ( -e 'out/debug' ) { # FIXME flag? + my $out; + foreach my $i ( @inputs ) { + warn dump( $i ); + next unless defined $i->{normalize}; + warn dump( $i->{normalize} ); + foreach my $normalize ( @{ $i->{normalize} } ) { + my $path = $normalize->{path}; + $out .= qq/\n##\n## $path\n##\n\n/; + $out .= read_file( $path ); + } + } + my $all = "out/debug/all-normalize.pl"; + write_file( $all, $out ); + warn "### all normalize for this input saved to: $all"; + }; + foreach my $input (@inputs) { my $input_name = $input->{name} || $log->logdie("input without a name isn't valid: ",dump($input)); + if ( $input->{skip} ) { + $log->info("skip $input_name"); + next; + } + next if ($only_input && ($input_name !~ m#$only_input#i && $input->{type} !~ m#$only_input#i)); my $type = lc($input->{type}); @@ -365,7 +400,7 @@ my @lookups = $parser->have_lookup_create($database, $input); - $log->info("working on input '$input_name' in $input->{path} [type: $input->{type}] using $input_module", + $log->info("working on $database/$input_name with $input_module on $input->{path}", @lookups ? " creating lookups: ".join(", ", @lookups) : "" ); @@ -375,9 +410,24 @@ delete($input->{modify_file}); } + if ( $mirror ) { + my $path = $input->{path} || die "no input path in ",dump( $input ); + + my $base = $path; + $base =~ s{/[^/]+$}{}; + mkpath $base unless -e $base; + + my $rc = LWP::Simple::mirror( "$mirror/$path", $path ); + if (LWP::Simple::is_error( $rc )) { + die "can't mirror $mirror/$path -> $path [$rc]"; + } else { + $log->info( "mirror ", $path, " [$rc] ", -s $path, " bytes" ); + } + + } + my $input_db = new WebPAC::Input( module => $input_module, - encoding => $config->webpac('webpac_encoding'), limit => $limit || $input->{limit}, offset => $offset, recode => $input->{recode}, @@ -422,7 +472,7 @@ my $maxmfn = $input_db->open( path => $input->{path}, - code_page => $input->{encoding}, # database encoding + input_encoding => $input->{encoding}, # database encoding lookup_coderef => $lookup_coderef, lookup => $lookup_jar, %{ $input }, @@ -463,7 +513,9 @@ my $report_fh; if ($stats || $validate) { - my $path = "out/report/${database}-${input_name}.txt"; + my $out_report = 'out/report'; # FIXME move to config + mkpath $out_report unless -e $out_report; + 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 ", @@ -478,14 +530,20 @@ my $marc; if ($marc_generate && $parser->have_rules( 'marc', $database, $input_name )) { + + my $out_marc = 'out/marc'; # FIXME move to config + mkpath $out_marc unless -e $out_marc; + $marc = new WebPAC::Output::MARC( - path => "out/marc/${database}-${input_name}.marc", + path => "$out_marc/${database}-${input_name}.marc", lint => $marc_lint, dump => $marc_dump, ); } - my $rules = $parser->normalize_rules($database,$input_name) || $log->logdie("no normalize rules found for $database/$input_name"); + my $rules = $parser->normalize_rules($database,$input_name); + $log->logwarn("no normalize rules for $database/$input_name") unless $rules; + $log->debug("parsed normalize rules:\n$rules"); # reset position in database @@ -555,56 +613,59 @@ next; # validation doesn't create any output } - my $ds = WebPAC::Normalize::data_structure( - row => $row, - rules => $rules, - lookup => $lookup_hash, - config => create_ds_config( $db_config, $database, $input, $mfn ), - marc_encoding => 'utf-8', - load_row_coderef => sub { - my ($database,$input,$mfn) = @_; + if ($rules) { + + my $ds = WebPAC::Normalize::data_structure( + row => $row, + rules => $rules, + lookup => $lookup_hash, + config => create_ds_config( $db_config, $database, $input, $mfn ), + marc_encoding => 'utf-8', + load_row_coderef => sub { + my ($database,$input,$mfn) = @_; #warn "### load_row($database,$input,$mfn) from data_structure\n"; - return $store->load_row( + return $store->load_row( + database => $database, + input => $input, + id => $mfn, + ); + }, + ); + + $log->debug("ds = ", sub { dump($ds) }); + + if ( $ds ) { + + $store->save_ds( database => $database, - input => $input, + input => $input_name, id => $mfn, - ); - }, - ); - - $log->debug("ds = ", sub { dump($ds) }); + ds => $ds, + ) if !$stats; - if ( $ds ) { + $indexer->add( + id => "${input_name}/${mfn}", + ds => $ds, + type => $config->get($indexer_config)->{type}, + ) if $indexer; - $store->save_ds( - database => $database, - input => $input_name, - id => $mfn, - ds => $ds, - ) if !$stats; - - $indexer->add( - id => "${input_name}/${mfn}", - ds => $ds, - type => $config->get($indexer_config)->{type}, - ) if $indexer; + foreach my $out ( @output_modules ) { + $out->add( $mfn, $ds ) if $out->can('add'); + } - 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!") unless $marc; } - - } else { - $log->warn("record $pos didn't produce any output after normalization rules!") unless $marc; } if ($marc) { my $i = 0; - while (my $fields = WebPAC::Normalize::_get_marc_fields( fetch_next => 1 ) ) { + while (my $fields = WebPAC::Normalize::MARC::_get_marc_fields( fetch_next => 1 ) ) { $marc->add( id => $mfn . ( $i ? "/$i" : '' ), fields => $fields, - leader => WebPAC::Normalize::_get_marc_leader(), + leader => WebPAC::Normalize::MARC::_get_marc_leader(), row => $row, ); $i++; @@ -641,7 +702,7 @@ close($report_fh) if ($report_fh); } - eval { $indexer->finish } if ($indexer && $indexer->can('finish')); + $indexer->finish if $indexer && $indexer->can('finish'); foreach my $out ( @output_modules ) { $out->finish if $out->can('finish');