use Data::Dump qw/dump/;
use Storable qw/dclone/;
use Pod::Usage qw/pod2usage/;
+use LWP::Simple;
-use Proc::Queue size => 1;
use POSIX ":sys_wait_h"; # imports WNOHANG
=head1 NAME
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
my $parallel = 0;
my $only_links = 0;
my $merge = 0;
+my $mirror;
my $help;
my $log = _new WebPAC::Common()->_get_logger();
"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,
"parallel=i" => \$parallel,
"only-links!" => \$only_links,
"merge" => \$merge,
+ "mirror=s" => \$mirror,
"help" => \$help,
);
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);
# 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 $start_t = time();
my @links;
if ($parallel) {
+ eval 'use Proc::Queue size => 1;';
+ die $@ if $@;
$log->info("Using $parallel processes for speedup");
Proc::Queue::size($parallel);
}
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) {
foreach my $output ( @outputs ) {
-warn '## output = ',dump( $output );
+#warn '## output = ',dump( $output );
my $module = $output->{module} || $log->logdie("need module in output section of $database");
$module = 'WebPAC::Output::' . $module unless $module =~ m/::/;
# 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");
+ }
}
$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});
- 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);
- $log->info("working on input '$input_name' in $input->{path} [type: $input->{type}] using $input_module",
+ $log->info("working on $database/$input_name with $input_module on $input->{path}",
@lookups ? " creating lookups: ".join(", ", @lookups) : ""
);
delete($input->{modify_file});
}
+ if ( $mirror ) {
+ my $path = $input->{path} || die "no input path in ",dump( $input );
+ $log->info( "mirror ", $path, " ", -s $path, " bytes" );
+
+ $log->warn( "$path not modified" )
+ if mirror( "$mirror/$path", $path ) == RC_NOT_MODIFIED;
+ }
+
my $input_db = new WebPAC::Input(
module => $input_module,
- encoding => $config->webpac('webpac_encoding'),
limit => $limit || $input->{limit},
offset => $offset,
recode => $input->{recode},
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 },
);
}
- 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
next; # validation doesn't create any output
}
- if ( 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) = @_;
- return $store->load_row(
+ if ($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(
+ 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,
- );
- },
- ) ) { # if
+ ds => $ds,
+ ) if !$stats;
- $log->debug("ds = ", sub { dump($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 $indexer;
-
- if ($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++;
+ foreach my $out ( @output_modules ) {
+ $out->add( $mfn, $ds ) if $out->can('add');
}
- $log->info("Created $i instances of MFN $mfn\n") if ($i > 1);
+ } else {
+ $log->warn("record $pos didn't produce any output after normalization rules!") unless $marc;
}
+ }
+
+ if ($marc) {
+ my $i = 0;
- foreach my $out ( @output_modules ) {
- $out->add( $mfn, $ds ) if $out->can('add');
+ 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++;
}
- } else {
- $log->warn("record $pos didn't produce any output after normalization rules!");
+ $log->info("Created $i instances of MFN $mfn\n") if ($i > 1);
}
}
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');
projects / webpac2 / blobdiff