--- trunk/lib/WebPAC/Normalize.pm	2006/01/08 21:16:27	371
+++ trunk/lib/WebPAC/Normalize.pm	2006/04/30 12:17:19	436
@@ -13,11 +13,11 @@
 
 =head1 VERSION
 
-Version 0.08
+Version 0.09
 
 =cut
 
-our $VERSION = '0.08';
+our $VERSION = '0.09';
 
 =head1 SYNOPSIS
 
@@ -137,6 +137,37 @@
 	$self ? return $self : return undef;
 }
 
+=head2 all_tags
+
+Returns all tags in document in specified order
+
+  my $sorted_tags = $self->all_tags();
+
+=cut
+
+sub all_tags {
+	my $self = shift;
+
+	if (! $self->{_tags_by_order}) {
+
+		my $log = $self->_get_logger;
+		# sanity check
+		$log->logdie("can't find self->{inport_xml}->{indexer}") unless ($self->{import_xml}->{indexer});
+
+		my @tags = keys %{ $self->{'import_xml'}->{'indexer'}};
+		$log->debug("unsorted tags: " . join(", ", @tags));
+
+		@tags = sort { $self->_sort_by_order } @tags;
+
+		$log->debug("sorted tags: " . join(",", @tags) );
+
+		$self->{_tags_by_order} = \@tags;
+	}
+
+	return $self->{_tags_by_order};
+}
+
+
 
 =head2 data_structure
 
@@ -172,19 +203,13 @@
 		$log->debug("cache miss, creating");
 	}
 
-	my @sorted_tags;
-	if ($self->{tags_by_order}) {
-		@sorted_tags = @{$self->{tags_by_order}};
-	} else {
-		@sorted_tags = sort { $self->_sort_by_order } keys %{$self->{'import_xml'}->{'indexer'}};
-		$self->{tags_by_order} = \@sorted_tags;
-	}
+	my $tags = $self->all_tags();
 
-	my $ds;
+	$log->debug("tags: ",sub { join(", ",@{ $tags }) });
 
-	$log->debug("tags: ",sub { join(", ",@sorted_tags) });
+	my $ds;
 
-	foreach my $field (@sorted_tags) {
+	foreach my $field (@{ $tags }) {
 
 		my $row;
 
@@ -198,13 +223,13 @@
 
 			my @v;
 			if ($self->{'lookup_regex'} && $format =~ $self->{'lookup_regex'}) {
-				@v = $self->fill_in_to_arr($rec,$format);
+				@v = $self->_rec_to_arr($rec,$format,'fill_in');
 			} else {
-				@v = $self->parse_to_arr($rec,$format);
+				@v = $self->_rec_to_arr($rec,$format,'parse');
 			}
 			if (! @v) {
 				$log->debug("$field <",$self->{tag},"> format: $format no values");
-#				next;
+				next;
 			} else {
 				$log->debug("$field <",$self->{tag},"> format: $format values: ", join(",", @v));
 			}
@@ -364,7 +389,7 @@
 
 			# we will skip delimiter before first occurence of field!
 			push @out, $del unless($found_any->{$fld_type} == 1);
-			push @out, $tmp;
+			push @out, $tmp if ($tmp);
 		}
 		$f_step++;
 	}
@@ -410,40 +435,6 @@
 	return $out;
 }
 
-=head2 parse_to_arr
-
-Similar to C<parse>, but returns array of all repeatable fields
-
- my @arr = $webpac->parse_to_arr($rec,'v250^a');
-
-=cut
-
-sub parse_to_arr {
-	my $self = shift;
-
-	my ($rec, $format_utf8) = @_;
-
-	my $log = $self->_get_logger();
-
-	$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);
-	return if (! $format_utf8);
-
-	my $i = 0;
-	my @arr;
-
-	my $rec_size = { '_' => '_' };
-
-	while (my $v = $self->parse($rec,$format_utf8,$i++,\$rec_size)) {
-		push @arr, $v;
-		warn "parse rec_size = ", Dumper($rec_size);
-	}
-
-	$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr);
-
-	return @arr;
-}
-
-
 =head2 fill_in
 
 Workhourse of all: takes record from in-memory structure of database and
@@ -505,15 +496,20 @@
 	# remove filter{...} from beginning
 	$filter_name = $1 if ($format =~ s/^filter{([^}]+)}//s);
 
-	# do actual replacement of placeholders
-	# repeatable fields
-	if ($format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found,$rec_size)/ges) {
-		$just_single = 0;
-	}
+	{
+		# fix warnings
+		no warnings 'uninitialized';
+
+		# do actual replacement of placeholders
+		# repeatable fields
+		if ($format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found,$rec_size)/ges) {
+			$just_single = 0;
+		}
 
-	# non-repeatable fields
-	if ($format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found,$rec_size)/ges) {
-		return if ($i > 0 && $just_single);
+		# non-repeatable fields
+		if ($format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found,$rec_size)/ges) {
+			return if ($i > 0 && $just_single);
+		}
 	}
 
 	if ($found) {
@@ -546,34 +542,47 @@
 }
 
 
-=head2 fill_in_to_arr
+=head2 _rec_to_arr
 
-Similar to C<fill_in>, but returns array of all repeatable fields. Usable
+Similar to C<parse> and C<fill_in>, but returns array of all repeatable fields. Usable
 for fields which have lookups, so they shouldn't be parsed but rather
-C<fill_id>ed.
+C<paste>d or C<fill_id>ed. Last argument is name of operation: C<paste> or C<fill_in>.
 
- my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]');
+ my @arr = $webpac->fill_in_to_arr($rec,'[v900];;[v250^a]','paste');
 
 =cut
 
-sub fill_in_to_arr {
+sub _rec_to_arr {
 	my $self = shift;
 
-	my ($rec, $format_utf8) = @_;
+	my ($rec, $format_utf8, $code) = @_;
 
 	my $log = $self->_get_logger();
 
 	$log->logconfess("need HASH as first argument!") if ($rec !~ /HASH/o);
 	return if (! $format_utf8);
 
+	$log->debug("using $code on $format_utf8");
+
 	my $i = 0;
+	my $max = 0;
 	my @arr;
+	my $rec_size = {};
 
-	my $rec_size;
-
-	while (my $v = $self->fill_in($rec,$format_utf8,$i,\$rec_size)) {
-		push @arr, $v;
-		warn "rec_size = ", Dumper($rec_size);
+	while ($i <= $max) {
+		my @v = $self->$code($rec,$format_utf8,$i++,\$rec_size);
+		if ($rec_size) {
+			foreach my $f (keys %{ $rec_size }) {
+				$max = $rec_size->{$f} if ($rec_size->{$f} > $max);
+			}
+			$log->debug("max set to $max");
+			undef $rec_size;
+		}
+		if (@v) {
+			push @arr, @v;
+		} else {
+			push @arr, '' if ($max > $i);
+		}
 	}
 
 	$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr);
@@ -649,7 +658,8 @@
 			if ($$rec->{$f}->[$i] =~ /HASH/o) {
 				my $out;
 				foreach my $k (keys %{$$rec->{$f}->[$i]}) {
-					$out .= '$' . $k .':' . $$rec->{$f}->[$i]->{$k}." ";
+					my $v = $$rec->{$f}->[$i]->{$k};
+					$out .= '$' . $k .':' . $v if ($v);
 				}
 				return $out;
 			} else {