--- trunk/lib/WebPAC/Normalize.pm	2006/01/08 20:32:06	368
+++ trunk/lib/WebPAC/Normalize.pm	2006/01/08 21:50:34	372
@@ -198,9 +198,9 @@
 
 			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");
@@ -308,7 +308,7 @@
 sub parse {
 	my $self = shift;
 
-	my ($rec, $format_utf8, $i) = @_;
+	my ($rec, $format_utf8, $i, $rec_size) = @_;
 
 	return if (! $format_utf8);
 
@@ -357,7 +357,7 @@
 		}
 
 		my $found = 0;
-		my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found);
+		my $tmp = $self->get_data(\$rec,$3,$4,$r,\$found,$rec_size);
 
 		if ($found) {
 			$found_any->{$fld_type} += $found;
@@ -410,37 +410,6 @@
 	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;
-
-	while (my $v = $self->parse($rec,$format_utf8,$i++)) {
-		push @arr, $v;
-	}
-
-	$log->debug("format '$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
@@ -462,6 +431,11 @@
 This method will automatically decode UTF-8 string to local code page
 if needed.
 
+There is optional parametar C<$record_size> which can be used to get sizes of
+all C<field^subfield> combinations in this format.
+
+ my $text = $webpac->fill_in($rec,'got: v900^a v900^x',0,\$rec_size); 
+
 =cut
 
 sub fill_in {
@@ -469,10 +443,13 @@
 
 	my $log = $self->_get_logger();
 
-	my $rec = shift || $log->logconfess("need data record");
-	my $format = shift || $log->logconfess("need format to parse");
+	my ($rec,$format,$i,$rec_size) = @_;
+
+	$log->logconfess("need data record") unless ($rec);
+	$log->logconfess("need format to parse") unless($format);
+
 	# iteration (for repeatable fields)
-	my $i = shift || 0;
+	$i ||= 0;
 
 	$log->logdie("infitite loop in format $format") if ($i > ($self->{'max_mfn'} || 9999));
 
@@ -496,12 +473,12 @@
 
 	# do actual replacement of placeholders
 	# repeatable fields
-	if ($format =~ s/v(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,$i,\$found)/ges) {
+	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,$2,0,\$found)/ges) {
+	if ($format =~ s/s(\d+)(?:\^(\w))?/$self->get_data(\$rec,$1,$2,0,\$found,$rec_size)/ges) {
 		return if ($i > 0 && $just_single);
 	}
 
@@ -535,20 +512,20 @@
 }
 
 
-=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();
 
@@ -556,10 +533,20 @@
 	return if (! $format_utf8);
 
 	my $i = 0;
+	my $max = 0;
 	my @arr;
+	my $rec_size = {};
 
-	while (my $v = $self->fill_in($rec,$format_utf8,$i++)) {
+	while ($i <= $max) {
+		my $v = $self->$code($rec,$format_utf8,$i++,\$rec_size) || next;
 		push @arr, $v;
+		if ($rec_size) {
+			foreach my $f (keys %{ $rec_size }) {
+				$max = $rec_size->{$f} if ($rec_size->{$f} > $max);
+			}
+			warn "max set to $max, rec_size = ", Dumper($rec_size);
+			undef $rec_size;
+		}
 	}
 
 	$log->debug("format '$format_utf8' returned ",--$i," elements: ", sub { join(" | ",@arr) }) if (@arr);
@@ -572,7 +559,7 @@
 
 Returns value from record.
 
- my $text = $self->get_data(\$rec,$f,$sf,$i,\$found,\$fld_occurances);
+ my $text = $self->get_data(\$rec,$f,$sf,$i,\$found,\$rec_size);
 
 Required arguments are:
 
@@ -592,20 +579,20 @@
 
 =item C<$i>
 
-index offset for repeatable values ( 0 ... $#occurances )
+index offset for repeatable values ( 0 ... $rec_size->{'400^a'} )
 
 =item C<$found>
 
 optional variable that will be incremeted if preset
 
-=item C<$fld_occurances>
+=item C<$rec_size>
 
-hash to hold maximum occurances of C<field\tsubfield> combinations
+hash to hold maximum occurances of C<field^subfield> combinations
 (which can be accessed using keys in same format)
 
 =back
 
-Returns value or empty string, updates C<$found> and C<fld_occurences>
+Returns value or empty string, updates C<$found> and C<rec_size>
 if present.
 
 =cut
@@ -618,7 +605,7 @@
 	return '' unless ($$rec->{$f} && ref($$rec->{$f}) eq 'ARRAY');
 
 	if (defined($$cache)) {
-		$$cache->{"$f\t$sf"} ||= $$#rec->{$f};
+		$$cache->{ $f . ( $sf ? '^' . $sf : '' ) } ||= scalar @{ $$rec->{$f} };
 	}
 
 	return '' unless ($$rec->{$f}->[$i]);